Documente Academic
Documente Profesional
Documente Cultură
0 EXECUTIVE SUMMARY
This report aims at analysing the impact of introducing LTE technology for PMR, PAMR, and MFCN (with
channel bandwidth of 1.4 MHz, 3 MHz and 5 MHz) and narrow band IoT (with channel bandwidth of
200 kHz) within the 410-430 MHz and 450-470 MHz sub-bands based on 3GPP Release 12. Further it aims
to analyse Broadband (BB) PPDR in the band 410-430 MHz with a view to give protection to radiolocation
and radioastronomy services. Compatibility studies between LTE based BB-PPDR systems at 410-430 MHz
and 450-470 MHz and others services were carried out in ECC Report 240[1][1].
The use of the 400 MHz band is very complex both in terms of which services already use the bands and
on the density of usage by country.
Introducing LTE into the sub-band 450-470 MHz for PPDR was considered in Report 240. This report
expands the analysis of PPDR use of LTE in 410-430 MHz and LTE introduction considering different usage
scenarios in 410-430 MHz and 450-470 MHz.
This report considers compatibility between LTE and existing usage of the bands.
Compatibility between narrow band IoT using a channel bandwidth of 200 kHz within the 410-430 MHz
and 450-470 MHz sub-bands and other services and applications.
Compatibility with PMR (TETRA, DMR, Tetrapol). This is considered in section 4 and the conclusion
is:
Compatibility with DTT above 470 MHz. This is considered in section 5 and the conclusion is: Commented [A11]: I don't recall doing any studies for
Compatibility with Military Radars in 410-430 MHz. This is considered in section 6 and the conclusion IoT devices. No parameters supplied or device
is: densities
Compatibility with Radio Astronomy at 406.1-410 MHz. This is considered in section 7 and the
conclusion is:
Compatibility with Fixed Service. This is considered in section 8 and the conclusion is:
Compatibility with Programme Making. This is considered in section 9 and the conclusion is:
Compatibility with Paging systems. This is considered in section 10 and the conclusion is:
Compatibility with SRD applications at 433 MHz. This is considered in section 11 and the conclusion
is:
Compatibility between LTE technology for PMR, PAMR, and MFCN using a channel bandwidth of 1.4
MHz within the 410-430 MHz and 450-470 MHz sub-bands and other services and applications.
Compatibility with PMR (TETRA, DMR, Tetrapol). This is considered in section 4 and the conclusion
is:
Compatibility with DTT above 470 MHz. This is considered in section 5 and the conclusion is: Commented [A12]: Check bandwidth of studies and
Compatibility with Military Radars in 410-430 MHz. This is considered in section 6 and the conclusion parameters.
is:
Compatibility with Radio Astronomy at 406.1-410 MHz. This is considered in section 7 and the
conclusion is:
Compatibility with Fixed Service. This is considered in section 8 and the conclusion is:
Compatibility with Programme Making. This is considered in section 9 and the conclusion is:
Compatibility with Paging systems. This is considered in section 10 and the conclusion is:
Compatibility with SRD applications at 433 MHz. This is considered in section 11 and the conclusion
is:
ECC REPORT <No> - Page 3
Compatibility between LTE technology for PMR, PAMR, and MFCN using a channel bandwidth of 3 MHz
within the 410-430 MHz and 450-470 MHz sub-bands and other services and applications.
Compatibility with PMR (TETRA, DMR, Tetrapol). This is considered in section 4 and the conclusion
is:
Compatibility with DTT above 470 MHz. This is considered in section 5 and the conclusion is: Commented [A13]: Check bandwidth of studies and
Compatibility with Military Radars in 410-430 MHz. This is considered in section 6 and the conclusion parameters
is:
Compatibility with Radio Astronomy at 406.1-410 MHz. This is considered in section 7 and the
conclusion is:
Compatibility with Fixed Service. This is considered in section 8 and the conclusion is:
Compatibility with Programme Making. This is considered in section 9 and the conclusion is:
Compatibility with Paging systems. This is considered in section 10 and the conclusion is:
Compatibility with SRD applications at 433 MHz. This is considered in section 11 and the conclus ion
is:
Compatibility between LTE technology for PMR, PAMR, and MFCN using a channel bandwidth of 5 MHz
within the 410-430 MHz and 450-470 MHz sub-bands and other services and applications.
Compatibility with PMR (TETRA, DMR, Tetrapol). This is considered in section 4 and the conclusion
is:
Compatibility with DTT above 470 MHz. This is considered in section 5 and the conclusion is: Commented [A14]: I think studies done for this
Compatibility with Military Radars in 410-430 MHz. This is considered in section 6 and the conclusion bandwidth?
is:
Compatibility with Radio Astronomy at 406.1-410 MHz. This is considered in section 7 and the
conclusion is:
Compatibility with Fixed Service. This is considered in section 8 and the conclusion is:
Compatibility with Programme Making. This is considered in section 9 and the conclusion is:
Compatibility with Paging systems. This is considered in section 10 and the conclusion is:
Compatibility with SRD applications at 433 MHz. This is considered in section 11 and the conclusion
is:
ECC REPORT <No> - Page 4
TABLE OF CONTENTS
1 Introduction .............................................................................................................................................11
1.1 The band 410-430 MHz .................................................................................................................. 11
1.2 The band 450-470 MHz .................................................................................................................. 11
5.3.2 UE out-of-band emissions level to protect portable DTT reception based on MCL
calculations ........................................................................................................................ 40
5.3.2.1 Assumptions (portable indoor reception)............................................................ 40
5.3.2.2 Methodology .......................................................................................................41
5.3.2.3 Out-of-band emissions calculation .....................................................................41
5.3.2.4 Results ................................................................................................................ 42
5.4 Monte Carlo (SEAMCAT) Analysis N1 on the impact of UE OOBE to fixed DTT reception .........47
5.5 Monte Carlo (SEAMCAT) Analysis N2 on the impact of Base station and User Equipment OOBE
level to fixed DTT reception ............................................................................................................47
5.5.1 Basic geometry and simulation steps ................................................................................ 47
5.5.1.1 Geometry of the systems.................................................................................... 47
5.5.2 Coexistence Scenario ........................................................................................................49
5.5.3 Simulation steps ................................................................................................................. 50
5.5.4 Impact of the LTE base station on DTT reception ............................................................. 52
5.5.5 5.5.4.2 Impact of the reduction of the DTT/LTE PMR guard band on the PI of DTT
reception at the DTT cell edge ........................................................................................... 52
5.5.6 5.5.4.3 Impact of the reduction of the LTE PMR base station power on the PI of DTT
reception at the DTT cell edge ........................................................................................... 53
5.5.7 Conclusion ......................................................................................................................... 54
5.5.8 Impact of LTE UE on DTT reception.................................................................................. 55
5.5.8.1 Principle of the analysis ...................................................................................... 55
5.6 5.5.5.2 Considerations on the time aspect in the assessment of interference ............................... 56
5.6.1 Description ......................................................................................................................... 56
5.6.2 Method of calculation with formulas ................................................................................... 56
5.6.3 5.5.5.3 Results of the simulations ...................................................................................... 57
5.7 Conclusion ......................................................................................................................................59
5.8 Overall conclusions IMPACT FROM LTE IN 400 MHZ TO DTT ABOVE 470 MHZ ....................... 60
5.8.1 Conclusion LTE 400 Base stations and DTT .....................................................................60
5.8.2 Conclusion LTE 400 UE and DTT ..................................................................................... 60
12 Conclusion.............................................................................................................................................135
12.1 LTE impact on DTT above 470 MHz ............................................................................................ 135
ANNEX 10: LTE impact on DTT above 470 MHz (EBU) .............................................................................259
ANNEX 11: LTE impact on DTT above 470 MHz (Ericsson) .....................................................................272
ANNEX 12: Guidance on means to solve interference cases between LTE and DTT ........................... 274
LIST OF ABBREVIATIONS
Abbreviation Explanation
ACCP Adjacent channel co-polarized (operational option in the fixed service)
ACIR Adjacent Channel Interference Ratio
ACLR Adjacent Channel Leakage Ratio
ACS Adjacent Channel Selectivity
BB Broadband
BS Base station
BW Band Width
CDMA Code Division Multiple Access
CEPT European Conference of Postal and Telecommunications Administrations
CNR Carrier to Noise Ratio
DGNA Dynamic Group Number Assignment
DL Downlink
DM Direct Mode
DNO Distribution Network Operator
DTT Digital Terrestrial Television
ECC Electronic Communications Committee
ECO European Communications Office
ECU Electronic Control Unit
e.i.r.p. equivalent isotropically radiated power
eMTC enhanced Machine Type Communication
ETSI European Telecommunications Standards Institute
FEC Forward Error Correction
FM Frequency Modulation
FS Fixed Service
GB Guard Band
GDO Garage Door Openers
GEO Geostationary Orbit
GPS Global Positioning System
IM Inter-Modulation
IoT Internet of Things
ITU-R International Telecommunication Union - Radiocommunication Sector
LEO Low Earth Orbit
LP Location Probability
LTE Long-Term Evolution
MCL Minimum Coupling Loss
MEADS Medium Extended Air Defence System
MEO Medium Earth Orbit
MFCN Mobile Fixed Communication Network
MS Mobile Station
MSS Mobile Satellite Service
NB Narrowband
NFD Net Filter Discrimination
NJFA NATO Joint Civil/Military Frequency Agreement
ECC REPORT <No> - Page 10
OOB Out-of-Band
OOBE Out-of-Band Emissions
OBV Outside Broadcast Van
Oth Overloading threshold
PAMR Public Access Mobile Radio
pl Probability of Interference
PMR Private Mobile Radio
PPDR Public Protection and Disaster Relief
PR Protection Ratio
QAM Quadrature Amplitude Modulation
QPSK Quarterly Phase Shift Keying
RAS Radio Astronomy Service
RB Resource Block
RES Radio Equipment and Systems
RF Radio Frequency
RKE Remote Keyless Entry
RR Radio Regulations
Rx Receiver
SAR Search And Rescue
SARP Search And Rescue Processor
SARR Search And Rescue Repeater
SEAMCAT Spectrum Engineering Advanced Monte Carlo Analysis Tool
S/N Signal to Noise
SNR signal-to-noise ratio
TETRA Terrestrial Trunked Radio
T Threshold (receiver parameter common in the fixed service, equal to sensitivity)
TPM Tyre Pressure Monitoring
Tx Transmitter
UE User equipment
UHF Ultra High Frequency
UL Uplink
WRC World Radiocommunication Conference
ECC REPORT <No> - Page 11
1 INTRODUCTION
This report aims at analysing the impact of introducing LTE technology for PMR, PAMR, and MFCN
networks in the bands 410-430 MHz and 450-470 MHz. Further, it aims at analysing impact of LTE based
BB-PPDR in the band 410-430 MHz. Compatibility studies between LTE based BB-PPDR systems at 410-
430 MHz and 450-470 MHz and others services were carried out in ECC Report 240 [1][1].
This report aims at analysing the impact of introducing LTE (with channel bandwidth of 1.4 MHz, 3 MHz and
5 MHz) within the 410-430 MHz sub-band.
The conducted studies in ECC Report 240 showed that there are severe co-frequency sharing difficulties
between LTE based BB-PPDR systems and radars leading to wide exclusions zones (more than 400 km for
airborne radars), therefore further studies were needed.
Operation in adjacent band requires exclusion zones of 2.3 km and 3.8 km for airborne and ground radars
respectively, due to saturation phenomenon which cannot be solved with filtering. Therefore co-located
operation of BB-PPDR networks and radars was not seen to be possible. Out of band emissions of LTE
systems falling into radiolocation band need to remain below -114 dBm/MHz in order to avoid
desensitisation. It shall be noted that MEADS (Medium Extended Air Defence System) will be mounted on
lorries and must therefore be considered as mobile. The handling of exclusion zones (e.g. by coordination)
will appear as difficult.
The complementary compatibility and sharing studies are aimed to determine the technical and operational
characteristics of LTE network and exclusion zone around radar and radio astronomy stations in order to
provide co-existence.
Also, generic compatibility calculations for LTE systems in the 410-430 MHz band and radio astronomy
operating in the 406.1-410 MHz band showed that physical separation is required between RAS and LTE BS
and UE to achieve compatibility. In the case of a single base station emitting at a direct line of sight of a RAS
station (i.e., the worst case scenario), a separation distance of 17 km is needed. In the aggregation study
with statistical Monte-Carlo simulation a separation distance of 1 km is obtained. For these calculations a
spurious emission limit of -96 dBm/100 kHz was adopted in the UE frequency range, meant for the protection
of the base station from its own receiver or other BS transmitters. It should be noted that it was assumed that
the spurious emission limit gained from this filter is maintained throughout the RAS frequency band 406.1-
410 MHz. Should the filter apply to the UE frequency range only, then the elevated spurious emissions of -26
dBm/MHz in the RAS band will result in increased separation distances of more than 500 km between a RAS
station and the base stations. For the outdoor user equipment the separation distances for single emitter
and aggregate cases become 78 km and 326 km, respectively. For indoor usage and additional wall
attenuation of 11 dB reduces the separation distances for single emitter and aggregate cases to 34 km and
190 km, respectively. Studies in the ECC Report 240 had shown that adding the terrain profile to the analysis
does not necessarily reduce the separation distances. The effect of applying different guard bands between
the edges of the LTE UE and RAS bands was also studied. Overall compatibility between the LTE UE and
RAS in this frequency band will be very difficult to achieve.
ECC considers the amendment of the regulatory framework for PMR/PAMR in the 400 MHz band whereby
the scope should not be limited to PMR/PAMR (may also cover other land mobile radio systems, e.g. MFCN)
and the development towards LTE technology should be taken into account. Commented [A15]: Confusing wording
Therefore compatibility and sharing studies are needed for the band 410-430 MHz and for the band 450-470
MHz on the basis of LTE technology to be used by land mobile systems in this band.
ECC REPORT <No> - Page 12
The framework of the studies used for BB-PPDR in ECC Report 240 is considered to be a useful starting
point. New studies are necessary because the impact on other services is not the same as for BB -PPDR
(ECC Report 240) due to some differences in the usage scenarios as well as different user equipment
densities and resource block usage (for reference, see table 3 in ECC Report 240).
Typical usage scenarios include the introduction of MFCN-like use in a PMR/PAMR environment, broadband
PMR/PAMR using LTE technology as well as the PMR/PAMR use by the utilities sector for M2M/IoT mission
critical wide-area networks (LTE based) such as for smart grids/smart metering applications where the
terminals are dominantly not portable but fixed installations in either local, metropolitan or regional networks
with cell ranges of up to 20-35 km.
LTE profiles of 1.4 MHz, 3 MHz, and 5 MHz bandwidths are requested to be investigated and the results Commented [A16]: IoT is 200 kHz bandwidth?
should indicate amongst other findings the filter and guard-band requirements for the protection of existing
services.
In addition, ETSI prepared TR 103 401 [2][2] for smart grid systems and other radio systems suitable for
utility operations, and their long-term spectrum requirement, also covering the 400 MHz band.
ECC REPORT <No> - Page 13
2 GENERAL CONSIDERATIONS
Some points have to be addressed in order to define the common assumptions. Commented [A17]: Needs editors note
This report is based on an anticipated roll-out for LTE systems from single site, multiple sites and large wide
area networks.
The baseline scenario under study is designed to offer coverage from fixed base stations that support the
operation of PMR/PAMR services. Both radio parameters and inter-site distance are set to meet a desired
level of service availability and protection ratio at cell edge. Commented [A18]: Point to table with this information
It is foreseen that such LTE networks will be deployed primarily for critical M2M / IoT type of services. Typical
characteristics of such networks are:
LTE Networks and devices are eMTC compatible (according to the 3GPP defined Cat-M1), which implies
that they are designed with higher fade margins and, as a consequence, larger radio cells;
Most devices are indoor and on fixed locations;
There's a large amount of devices each sending or receiving only very small amounts of data at a time; Commented [A19]: There are a large number
Devices having transmitted powers that are generally lower than those used for common MFCN terminals
or PPDR devices. Commented [A110]: Need information on power
distribution otherwise this is just speculative
In comparison with LTE networks built for PPDR services, these types of networks will have significantly less (parameters)
impact on other systems in the 400 MHz band. Hence, the conclusions from ECC Report 240 [1][1], Commented [A111]: But there are many hundreds more
considering the impact of LTE PPDR networks in the 400 MHz band, may be considered as the worst case. devices so how is this justified?
Commented [A112]: Where is the evidence for this? For
example for DTT portable IoT is much worse than
2.2 FREQUENCY ALLOCATION PPDR as PPDR is tyypically out door whilst IoT/M2M
will be mainly on door
ECC Decision (04)06 [4][4] decides that Wide Band Digital Land Mobile PMR/PAMR systems in the bands
410-430 MHz and/or 450-470 MHz shall be with 10 MHz duplex spacing between the transmit frequencies of
mobile stations (410-420 MHz and 450-460 MHz) and the transmit frequencies of base stations (420-430
MHz and 460-470 MHz).
Figure 1: Illustrative mobile service allocations in the 400 MHz range Commented [A113]: What is SI, scale of freq blocks not
consistent
The frequency band 406.1-410 MHz is allocated to fixed, mobile (except aeronautical) and radio astronomy
services on a primary basis. According to RR footnote No. 5.149 [3][3], administrations are urged to take all
practicable steps to protect the radio astronomy service from harmful interference. It should be noted that 14
CEPT countries registered RAS in Europe (see Table 27Table 28).
In the band 420-430 MHz, there is a secondary allocation to radiolocation service for 29 CEPT countries out
of 48 and the only exception is the UK where it is allocated on a primary basis based on RR footnote No.
ECC REPORT <No> - Page 14
5.269. It should be noted that only UK and France reported real radar usage in part of the 420-430 MHz
band, according to the EFIS (ECO Frequency Information System) database1.
The allocation of the band 420-430 MHz for radiolocation service in Europe according to EFIS (2016) is in
Figure 2Figure 2.
The allocation of the band 406.1-410 MHz for radio astronomy service in Europe according to EFIS (in 2016)
is in Figure 3Figure 3 2.
1 http://www.efis.dk
2
http://www.efis.dk/include2/graphTool.jsp?lowRange=406.1+MHz&highRange=410+MHz&action=search&specifyRange=1&low=40
6.1&high=410&unit=MHz&user=1&languages=English&searchOption=Allocation&orientation=horizontal
ECC REPORT <No> - Page 15
Figure 3: Radio astronomy allocations in the band 406.1-410 MHz according to EFIS
Non-exhaustive options for spectrum arrangements in the frequency band 450-470 MHz:
Note that many of these options may not be very spectrum efficient in a PMR environment with small local
PMR networks and that only a smaller number of PMR systems in a given area can be realised.
Editor's Note 1: Figure concerning the band 410-430 MHz and adjacent bands
Scenario specific parameters are applied for compatibility studies in this report in order to reflect the typical
characteristics of LTE networks mentioned in section 2.12.1 above, other than those deployed for PPDR
purposes and described in ECC Report 240 [1][1]. The tables below provide a summary of parameters for
BS and UE as applied to eMTC LTE (PMR / PAMR)/ Application. To facilitate a clear comparison, the
parameters from ECC Rep. 240 (LTE450 PPDR) are also given.
ECC REPORT <No> - Page 16
Table 1: System parameters for BS (including those for PPDR for comparison)
Spurious Emissions
Frequency Reuse
1 1
Factor
Antenna Pattern/
Number of Sectors Directional / 3 Directional / 3
(Note 2)
45 dB (see Error! Reference source 45 dB (see Error! Reference source
ACLR
not found.[1], table 6.6.2.1-1) not found.[2], table 6.6.2.1-1)
-135 dBm, see Error! Reference
source not found.[6] and Error!
Reference source not found.[5]
-107,77 dBm used in ECC Rep. 240 (3GPP LTE Reference Sensitivity
Reference Sensitivity (TS 36.104 V12.3.0 states -103 dBm refers to certain throughput; rather not
(QPSK) (3 MHz) not -107.7 dBm, see Table relevant for PMR / PAMR / eMTC.
7.2.1-1 in [1]) Instead supporting of severe propaga-
tion conditions. Therefore use of
minimum conditions to maintain
relevant PHY channel (DL, UL) under
severe propagation conditions)
Deployment Density/ 0.0091 km-2 / (7.5 km cell range) 0.0018 km-2 / (17 km cell range) for
ECC REPORT <No> - Page 17
Note 1: Considering two power amplifiers transmit power is up to up to 47 dBm. (For one amplifier, the power is 44 dBm.)
Note 2: Antenna gains between 6.5 to 13 dBi are considered in this Report. In ECC Report 240, 13 dBi was used, typical directive
antenna (15 dBi) including cable loss (2 dB). Kathrein product datasheet (742 242) used to created SEAMCAT antenna patterns.
Note 3: Report ITU-R M.2292 defines 3 dB, but in ECC Report 240 value of 2 dB was used, since according to manufacturers it is
more realistic for the 400 MHz range
Note 4: -96dBm/100kHz is the 3GPP requirement for protection of own UL band. BS emissions are expected to be equal or lower than
this level at the RAS frequencies due to rejection from the TX filter.
Table 2: System Parameters for UE (including those for PPDR for comparison)
Baseline Value (UE) from ECC PMR / PAMR Value for UE (eMTC
LTE UE Parameters
Rep. 240 for Comparison Purpose Application)
Baseline Value (UE) from ECC PMR / PAMR Value for UE (eMTC
LTE UE Parameters
Rep. 240 for Comparison Purpose Application)
In order to define the interferer density, cell range calculations at 450 MHz in an urban environment have
been performed using the radio parameters of the different system and the extended Hata propagation
model as available within SEAMCAT. For DTT considerations, the cell range has been calculated at
474 MHz (central frequency of Channel 21) using the propagation model specified in Recommendation ITU-
R P.1546 [10][10].
ECC REPORT <No> - Page 19
For PMR systems,cell ranges have been calculated in order to guarantee a 75 % confidence level at the cell
fringe. Cell range depends on applications and environments considered. An example of PMR cell range
calculation is presented below.
Table 3Table 3 shows UE densities with 5.868 km Cell Range (urban indoor).
This report covers several systems. Their characteristics and special features relevant to these studies are
discussed in this section.
3.1 PMR
PMR systems are characterised by being privately owned and operated under licensed conditions, offering
professional group-communication facilities, tailor-made UE design, with deployment of predominantly
portable devices allowing users to have full control over their activities and staying very competitive.
PMR services include Group Call Voice Services (commonly called 'all informed net' and 'talk group call'),
Pre-Emptive Priority Call (Emergency Call), Call Retention, Priority Calling, Dynamic Group Number
Assignment (DGNA), Ambience Listening, Call Authorized by Dispatcher, Area Selection, Late Entry, Direct
Mode, Short Data Service, Packet Data service and smooth migration from analogue to digital platforms.
The study in this Report takes into account that BB-PMR deployment will consist in many independent stand-
alone licensed networks that are un-synchronised..
3.2 PAMR
PAMR (Public Access Mobile Radio) networks are intended for those users who need PMR type services but
cannot afford to build their own network. Operators of PAMR networks can allocate resources to different
closed user groupsincluding security companies, transportation or construction companies etc. Many of the
current PAMR networks are based on TETRA technologies. The trunking principle allows mobile stations to
choose any available free channel rather than having to wait a certain channel to become available.
Networks can be regional or nationwide.
3.3 PPDR
PPDR networks are based on cellular type architecture augmented, where necessary, by vehicle mounted
relay stations and direct mode operation.
3.4 M2M/IOT
Many forecasts and studies show that Internet of Things (IoT) and Machine-to-Machine (M2M)
communications will play an increasingly important role in Europe in the near future and will be a main
contributor to economic development. A significant portion of IoT/M2M applications will be used for critical
applications in the private and government related sectors like utilities, city and national infrastructures and
other sectors such as production, transport/logistics as well as in healthcare. Critical IoT/M2M applications
require telecommunication networks with deep indoor coverage, robust resilience and high service
availability.
ECC REPORT <No> - Page 21
New deployments of 450 MHz networks focussing on IoT/M2M are using either CDMA450 or LTE450.
CDMA450 technology is widely used worldwide and is already customised for IoT/M2M. LTE450 is a next
generation technology that promises substantial technical benefits for IoT/M2M with regards to coverage,
performance, cost and power consumption.
3.5 RLOC
420-450 MHz is the tuning range of airborne and ground radars, as described in Recommendation ITU-R
M.1462. This frequency range is harmonized for NATO ground, air and naval military radar systems as
described in the NATO Joint Civil/Military Frequency Agreement (NJFA). Spectrum agility in this range is an
essential requirement to ensure coordination with other systems using this band and to operate in an
Electronic Warfare context. Thus, radiolocation systems cannot be limited to some specific frequencies in
particular in the 420-430 MHz frequency band, where radiolocation is a secondary service, and in the 430-
440 MHz frequency band, where radiolocation is a primary service.
Airborne radars are used on aeronautical platforms and are operating daily over wide areas that consist on
several hundreds of kilometres over land and sea.
Ground radars are transportable systems used to enhance the protection of specific areas. In practice, these
radars could be co-located with military forces that are operating communication systems derived from
current and future PPDR technology. Military forces are equipped with narrow band PPDR systems to
interoperate with other governmental actors for national missions. These systems are also used for military
operation in addition to hardened tactical communication systems.
3.6 RAS
The band 406.1-410 MHz is one of the preferred frequency bands for continuum observations for the radio
astronomy service. The service involves only passive systems, which are very sensitive. Radio astronomers
have no control over the power, the frequency or other characteristics of the emissions able to cause harmful
interferences. To meet the needs of radio astronomy, there may be need to limit the unwanted emission
levels of the radio transmitters and their use in the vicinity of the radio astronomy observatories.
3.7 DTT
DTT (Digital Terrestrial Television) is based on DVB-T or on the newer DVB-T2 technology, which are based
on OFDM principle. Different sets of technical characteristics can be chosen among which quadrature
amplitude modulation (QAM) levels, forward error correction (FEC) rates, guard intervals, number of carriers
and bandwidths. Different combinations can be designed to match requirements such as between quality
and capacity, for example. In addition to traditional TV programs, DTT networks can deliver also different
data services. Commented [A115]: review
3.8 PAGING
These applications have to provide also a reliable indoor reception in case of lacking other communication
alternatives or if the availability of service has to be increased.
ECC REPORT <No> - Page 22
3.9 SRD
This section investigates and describes the impact of the LTE-based mobile PPDR unit which operates in the
frequency range from 410 430 MHz on wireless signal transmission systems on the bases of non-specific
SRDs (Short Range Devices) in accordance with band g1 - g3 (i.e. 433.05 434.79 MHz) of Annex 1 of ERC
REC 70-03 [7][7] and ETSI EN 300 220-1 and -2 [8][8]. These frequency bands are directly adjacent to the
frequency band 410 430 MHz.
The 433 MHz frequency band is of eminent importance for all kinds of non-specific SRD applications since
this band is amongst the few bands for these kind of applications that are globally available harmonised and
available. For example, many automotive applications like Garage Door Openers (GDO), Block Heater
Remote Controls, Remote Keyless Entry (RKE) or Tyre Pressure Monitoring (TPM) Systems make use of the
band g1 (433.05 434.79 MHz) according to Annex 1 of ERC REC 70-03.
Automotive Remote Keyless Entry Systems (RKE-Systems) are combining safety and comfort-related
functions. They are designed to prevent unauthorised access to the vehicle and for theft-prevention. The
comfort-related aspect of these systems is to remotely and effortlessly lock and unlock the doors of a car.
Such systems consist of an external remote Key (Transceiver) as well as of an antenna Module and an
electronic control Unit (ECU) (Transceiver) built into the vehicle itself.
The particular RF-Parameters depend on the model and the manufacturer and therefore have a wide
variation. For this investigation a contemporary complex keyless-Go system and a legacy keyless-Entry
system have been chosen to represent this kind of application.
3.10 FS
The fixed service (FS) in the 400 MHz is dedicated to single narrowband radio links. It is not intended to
assemble network structures. In case of not having another option for communication, these links are
considered as an alternative. The applications are mainly used by e.g. utilities and public service providers
for metering and process control purposes.
Even these applications are not covered by the Harmonized Standard ETSI EN 302 217-2-2 [24][24] they
can be considered as specific in some countries
Within all areas of PMSE productions, there is a need for service links for activities such as communications,
control of staff operating cameras and machinery; these are either conventional push-to-talk systems or
constant carrier base unit. The SAP/SAB applications include both ENG/OB and SNG/OB applications and
also the communication links that may be used in the production of programmes, such as talk-back or
personal monitoring of telecommand, remote control and similar applications.
Such use can be land-, air-, or see-based, and some of them are internationally harmonised for aerial use to
ease cross-border coordination.
The demands of broadcast productions make the full-duplex operation of wireless intercom systems an
absolute necessity for stage managers, lighting and audio technicians or any professional who has to deal
with the breakneck speed and complexity of television productions. Wireless intercoms encompass all sorts
of systems from the most basic pair of walkie-talkies (PMR) to dedicated professional full duplex wireless
intercom products. Wireless intercom systems are employed where the limitations of wireless systems, which
can include fidelity, interference, lack of range, lack of security (real or perceived) and battery life limitations ,
are outweighed by the freedom of being cordless. This freedom is essential in many applications.
ECC REPORT <No> - Page 23
Wireless intercom systems can be designed, installed, configured and operated in partyline3 or matrix4
configurations, and may very likely be connected to a hard-wired matrix intercom system at some point. For
daily use in event and sport productions, OBVs (Outside Broadcast Van) needs at least 2x duplex (open
microphone production/engineers with talkback) and 2x simplex (simplex and interrupted foldback systems).
High quality wireless intercoms offer a distinct advantage over traditional, two-way radios in that they offer a
more natural full duplex operation. This enables all users on the system to speak and hear other users Commented [JC16]: Please further explain this concept
simultaneously without covering other users transmissions.
Certain types of frequency use are subject to licensing requirements during events. To prevent harmful
interference or breakdown of equipment as much as possible, it is vital that the frequencies used for wireless
communications are mutually coordinated with temporary and permanent users.
The following figure shows a typical application that is in operation in the environment of the content and
event production, where the management system will be contained in a van with a Base Station antenna on
3 Wireless intercom loop system that people can join just by plugging in their unit and select who he wants to talk to.
the roof. The management system is connected to various mobile stations that can be located in helicopters,
airplanes or on motorbikes.
PMSE links use both simplex, duplex, and semi-duplex (constant carrier base unit) in both identified bands
410-430 MHz and 450-470 MHz. Frequencies used by airplanes or helicopters are internationally
harmonised and coordinated; these include 419.8375 MHz, 419.8625 MHz, 419.9250 MHz, 419.9500 MHz,
419.9750 MHz, 429.8375 MHz, 429.8625 MHz, 429.9250 MHz, 429.9500 MHz, 429.9750 MHz.
Studies concerning impact from LTE on narrow-band PMR systems are in ANNEX 2:ANNEX 2: to ANNEX
8:ANNEX 8:
LTE transmitters may cause interference to PMR receivers in adjacent frequencies. There are three main
interference mechanisms, which are described in the following sections:
Receiver blocking;
Transmitter out-of-band emissions (OOBE); and
Receiver intermodulation.
Receiver blocking occurs when the interfering signal strength at the victim receiver is higher than the
blocking level specified for the victim receiver. In such case, the receiver is overloaded to the point that
reception is not possible as the receiver is operating beyond its linear range. Blocking performance depends
on the receiver selectivity requirements specified and usually improves at greater frequency separation from
the interfering transmitter, as the selectivity of the receiver helps to attenuate the signal received from the
interfering transmitter. PMR receiver specifications for blocking performance are for example -25 dBm at a
frequency separation greater than 500 kHz for TETRA. Simulation results based on MATLAB and SEAMCAT
presented in the Annexes 03, 04 and 05 take such specifications into account.
The effect of blocking is confined to locations close to the LTE transmitter. Generally other effects such as
transmitter OOBE and receiver intermodulation may also be present at such locations.
No transmitter is able to confine all of its transmitted energy within its defined wanted bandwidth.
Thereforeout-of-band emissions, which are a combination of spurious emissions and wideband noise, are
present in adjacent frequency bands where they can be the cause of undue interference to receivers
belonging to other systems. Due to the frequency allocations in 400 MHz described in section 2.22.2 of this
Report, narrowband receivers will be operating in adjacent frequency bands to broadband transmitters, thus
the level of spurious and noise emissions from LTE transmitters will be according to that specified in 3GPP
specifications.
In the case of undue interference, out-of-band emissions from the LTE transmitter can be mitigated by
additional filtering applied to the transmitter. This can reduce the level of emissions that interfere wit h
narrowband receivers until reaching a point where the overload effect in a receiver (blocking or
intermodulation) becomes dominant. Beyond this point, further filtering does not add any benefits. ANNEX
7:ANNEX 7:contains results of simulations for varying levels of filter attenuation.
The duplex filter needed by an LTE base station is an example of an external filter which can improve the
out-of-band emissions performance. This will provide substantial attenuation of OOBE effects at the LTE
receiver band (5 MHz from the transmitter band), and will also provide variable degrees of attenuation in an
adjacent PMR receiver band. Close to the LTE transmitter frequency, the effects will be minimal (to avoid
attenuating the wanted LTE transmitter signal); but attenuation will increase at increasing frequency from the
LTE transmitter band.
ECC REPORT <No> - Page 27
[SE7(17)044]
Intermodulation distortion in receivers is caused by non-linearity and limited selectivity, which are inherent to
any real receiver implementation. Multiple strong signals at different frequencies or a broadband signal (LTE,
CDMA, DTT etc.) appearing at a non-linear receiver input generate intermodulation products.
The 3rd order intermodulation products are of particular interest with regard to the receiver performance
testing because they are stronger than higher odd order interference products which may fall into the
receiver band when the receiver band is adjacent to the transmitter frequency. Since it is not feasible to
implement a perfectly linear receiver frontend, technical standards define minimum performance
requirements based on 3rd order intermodulation products. As a consequence of these minimum
requirements, receivers of certified (standardized) equipment may experience some intermodulation
distortion caused by operation of a radio system in neighbouring bands.
Receiver intermodulation performance degradation can be mitigated by the design of receiver selectivity that
suppresses signals from neighbouring bands and receiver gain control. Receiver design generally attempts a
compromise between high sensitivity (which can impact intermodulation performance) and adherence to
intermodulation specifications.
Receiver intermodulation is a factor causing interference from both narrowband networks and broadband
networks. In a particular situation where a sufficient amount of spectrum is assigned to one narrow band
system operator, the problem can be alleviated in narrowband networks by variations of frequency and
transmission times. However, such mitigation measures come with the cost of an inefficient use of spectrum
and are only applicable for lightly loaded networks, since some frequency combinations o r transmission
times have to be excluded. Furthermore, for typical uncoordinated small PMR networks such measures
cannot be applied, since the frequency and transmission times are not under control of a single PMR
operator.
Unwanted signals which are received by a victim receiver with sufficient signal strength may produce an
intermodulation product in the victim receiver, but it only causes interference if the frequency relationship is
such that the intermodulation product falls within the wanted receiver bandwidth. This could be the case in
uncoordinated PMR or in heavily loaded PMR networks where concerned combinations of narrow band
frequencies and transmission times cannot be excluded. A broadband transmitter generates intermodulation
products across a broad frequency range; hence it is not possible to avoid energy from the intermodulation
process falling within the wanted receiver bandwidth. However, with broadband interference only a fraction of
the interference power is captured in narrowband receivers given their selectivity..
Embarking upon modelling intermodulation distortion from wideband systems proves to be a very complex
exercise. Currently, the model provided in SEAMCAT is applicable to narrowband systems, but does not yet
account for the complexity of a wideband system. Further development of IMD modelling in SEAMCAT that
is widely accepted, is certainly needed. This is challenging as there is no standardized nonlinearity model for
receivers covering a large amplitude range, whereas the classical 3rd order intermodulation model is only
valid for weak non-linearity. Several factors are to be taken into account that are often overlooked, the most
important ones being:
Gain compression, that drastically reduces the validity range of weak non-linearity (the range in which the
relationship between the input power and the third order intermodulation product can be supposed to be 3
dB/1 dB)
The fact that the ratio between the fundamental signal and the 3rd order intermodulation product will never
be lower than 13 dB
The attenuation of the interfering signal due to receiver selectivity, which can be shown to be >15 dB for
signals with more than 500 kHz frequency offset.
ECC REPORT <No> - Page 28
These factors are described in the following sections, after which a realistic set of assumptions and
parameters for SEAMCAT simulations are given.
In a receiving system, the received wanted signal (or fundamental component) is amplified by the Receiver
Line-up to an output level that is in principle linear to the input level of the wanted signal.
Due to non-linearity of the receiver line-up characteristics, 3rd order intermodulation products may be
produced within the wanted receiver bandwidth as well, the output level of which is approximately linearly
proportional to the input level of the wanted signal but with a steeper slope, increasing with 3 dB for every dB
increase of the input level of the interfering signal. This theoretical concept suggests that there would be a
3rd order intercept point above which the 3rd order intermodulation product would be larger than the
fundamental component, as depicted in figure 1:
This however constitutes a theoretical concept and the intercept point can never be reached in reality, since
at higher input levels, the Receiver Line-up will saturate and the output levels dont increase further with
higher input levels. Moreover, its a well known fact, both from theory and from measurements, that 3rd order
intermodulation products always remain well below the signal level of the fundamental component. This is
clearly described in [1] and [2], where its also shown that the worst case ratio between the fundamental
component and the 3rd order intermodulation product cannot be lower than 13 dB (figure 2).
ECC REPORT <No> - Page 29
Figure 2: Ratio between fundamental component and 3rd order Intermodulation Product in the realistic case
of Gain Compression
If this phenomenon is ignored in simulations, the impact of the Intermodulation on system performance will
be highly overestimated.
Intermodulation products are built up from spectral components, the frequencies of which will have a certain
frequency offset from the receivers centre frequency. Due to receiver selectivity, these signal
components will be attenuated, the attenuation depending on the frequency offsets involved.
For that reason, its mandatory to take the TETRA receivers selectivity into account. The receivers
architecture and selectivity typically isnt known directly but can be derived indirectly from high amplitude
signal specifications and from standard performance measurements, for which blocking performance
tests are performed.
Such a test is performed with a wanted signal 3 dB above the reference sensitivity, in the presence of a Single
Tone Interferer at different frequency offestes from the wanted signal [3]. The system performance degradation shall not occur up to
the following levels of the interfering signal for different frequency offsets:
50 100 -40
100 200 -35
200 500 -30
500 -25
Without the selectivity, the level for blocking would be the same for all defined frequency offsets. Thus the
receiver sensitivity can be derived from these results. For signals with a frequency offset of 500 kHz, the
attenuation amounts to 15 dB (being the difference between the level of the interferer signal at >500kHz
and that at 50-100kHz frequency offset). In case of intermodulation from an LTE channel with a typical
bandwidth of 3 MHz, the offset frequencies will generally be far above this figure, hence a 15 dB
attenuation of the interfering components entering the non-linear TETRA receiver is a conservative
assumption that could be applied in intermodulation calculations.
ECC REPORT <No> - Page 30
The effects of gain compression and the minimum ratio of 13 dB between the fundamental component and
the 3rd order intermodulation products must be taken into account when running simulations in SEAMCAT to
determine the impact of intermodulation of wideband LTE into a neighbouring TETRA receiver. The upper
limit for the weak non-linearity range can be derived from extrapolation of the saturation asymptotes for the
fundamental and the 3rd order IM products to lower values, as shown in figure 3. The weak non-linearity
range, i.e. the linear range of the 3rd order intermodulation product, is always at least 13 dB below the
fundamental component.
Figure 3: Piecewise linear approximation of the output curves of both the fundamental component and 3rd
order intermodulation product, being somewhat pessimistic but pragmatic.
As can be seen from figure 3, this approach means that the validity range for weak non-linearity is extended
a little bit. Hence, this approach is somewhat pessimistic, but the impact on the outcomes of the
simulations is expected to be within acceptable limits.
In addition, the levels of the interfering signal entering the non-linear TETRA receiver are reduced by 15 dB
due to the selectivity of the TETRA receiver derived from frequency offset dependant TETRA receiver
blocking specifications.
4.4.5 REFERENCES
[1] GEC Plessey Semiconductors 1993 Publication No AN-156 Issue No 2.0, September 1993:
Intermodulation, Phase Noise and Dynamic Range (http://www.ab4oj.com/dl/plessey_an156.pdf)
[2] George S.F and Wood, 3W, Washington D.C.: U.S. Naval Research Laboratory AD266069: Ideal
Limiting Part 1, 2nd October, 1961 (https://catalog.hathitrust.org/Record/009204912)
The simulations outlined in Annexes 2 to 8 show the effect of LTE transmitters causing interference to
narrowband PMR receivers in adjacent frequency spectrum Other results based on MATLAB and SEAMCAT
are presented in the Annexes 03, 04 and 05. It can be seen that the interference probabilities for low to
medium BS and MS densities are generally within 1 % or less, albeit that a 25 dB duplex filtering at the BS
may be required to keep the interference from the LTE BS into the PMR BS at these low levels. Even lower
interference probabilities are expected if the bursty nature of the traffic is included in the models. Hence,
compatibility of LTE with other technologies in the 400 MHz is easily achieved.
The effects are greatest when the victim receiver is closest to the LTE transmitter, and when the victim
receiver frequency is closest to the edge of the LTE transmitter bandwidth.
According to Annex 2, the dominant effect is that caused by transmitter out-of-band emissions. This can be
mitigated by additional filtering on that transmitter, such as the duplex filter which is used to protect the LTE
receiver. The filtering will have greatest effects at the greatest frequency separations from the LTE
transmitter. However, according to this model, once filtering on the transmitter has attenuated OOBE by
more than approximately 20 dB, receiver intermodulation becomes the dominant effect, and this sets the limit
on the distance at which interference is manageable. The accuracy of this model for intermodulation
distortion in narrow band receiver is questionable since the model was applied for high amplitude signals.
Furthermore, narrow band receiver selectivity that attenuates signals in adjacent bands and thus significantly
contributes to the mitigation of this distortion is not taken into account.
ECC REPORT <No> - Page 32
[Ericsson:]
Three different studies were done to assess the possible impact of LTEin400 on DTT systems above 470
MHz.
Annex 11 describes a Monte Carlo study performed by Ericsson. The study notes that ECC Report 240
concluded that Monte Carlo simulations demonstrated limited interferences to DTT for high power UE (37
dBm) with improved ACLR (79 dB, i.e. OOBE of -42 dBm/8 MHz) in DTT Channel 21. The Monte Carlo study
in Annex 11 uses two scenarios. In the first scenario the DTT receiver is placed at the DTT cell edge, while in
the second scenario the DTT receiver is randomly placed within the DTT cell range. In both scenarios the
LTE UEs are placed within 50 meters of the DTT receiver. The study assumes UE unwanted emission level
above 470 MHz of -42 dBm/8MHz. The Monte Carlo simulations show limited probability of interference. Commented [JC17]: please specify the meaning of
"limited" comparing the result of the probability of
Annex 9 describes a Monte Carlo study performed by FranceR.: The simulation results show that: interference with the criterion chosen
Commented [JC18]: in the final report, there is no need
the impact of the LTE PMR400 (5 MHz) BS with an e.i.r.p of 59 dBm on the DTT reception above 470 MHz, to mention the source of the study, as it is agreed by
with a guard band of 3 MHz, is similar to that of the LTE BB PPDR400 (3 MHz) BS with an e.i.r.p of 60 dBm. the whole group
The improvement of PMR BS ACLR from 45 to 67 dB/8 MHz reduces the PI to DTTB reception on channel
21 from 44 % to 8 %. The improvement of the ACLR beyond 67 dB, corresponding to an OOBE level of -8
dBm/8 MHz, does not improve notably the PI to DTTB reception due to the limiting factor of DTT ACS, which
is 65 dB. The reduction of the DTT/LTE PMR guard band from 3 MHz to 2.5 MHz has a negligible impact on
the PI to DTT reception at the DTT cell edge. It increases the PI to DTT reception only from 8.24 % to 8.66 %
for a LTE PMR BS ACLR of 67 dB/8 MHz.
with a LTE PMR UE ACLR of 65 dB/8 MHz resulting in a UE out-of-band emissions level of -42 dBm/8
MHz, the interference from LTE UE to DTT receiver is very low, in particular when the UE Tx power range is
from -40 to 23 dBm (use of the LTE power control) For all the used UE densities, the probability of
interference (PI) and probability of disruption (Pd) to DTT reception are less than 0.01 % and 3 %
respectively.Note that the PI to DTT reception increases with the increase of the number of active UE per
cell, but this increase is mainly due to the in-band power of UE. The improvement of the OOBE of the UE
does not improve notably the PI due to the limiting factor of the UE in-band power (23 dBm).
Consequently, it is concluded that the out-of-band emissions limits defined, for the BS and UE of LTE BB
PPDR 400 MHz, in ECC Report 240, as well as in ECC Decision(16)02, also apply to the BS and UE of LTE
PMR 400 MHz. In the case of interference from LTE PMR base stations to DTT reception despite the above
measure, the interference can be combated by implementing an adequate mitigation technique on a case by
case basis at national level.
Annex 10 describes a MCL study performed by EBU. It is shown that the unwanted emissions above 470 Commented [JC19]: in the final report, there is no need
MHz need to be adequately limited in order to minimize interference. The MCL study concludes that in order to mention the source of the study, as it is agreed by
to protect DTT in channel 21 on the criteria of a receiver sensitivity degradation limited to 1 dB, the LTE the whole group
unwanted emission level should not exceed -70 dBm/8 MHz for fixed reception and -75 dBm/8 MHz for
portable reception. This is derived under the condition of DTT ACS values of 80 dB and 85 dB, respectively.
It is noted that the values in Table 144 for LTE Base Station OOBE e.i.r.p levels for protection of DTT above
470 MHz was concluded in ECC Report 240. Commented [A120]: Needs removing as individual
annexes will disappear
5.1 INTRODUCTION
The study analyses the adjacent band compatibility between LTE PMR/PAMR systems operating in the 450-
470 MHz with DTT receivers services operating at DTT channel 21 (in the broadcast band 470- 694 478
MHz) needs to be considered as. Bboth BS and UE may create interference into the DTT receiver system
and the LTE receivers may be subject to interference from DTT transmitters. , therefore tThe two cases of
ECC REPORT <No> - Page 33
interference from BS and UE into DTT and DTT to LTE receivers has been are previously evaluated and
been presented in ECC Report 240 for the scenario where the 450 - 470 band was used for PPDR. .
Although Whilst the study of the impact of DTT on LTE BS and UE has already been done and the results
presented in ECC Report 240 remain applicable, this new study considers a new set of parameters,
applicable to the new use case, PMR/PAMR, MFCN and M2M/IoT, which are different from those of LTE BB-
PPDR considered in ECC report 240. In particular, in order to be aligned with 3GPP channel 31, a guard
band of 2.5 MHz between the LTE BS channel upper limit and DTT channel 21 lower limit is investigated.
In Summary,For the purpose of this study use of the band 450 - 470 MHz is aligned with 3GPP channel 31,
that is the LTE uplink band is startsing at 452.5 MHz and the downlink band starts at 462.5 MHz as
illustrated by the followingshown in Figure 8Figure 9. The basis is LTE Band 31.
Figure 89: Illustrative frequency allocation of the 450-470 MHz Commented [A121]: Show upper freq of downlink
The tTwo main methods are used for assessing compatibility in the studies are:
The MCL (Minimum coupling Coupling Lloss (MCL) method.- this method of calculation, being based on a
static geometry and parameters between the interfering and the wanted systems and as such provides an
upper bound to the level of expected interference. Generally interference may be lower in systems where the
level of interference varies with with time, such as with LTE UE because of Transmit Power Control (TPC)
mechanism and the fact they move and transmit in an intermittent manner. In such cases a statistical
approach, a Monte Carlo simulation, may provide a better indication of likely interference.This method of
calculation provides an upper bound of the interference assessment for a given static geometry between the
interfering and the wanted systems. It is simple enough to allow understanding the effect of individual
technical parameters on the result. However, it doesn't allow taking into account the inherent statistical
nature of some parameters, like the location variation of field strength due to propagation and to possible
Transmit Power Control (TPC) mechanism and the moving and intermittent nature of moving sources or
interference (the case of User Equipment)
The MC (Monte Carlo (MC) method. - To account for the varying nature of interference Tthis method of
calculation is performed used to provide an indication of assess the probability of interference faced by the
victimthat a DTT receiver may be subject to interference. The MC calculations as perfomed are typically only
relevant for one moment in time, which whilst useful for understanding loss of data throughput in for example
a system, do not characterise the interference that a DTT viewer may experience whilst watching television
over a period of time. For this further calculations have been performed to give an indication of probability of
interference in a given time interval.
The Monte Carlo analysis for the UE case enables to compare the interference with the MCL analysis where
the UE is considered to be located at the position that maximizes the interference. Indeed, an MCL analysys
does not encompass all real life situations where devices may be located anywhere in the vicinity of the DTT
Rx. Therefore, if only the conclusions of the MCL analysis are considered, they may lead to undue
constraints for the operation of the UE. The time aspect of the interference from UE to DTT Rx is considered.
As indicated in ECC Report 240 (section 3.5.1.3), the Monte Carlo simulation method used in this study has
been usedwas originally used to determine the OOBE emission limits of LTE 800 MHz Base Stations in the
UHF broadcasting band. The method is described in detail in ANNEX 4 of ECC Report 240.
Also according to ECC Report 240, several different Monte Carlo simulation methods were have previously
been used to determine the OOBE emission limits of LTE800 and LTE700 User equipment. Studies
conducted in ITU-R/JTG 4-5-6-7 as well as in CPG/PTD have already recognized the insufficiency of the
calculation of the interference probability (IP) vis--vis interference into the broadcasting service and the
ECC REPORT <No> - Page 34
need to take the time into account when dealing with IMT UE interference. The studies in this report attempt
to address this deficiency.
There is a need to take the time into account when dealing with IMT UE interference, since [].
Studies on the protection of DTT above 470 MHz from LTE BS in the 450 470 MHz range were carried out
in ECC Report 240 section 3.5.3.1 (Minimum Coupling Loss) and 3.5.1.4 (Monte Carlo).
Especially for the Monte Carlo simulations the interference probability is likely to be different. However, using
the same method of interpreting the results, the same conclusion can be drawn.
In particular, the following text, quoted from the executive summary of ECC Report 240 related to PPDR
Base station impact on DTT above 470 MHz, applies to LTE450 base station impact to DTT above 470 MHz:
"The results of the theoretical co-existence analyses with DTT demonstrate interferences from the PPDR
LTEin400 system to DTT reception when the PPDR system is adjacent in the frequency domain to the lower
DTT Channel, i.e. Channel 21. Nevertheless, the risk of interference can be reduced by at a set of technical
measures including a guard band of up to 3 MHz between DTT and PPDR BSs and an appropriate limit of
the corresponding PPDR BS out-of-band emissions. Furthermore additional mitigation measures may be
required to solve possible residual interference from PPDR BSs on a case by case basis in a manner similar
to the situation between LTE800 and DTT".
LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz are given in Table 161 5 Field Code Changed
below.
Table 65: PPDR 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz
Maximum mean
Condition on Base OOBE e.i.r.p
Measurement
Frequency range station in-block e.i.r.p, (dBm/cell)
bandwidth
P (dBm/cell)
See ANNEX 2:ANNEX 2: for a list of possible mitigation techniques (list of mitigation measures).
Studies carried out for compatibility between LTE800 and portable DTT reception concluded that portable
DTT reception is less susceptible to interference from base stations 5. Additional studies are not required. If
fixed DTT reception is protected from base station interference, portable DTT reception is automatically
protected.
ECC Report 240 considered a LTE UE an unwanted emission level of -42dBm/8MHz, as this level was
initially agreed for allowing coexistence with DTT below 694 MHz with UE transmitting above 703 MHz.
Furthermore, ECC Report 240 concluded that Monte-Carlo simulations demonstrated limited probability of
interference into to DTT reception for high power UE (37 dBm) with improved ACLR (79 dB, i.e. OOBE of -42
dBm / 8 MHz) in Channel 21.
5.3 MCL (MINIMUM COUPLING LOSS) ANALYSIS: UE OOBE LEVEL TO PROTECT FIXED AND
PORTABLE DTT RECEPTION
5.3.1 UE out-of-band emissions level to protect fixed DTT reception based on MCL calculations
The UE out-of-band emissions level necessary to protect limit the increase in interference to 1dB to a TV
receiver using a fixed rooftop antenna from interference from a UE located outdoors is calculated in the
following chapters using a MCL analysis.
In some studies, the effect of body loss was taken into account for the LTE UE by an additional attenuation
of 4 dB (taken from Report ITU-R M.2292), in order to simulate e.g. handheld devices (mobile terminals). In
other studies, this effect was not applied in order to simulate devices not used very close to the human body,
e.g. Wifi Routers or nomadic installations.
The following assumptions have been used in the analysis of the out-of-band emissions level needed to
protect fixed DTT reception.
TV Receiver
Noise figure 6 dB
TV Receiver
Antenna height 10 m
Note that the same directional pattern is used both in azimuth and elevation, i.e., the curves represent
g,(TV)() or g,(TV)() where and are azimuth and elevation offsets from bore sight.
20
g,(TV)()
g,(TV)()
60
UE Transmitter
UE Transmitter
In some studies, the effect of body loss was taken into account for the LTE UE by an additional attenuation
of 4 dB, in order to simulate e.g. handheld devices (mobile terminals). In other studies, this effect was not
applied in order to simulate devices not used very close to the human body, e.g. broadband wireless
terminals and mobile TV receivers.
5.3.1.2 Methodology
A MCL analysis is used for evaluating the impact of adjacent-channel interference from UEs to DTT
receivers. The situation is considered where the DTT signal is received at the reference sensitivity level, the
worst case separation distance between the TV antenna and the UE is established, accounting for both the
path-loss and the elevation pattern of a typical TV antenna, and the out-of-band emissions level which would
result in a 1 dB desensitization of the TV receiver is then evaluated.
It is assumed that the TV antenna is roof mounted (at a height of 10 m) and that the UE is outdoors (at a
height of 1.5 m).
The worst-case UE to TV antenna horizontal separation distance is established by considering both the path-
loss between the UE and the TV antenna and the elevation pattern of the TV antenna.
For the path-loss the free-space model is used together with the TV antenna elevation pattern from ITU-R
BT.419-3 [18][18], see below.
The path gain between the UE and the TV receiver is calculated as follows:
where:
ECC REPORT <No> - Page 38
As can be seen, the worst-case occurs at a horizontal separation distance of 22 m, where the total coupling
gain between the UE and the TV receiver is 44.3 dB.
Having established the total path gain for the worst-case horizontal separation between the UE and TV
antenna, the out-of-band emissions needed to meet the 1 dB desensitisation criteria is calculated.
where:
k = Boltzmanns constant
T = Temperature (290 K)
B = Noise equivalent bandwidth of the TV receiver (7.6 MHz)
NF = DVB-T2 receiver noise figure (6 dB)
The interference power in the TV receiver adjacent channel is calculated from a combination of the UE in-
band power (23 dBm) and the total path gain (including 4 dB body loss at the UE) at the worst-case distance
as follows:
ECC REPORT <No> - Page 39
PAC PUE , IB GPG,(UE ,TV ) GBL 23.0 (44.26) (4.0) 25.26 dBm
From the above the adjacent-channel interference ratio (ACIR) can be established as follows:
Without body loss (e.g. for a broadband wireless internet terminal) this would be 83.74 dB.
ACIR is related to the adjacent channel selectivity (ACS) of the victim and to the adjacent-channel
interference ratio (ACLR) of the interferer via the following expression (linear units):
However, with an assumption about reasonable improvement in TV receiver ACS by means of additional
external filtering in the antenna down lead it can be concluded that an ACS figure of 80 dB or better is
achievable. Also measurement as reported in ECC Report 240 showed that 80 dB is achievable with current
receiver design.
Thus for the purposes of this calculation an ACS value of 80 dB has been used.
1 1
ACLR 10 log 10 1 79.78 / 10 80 / 10 92.84 dB
10 10
Thus for a UE transmitting at 23 dBm EIRP the out-of-band emissions will be:
This value can be rounded to -70 dBm /8 MHz. This means that an LTE user equipment BEM out-of-band
emissions limit of -70 dBm/(8 MHz) for frequencies below 790 MHz is necessary to protect fixed DTT
reception.
Rx height m 10 hRx
Path distance m 23.6 D=sqrt(dh2+(hRx-hTx)2)
Free space propagation dB 52.96 LFS
Rx antenna elevation
dB 0.45 GDir
discrimination
Rx antenna bore-sight gain dB 9.15 GRx
Body loss dB 4 LBody
Wall loss dB 0 LWall
Total coupling gain dB 48.25 Gtot = -LFS+GDir+GRx-LBody-LWall
I/N dB -5.87
Receiver desensitisation (C/N
dB 1.00 D=10log(1+10(I/N)/10)
degradation)
ACS dB 70.00
Additional filtering dB 10
Total ACS dB 80.00
ACIR dB 79.78
Interference power dBm -105.04 PI=Pn+I/N
ACLR dB 92.84
OOBE (TX) dBm/8 MHz -69.84 OOBE=PTx-ACLR
5.3.2 UE out-of-band emissions level to protect portable DTT reception based on MCL calculations
The UE out-of-band emissions level necessary to protect portable TV reception from interference from a UE
is calculated in the chapters below using MCL analysis.
TV Receiver
Noise figure 6 dB
UE Transmitter
General
5.3.2.2 Methodology
An MCL analysis is used for evaluating the impact of adjacent-channel interference from UEs to DTT
receivers. The situation is considered where the DTT signal is received at the reference sensitivity level. The
victim TV antenna and the interfering UE are assumed to be in the same building. Some of the MCL
calculations assume that they are separated by one internal wall. It can be argued that if the victim and
interferer are in the same room then the users of both devices can negotiate a local solution in case of
interference, e.g. one of them can move to increase the distance between the victim and interferer, or, if
necessary, move to another room. For various assumed values of the UE out-of-band emissions level, the
separation distance needed to meet the 1 dB desensitisation criteria is evaluated (taking account of the wall
loss). A value for the out-of-band emissions level is then chosen which balances the need to minimise the
separation distance and be achievable in a realistic terminal design.
where:
k = Boltzmanns constant
T = Temperature (290 K)
B = Noise equivalent bandwidth of the TV receiver (7.6 dB)
NF = TV receiver noise figure (6 dB)
ECC REPORT <No> - Page 42
The interference power at the source UE (P I,(UE)) is a combination of the UE in-band power (PIB,(UE) = 23
dBm) the ACS of the victim TV receiver and out-of-band emission power of the UE (P OOB,(UE)) within the
victim receivers channel as follows:
PI ,(UE) 10 log 10 10
PIB ,(UE ) ACS 10
10 OOB,(UE )
P 10
For the purposes of this calculation a minimum achievable ACS value of 85 dB has been assumed. This
takes into account that an ACS is achievable with current receiver design as shown in ECC Report 240 and
some rejection in the TV receiver antenna.
Results have also been calculated for an ACS value of 100 dB to demonstrate the impact of additional
rejection filters at the portable TV receiver.
The minimum allowed coupling gain between the interfering UE and the victim TV is therefore the difference
between the target interference power (PI) and the interference power at the source UE (P I,(UE)).
GCG PI PI ,(UE )
The total path gain between the interfering UE and the victim TV (G PG,(UE,TV)) is given by the allowed
coupling gain GCG minus the wall loss (GWL = -10.4 dB) minus the body loss at the UE (GBL = -4 dB) minus
the TV antenna gain (GA,(TV) = 2.15 dBi).
From the total path gain we can then calculate the minimum separation distance needed to meet the 1 dB
desensitisation criteria using the free-space path-loss model.
5.3.2.4 Results
As indicated above, for various assumed values of the UE out-of-band emissions level, the separation
distance needed to meet the 1 dB desensitisation criteria has been evaluated. Results have been obtained
for assumed TV ACS values of both 85 dB and 100 dB (to assess the impact of rejection filters at the
portable TV receiver).
TV ACS = 85 dB
The graph below illustrates the relationship between separation distance and out-of-band emissions. The
lower blue curve takes into account -10.4 dB wall loss whereas the upper pink curve does not.
ECC REPORT <No> - Page 43
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
As can be seen, the curves have essentially flattened out for a out-of-band emissions level of -75
dBm/(8 MHz) and below i.e. for out-of-band emissions levels lower that -75 dBm/(8 MHz) there is minimal
improvement in separation distance. From this it is concluded a UE out-of-band emission level of -75
dBm/(8 MHz) is optimal.
The following table summarises the calculation of separation distance for the situation where the assumed
TV receiver ACS is 85 dB and the out-of-band emissions is set to -75 dBm/(8 MHz) for the various
combinations of wall loss and body loss.
Target performance
Receiver NF dB 6.00 6.00 6.00 6.00 NF
Thermal Noise floor (9
dBm -99.17 -99.17 -99.17 -99.17 Pn= 10 log(kTB) + NF + 30
MHz)
INR dB -6.00 -6.00 -6.00 -6.00 INR
Target interference
dBm -105.17 -105.17 -105.17 -105.17 PItarget= Pn + INR
power
Victim's performance
Receiver selectivity
dB 85.00 85.00 85.00 85.00
(ACS)
BEM limits
In-block transmit dBm/
23.00 23.00 23.00 23.00 Pib,tr
power 10MHz
Interferer antenna gain dBi 0.00 0.00 0.00 0.00 Ga,i
ECC REPORT <No> - Page 44
dBm/
EIRP 23.00 23.00 23.00 23.00 Pib
10MHz
dBm/
Out-of-block -75.00 -75.00 -75.00 -75.00 Poob
8MHz
"Total" interference at Linear: Px = Pib/ACS + Poob,
dBm -61.79 -61.79 -61.79 -61.79
"source" where PItarget = G Px
ACIR calculation
ACLR dB 98.00 98.00 98.00 98.00 Pib - Poob
Linear = 1/((1/ACLR) +
ACIR dB 98.21 98.21 98.21 98.21
(1/ACS))
Coupling calculation
Coupling gain dB -43.48 -43.48 -43.48 -43.48 Linear: G = PItarget -Px
Link budget
Interferer body gain dB -4.00 -4.00 0.00 0.00 Gb,I
Wall gain dB -10.40 0.00 -10.40 0.00 GWl
Victim body gain dB 0.00 0.00 0.00 0.00 Gb,v
Victim ant. Elevation
dB 0.00 0.00 0.00 0.00 gb,v (assumed zero)
pattern
Victim antenna gain dB 2.15 2.15 2.15 2.15 Ga,v
GpI = G -Gb,I -GWl -gb,v -Ga,v
Path gain dB -31.13 -41.53 -35.13 -45.53
-Gb,v
Geometry
d, where GpI = 147.56 -
Protection distance m 1.89 6.26 3.00 9.92
20log10(f) - 20log10(d) dB
TV ACS = 100 dB
In order to assess the impact of a rejection filter fitted to the portable TV receiver a further set of results are
calculated but with an ACS value of 100 dB (rather than 85 dB assumed above).
The graph below provides results where the UE body loss is set to 4 dB.
ECC REPORT <No> - Page 45
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
The graph below provides results where the UE body loss is set to zero.
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
The following table summarises the calculation of separation distance for the situation where the assumed
TV receiver ACS is 100 dB and the out-of-band emissions is set to -75 dBm/(8 MHz) for the various
combinations of wall loss and body loss.
Target performance
Receiver NF dB 6.00 6.00 6.00 6.00 NF
Thermal Noise floor (9
dBm -99.17 -99.17 -99.17 -99.17 Pn= 10 log(kTB) + NF + 30
MHz)
INR dB -6.00 -6.00 -6.00 -6.00 INR
Target interference
dBm -105.17 -105.17 -105.17 -105.17 PItarget= Pn + INR
power
Victim's performance
Receiver selectivity
dB 100.00 100.00 100.00 100.00
(ACS)
BEM limits
In-block transmit dBm/
23.00 23.00 23.00 23.00 Pib,tr
power 10MHz
Interferer antenna
dBi 0.00 0.00 0.00 0.00 Ga,i
gain
dBm/
EIRP 23.00 23.00 23.00 23.00 Pib
10MHz
dBm/
Out-of-block -75.00 -75.00 -75.00 -75.00 Poob
8MHz
"Total" interference at Linear: Px = Pib/ACS +
dBm -72.88 -72.88 -72.88 -72.88
"source" Poob, where PItarget = G Px
ACIR calculation
ACLR dB 98.00 98.00 98.00 98.00 Pib - Poob
Linear = 1/((1/ACLR) +
ACIR dB 102.12 102.12 102.12 102.12
(1/ACS))
Coupling calculation
Coupling gain dB -32.29 -32.29 -32.29 -32.29 Linear: G = PItarget -Px
Link budget
Interferer body gain dB -4.00 -4.00 0.00 0.00 Gb,I
Wall gain dB -10.40 0.00 -10.40 0.00 GWl
Victim body gain dB 0.00 0.00 0.00 0.00 Gb,v
Victim ant. Elevation
dB 0.00 0.00 0.00 0.00 gb,v (assumed zero)
pattern
Victim antenna gain dB 2.15 2.15 2.15 2.15 Ga,v
GpI = G -Gb,I -GWl -gb,v -
Path gain dB -20.04 -30.44 -24.04 -34.44
Ga,v -Gb,v
Geometry
d, where GpI = 147.56 -
Protection distance m 0.53 1.75 0.84 2.77
20log10(f) - 20log10(d) dB
ECC REPORT <No> - Page 47
The MCL studies have shown that the unwanted emissions above 470 MHz need to be adequately limited in
order to minimize interference.
To protect DTT in channel 21 on the criteria of a receiver sensitivity degradation limited to 1 dB, the MCL
analysis shows that the LTE unwanted emission level should not exceed -70 dBm/8 MHz for fixed reception
and -75 dBm/8 MHz for portable reception. This is derived under the condition of DTT ACS values of 80 dB
and 85 dB, respectively.
5.4 MONTE CARLO (SEAMCAT) ANALYSIS N1 ON THE IMPACT OF UE OOBE TO FIXED DTT
RECEPTION
SEAMCAT simulations between IoT UE and DTT Channel 21 have been performed using the parameters
described in A1.1A1.1 and A1.6A1.6 and assuming a UE IoT unwanted emission level above 470 MHz
of -42 dBm/8MHz. BB PPDR devices were used in ECC Report 240 which compared to IoT devices, can use
many RBs simultaneously, have higher transmit power and are mainly outdoor.
Two scenarios have been simulated. Scenario 1 describes the worst case, on which fixed DTT receivers are
located at the DTT cell edge while in scenario 2 DTT receivers are randomly located within the DTT cell
area, representing a more realistic scenario. The IoT network is intentionally placed around the DTT receiver
to ensure proximity between the IoT UE and the DTT receiver. In both scenarios the IoT UEs are placed
within 50 meters of the DTT receivers. 3 MHz bandwidth is used for the LTE base station where 15 UEs are
actively transmitting in both of the scenarios with one RB each of 180 kHz. Transmit power for the IoT UEs
are power controlled between -40 to 23 dBm.
Pinterference
Scenario
(%)
Scenario 1 4.59 %
Scenario 2 0.00 %
5.5 MONTE CARLO (SEAMCAT) ANALYSIS N2 ON THE IMPACT OF BASE STATION AND USER
EQUIPMENT OOBE LEVEL TO FIXED DTT RECEPTION
The DTT transmitter is placed at the centre of the coverage area as depicted in Figure 139. Field Code Changed
ECC REPORT <No> - Page 48
DTT coverage radius for DTT high power transmitters (85.15 dB)
The LTE PMR base station (BS) is placed at the centre of the cell. Each LTE cell is composed of three
sectors as depicted in the following Figure.. The Cell Radius (R) corresponds to the edge of the hexagon,
following the definition of 3GPP. The Cell Range is therefore two times the value of the cell radius (2*R).
Figure 1617: Hexagonal three-sector cell layout (R= Cell radius, 2*R= cell range)
ECC REPORT <No> - Page 49
This LTE cell is repeated to build up a perfectly homogeneous single frequency LTE cluster composed of 7
cells (BS) as depicted in Figure 18. A cluster of size 7 is composed of 21 (7 x 3) hexagonal-shaped sectors.
The coexistence scenario for the analysis of potential interference from LTE BS or UE to DTT receiver uplink
is illustrated in Figure 19.
The DTT and LTE network parameters are summarised in Annex 1. The LTE network cluster is placed at a
separation distance D of the DTT receiver (D is measured between the DTT receiver and LTE cluster centre
reference cell site).
The propagation models used in the simulations for each type of link are summarised in the table below:
ECC REPORT <No> - Page 50
1 The DTT receiver is randomly positioned, following a uniform polar distribution, in the DTT cell or in a
pixel of 100 m x 100 m at the edge of the DTT cell as depicted in Figure 143. The azimuth orientation of Field Code Changed
the TV receiver antenna is directed toward the DTT transmitter in case of fixed rooftop reception.
2 Around the DTT receiver within a radius of Dmax, a LTE PMR cluster is randomly positioned following a
uniform angular distribution. The position of the cluster is defined by the position of the central cells BS
as depicted in Field Code Changed
32 Figure 144.
43 The active LTE user equipment (UE) is randomly positioned, following a uniform distribution, within each
cell of the LTE cluster.
54 The probability of interference is calculated for each simulation based on a high number of events
generated.
200000 events for assessing the impact of LTE PMR BS on DTT reception, resulting in an accuracy
which cannot be better than < 5.26E-06;
500000 events for assessing the impact of LTE PMR UE on DTT reception, resulting in an accuracy
which cannot be better than < 2.1E3-06. Commented [A122]: Tidy up
ECC REPORT <No> - Page 51
DTT Tx
Position of the LTE PMR cluster around the victim DTT receiver (a single Monte Carlo event):
TxDTT
rDTT
The considered DTT Rx protection criterion is C/(N+I) = 21 dB. For a good statistical accuracy, 200 000
events are generated for each simulation.
Assesment of compatibility between DTT and LTE PMRThe initial analysis is based on the following
assumptions:
The DTT Rx at the DTT coverage edge, uniformly located in an area of 100m x 100m (one pixel). This
corresponds to the most sensitive case for DTT reception scenario.
The position of the interfering link (IL) is set to be dynamic relative to the victim DTT Rx so that the LTE
cluster reference cell is located in the vicinity of the DTT receiver, uniformly positioned within an area
limited by a circle of radius Dmax = PMR BS cell range whose centre is the DTT Rx.
The outcome of the Monte Carlo simulation provides:
The received interference from the LTE BS unwanted emissions (iRSS_unwanted);
The received interference from the DTT receiver blocking response (iRSS_Blocking);
The probability of interference of the DTT receiver (PI), as well as the probability of DTT coverage with
interference.
The reported received useful and interfering signal levels (dRSS, iRSS Unwanted and iRSS Blocking )
correspond to the median values. They are reported in order to evaluate which contribution impacts more the
DTT Rx. Note that at the DTT cell edge the median useful signal level (dRSS) is always about -68 dBm/8
MHz..The calculated probability of interference takes into account mutually the interference from the LTE
unwanted emissions, the DTT Rx blocking and the overloading.
5.5.5 5.5.4.2 Impact of the reduction of the DTT/LTE PMR guard band on the PI of DTT reception
at the DTT cell edge Commented [A123]: numbering
A guard band of 3 MHz between DTT and BB PPDR400 has been defined in ECC Report 240 to minimise
the interference from BB PPDR400 base stations to DTT reception. While, 3GPP E-UTRA Operating Band
31 provides only a guard band of 2.5 MHz between the E-UTRA BS upper channel and DTT channel 21.
Here, the impact of the reduction of the DTT/LTE PMR guard band, from 3 MHZ to 2.5 MHz, on the PI of
DTT reception at the DTT cell edge has been assessed.
The results of the simulations are presented in Figure 22 and Table 17.
ECC REPORT <No> - Page 53
Figure 22: Impact of the DTT/ LTE PMR guard band on the DTT reception at the DTT cell edge
Table 17: Impact of the DTT/ LTE PMR guard band on the DTT reception at the DTT cell edge
Probability of interference to DTTB reception at the DTTB cell edge (dRSS=-68 dBm)
LTE PMR 5 MHz BS interfering signal (e.i.r.p = 59 dBm); PMR cell range = 1.74 km;
PMR BS PMR BS
P of non-coverage* (%)
ACLR OOBE P of non-coverage* (%)
PI (%) PI (%)
with interference
with interference
(dB/8 MHz) (dBm/8 MHz)
5.5.6 5.5.4.3 Impact of the reduction of the LTE PMR base station power on the PI of DTT
reception at the DTT cell edge
Report ITU-R M.2292 defines the maximum IMT (LTE) base station output power as 43 dBm/5 MHz, 46
dBm/10 MHz and 46 dBm/20 MHz. Actually, in some cases, for the same cell range defined by the uplink
budget, the reduction of the system bandwidth may imply the reduction of the BS power. That is the
reduction of the BS power as a function of the system BW. This power redaction can be justified from the fact
that the reduction of the system BW means the increase of the user terminal (UE) sensitivity. However, one
can also keep the BS power unchanged to benefit from an additional S/N margin on downlink.
ECC REPORT <No> - Page 54
Indeed, it is obvious that reducing the PMR BS power while keeping cell range unchanged will results in a
lower PI to DTT reception. Nevertheless, simulations have been carried out to get an overview on this point
for PMR system bandwidths (BW) of 5, 3 and 1.4 MHz, with BS e.i.r.p of 59, 57 and 53.5 dBm/BW
Table 18: impact of the LTE BS transmitted power on DTT reception at the DTT cell edge for a DTT/
LTE PMR
Probability of interference to DTTB reception at the DTTB cell edge (dRSS=-68 dBm)
LTE PMR BS interfering signal (e.i.r.p = 59 dBm); PMR ACLR = 67 dB; PMR cell range = 1.74
km;
5.5.7 Conclusion
The simulation results show that the impact of the LTE PMR400 (5 MHz) BS with an e.i.r.p of 59 dBm on the
DTT reception above 470 MHz, with a guard band of 3 MHz, is similar to that of the LTE BB PPDR400 (3
MHz) BS with an e.i.r.p of 60 dBm. The improvement of PMR BS ACLR from 45 to 67 dB/8 MHz reduces the
PI to DTTB reception on channel 21 from 44 % to 8 %. The improvement of the ACLR beyond 67 dB,
corresponding to an OOBE level of -8 dBm/8 MHz, does not improve notably the PI to DTTB reception due to
the limiting factor of DTT ACS, which is 65 dB.
The reduction of the DTT/LTE PMR guard band from 3 MHz to 2.5 MHz has a negligible impact on the PI to
DTT reception at the DTT cell edge. It increases the PI to DTT reception only from 8.24 % to 8.66 % for a
LTE PMR BS ACLR of 67 dB/8 MHz.
Consequently, it is concluded that the maximum out-of-band emissions limit of -7 dBm/8 MHz, with an ACLR
of 67 dB, defined for LTE BB PPDR 400 MHz base stations in ECC Report 240, as well as in ECC
Decision(16)02, also applies to LTE PMR 400 MHz base stations.
ECC REPORT <No> - Page 55
In the case of interference from LTE PMR base stations to DTT reception despite the above measure, the
interference can be combated by implementing an adequate mitigation technique on a case by case basis at
national level.
For this scenario, the interfering UE is randomly positioned, following a uniform distribution, within each cell
of the LTE PMR cluster around the DTT Rx and within an area that is determined by the LTE cell range (see
Figure 23). The Monte Carlo analysis is performed to assess how much the DTT receiver is statistically
impacted by the variations of the operation of the UE, which implies mutually:
The UE position around the DTT Rx
The UE effective power through the power control mechanism of the system
The number of simultaneously transmitting UE
The impact of the propagation channel variations
Figure 23: Position of interfering UEs around the DTT receiver inside the cluster
The simulations have been carried out with a guard band of 3 MHz for three different UE densities as
shoving in Table 20.
Environment Urban
2 12 1,017043677
4 6 2,034087353
UE OOBE limits are defined for full channel bandwidth occupation in 3GPP specification TS36.101. Both
simulations and laboratory measurements have shown that when UE is transmitting in partial band
depending on the number of resource blocks used, UE OOBE level is reduced, thus a correction factor
should be applied. The correction factor of UE OOBE form 25 RB (1 UE) to 12 RBs or more (2 UE or more)
is 7 dB.
Most of the simulations are carried out for 5 MHz (25 RB) LTE PMR system. The correction factors as a
function of the number of used RB used are presented in Annexe 1.
5.6.1 Description
The objective of this section is to reconcile the use of Monte Carlo approach with the need to take into
account time element by converting the Probability of Interference (PI) into a probability which would better
reflect the impact of interference on the TV viewer.
If IP is the interference probability derived from the Monte Carlo simulations and C is the number of network
state changes during a certain time window (TW), assuming that two consecutive network states are
independent (not correlated), then the probability of TV viewer observing LTE UE causing at least one
harmful interference to DTTB reception is given by:
= 1 (1 ) (1)
C is calculated as follows
C = TW/DT
DT: average decorrelation time between two consecutive network states for one active uplink data user.
The average decorrelation time reflects the fact that when a terminal is interfering with the broadcasting
receiver, it will keep the resource of the network for a certain time before this resource is allocated to another
terminal which may, or may not, cause interference to the broadcasting receiver.
Some contributions to the third mobile-DTT correspondence group meeting in October 2013 indicate a TW
equal to one hour. The basis for this value could be an average viewing time for a given TV program.
from 1 ms which is the subframe time: it is not realistic to assume that each terminal will transmit;
to the full time window. If this time window is as large as one hour, this is neither realistic since it would
assume that each terminal is permanently transmitting traffic data (other than signalling). In addition, for
such large time, the movement of the terminal would also create another dimension of decorrelation,
since the interference potential could significantly vary between the positions of the terminal during one
hour. Commented [A125]: The concept of decorrelation time
doesn't seem unreasonable in a model dependent on a
Moreover, DT depends on the UE density and the services used by the latter as well as their mobility. For a number of simplifying assumptions
given service and UE density: The problem is what should the decorrelation time be?
I suspect that it will be a function of the network type,
Time interval has to be short enough so that the network state does not change during the time interval; network loading and mobility of devices. So an IoT
Time interval has to be large enough so that the consecutive network states are de-correlated. network will be different to an MFCN network.
But it is felt possible to derive an average decorrelation time considering the various IMT services.Time
window method has been used to evaluate the impact of LTE 700 MHz and BB PPDR 700 MHz on DTT
reception:
In the first case, based on the information provided by the mobile operators, DT of 1 s, 10 s and 100 s
were used to calculate the probability of disruption (Pd) to DTT reception. Commented [A126]: No idea what it should be so we
In the second case, the operation of the BB PPDR UE is only considered effective during short-term chose a value that justifies the -42 dBm/8 MHz
interventions, the birth and death process of the UEs represented by Markov chain. Consequently, DT
used were very long: 1.85 to 20 min (see ECC Report 239)! Commented [A127]: I never really understood this in
239 - the DT in this case refers to a specific PPDR
example in Paris and is not relevant for the scenarios
under consideration
In this study we consider that the operation of the LTE PMR UE will not be limited only to short-term
interventions as BB PPDR. On the other hand, their density and the services used by the latter as well as
their mobility will be more limited than LTE 700/800 MHz. Consequently, DT of 10 s, 30 s and 60 s have
been used to calculate the LTE PMR Pd to DTT reception. DT of 1 s is considered to be too short in the case
of LTE PMR networks.
Probability of interference to DTTB reception at the DTTB cell edge (dRSS=-68 dBm)
LTE PMR 5 MHz UE interfering signal (e.i.r.p = 20 dBm); DTT/ PMR GB= 2.5 MHz; PMR cell range 1.74
km
Probability of disruption of DTTB reception during a time window at the DTTB cell edge (dRSS=-68 dBm)
PMR 5 MHz UE interfering signal (e.i.r.p = 20 dBm); DTT/ PMR GB= 2.5 MHz; PMR cell range 1.74 km
1 25 3,50E-05 0,00350
2 12 5,80E-05 0,00580
4 6 8,20E-05 0,00820
Decorrelation C (N of network
10 360
time (s) state changes)
1 25 0,01252 1,25
2 12 0,02066 2,07
4 6 0,02909 2,91
Decorrelation C (N of network
30 120
time (s) state changes)
1 25 0,00419 0,42
2 12 0,00694 0,69
ECC REPORT <No> - Page 59
1 25 0,00210 0,21
2 12 0,00491 0,49
4 6 0,01548 1,55
The results of the simulations clearly show that the PI to DTT reception increases with the increase of the
number of active UE per cell. It is important to know if this increase is mainly due to the in-band emission or
OOBE of the UE. This point has been verified by simulation. The obtained results are presented in Table 23
Impact of the LTE PMR UE OOBE on DTT reception (dRSS=-68 dBm) in the presence of
more than 1 UE per BS cell (e.i.r.p = 20 dBm)
5.7 CONCLUSION
In the case of a moving source of interference (PPDR user equipment), the implementation of interference
mitigation techniques is much more difficult than in the case of a fixed source of interference (LTE PMR base
station) as the interference is transient and therefore difficult to identify. Also, because of the unpredictable
nature of the interference coming from a moving source, the PI to DTT reception needs to be low.
ECC REPORT <No> - Page 60
The results of the study reveal that, with a LTE PMR UE ACLR of 65 dB/8 MHz resulting in a UE out-of-band
emissions level of -42 dBm/8 MHz, the interference from LTE UE to DTT receiver is very low, in particular
when the UE Tx power range is from -40 to 23 dBm (use of the LTE power control) For all the used UE
densities, the probability of interference (PI) and probability of disruption (Pd) to DTT reception are less than
0.01 % and 3 % respectively.
Note that the PI to DTT reception increases with the increase of the number of active UE per cell, but this
increase is mainly due to the in-band power of UE. The improvement of the OOBE of the UE does not
improve notably the PI due to the limiting factor of the UE in-band power (23 dBm).
Consequently, it is concluded that the OOBE limit of -42 dBm/8 MHz, defined for the UE of LTE BB PPDR
400 MHz in ECC Report 240 as well as in ECC Decision(16)02, also applies to the UE of LTE PMR 400
MHz.
Given the high number of events considered, it is reasonable to assume that this analysis and the resulting
conclusions above encompass the MCL worst case study.
5.8 OVERALL CONCLUSIONS IMPACT FROM LTE IN 400 MHZ TO DTT ABOVE 470 MHZ
The studies carried out for PPDR base stations in 400 MHz in ECC Report 240 concluded on a set of out-of-
band emissions from base stations. The scenario for this report is similar and therefore the same limits
should apply. However, additional mitigation measures may be required to solve possible residual
interference from LTE to DTT receivers on a case by case basis in a manner similar to the situation between
LTE800 and DTT (see also ANNEX 2:ANNEX 2: (list of mitigation measures)).
LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz are given in Table below.
Table 1920: LTE Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz (ECC Report
240)
Maximum mean
Condition on Base OOBE e.i.r.p
Measurement
Frequency range station in-block e.i.r.p, (dBm/cell)
bandwidth
P (dBm/cell)
The MCL studies have shown that the unwanted emissions above 470 MHz need to be adequately limited in
order to minimize interference.
To protect DTT in channel 21 on the criteria of a receiver sensitivity degradation limited to 1 dB, the MCL
analysis shows that the LTE unwanted emission level should not exceed -70 dBm/8 MHz for fixed reception
and -75 dBm/8 MHz for portable reception. This is derived under the condition of DTT ACS values of 80 dB
and 85 dB, respectively.
ECC REPORT <No> - Page 61
It should be noted that LTE includes power control for the UEs and that the UEs are moving devices. Monte
Carlo simulations are therefore useful to assess the impact of LTE UE into DTT, provided however that time
can be taken into consideration (i.e to assess the probability of disruption to DTT reception in 1 hour).
From the Monte Carlo simulations that consider time, it is concluded that the OOBE limit of -42 dBm/8 MHz,
defined for the UE of LTE BB PPDR 400 MHz in ECC Report 240 as well as in ECC Decision(16)02, also
applies to the UE of LTE PMR 400 MHz.
SEAMCAT simulations, which do not take time into account, show limited probability of interference
assuming a DTT protection level of -42 dBm/8MHz. For all the used UE densities, the probability of
interference is less than 0,01 %. This is in line with the conclusion of BB PPDR UE emissions towards DTT
in ECC Report 240.
ECC REPORT <No> - Page 62
6.1 INTRODUCTION
This report analyses the conditions required for the deployment of LTE based BB-PPDR systems in the 410-
430 MHz band without causing harmful interference to radiolocation systems operating in the overlapped
420-430 MHz frequency band or in the adjacent frequency band (430-440 MHz). The analysis focuses
mainly on the impact of LTE BS on radar receiver. The effect of the LTE PPDR UL on the radiolocation
system was already addressed in the ECC Report 240 with 37 dBm e.i.r.p. of UE. Additionally, in ECC
Report 240 it is demonstrated that compared to saturation, desensitization of the radar receiver is the most
constraining impact from LTE systems.
It is recalled that in Article 5 of the Radio Regulations (RR), the allocation in Region 1 to the radiolocation
service is secondary in the band 420-430 MHz and primary in the band 430-440 MHz. Therefore, the base
station transmitting in the band 420-430 MHz operates under a co-channel basis with radars within this band
whereas it operates under adjacent band basis when the radar operates above 430 MHz. Given the
allocation in the RR, the protection of radiolocation systems is mandatory only for the band 430-440 MHz,
although minimization of interference within the band 420-430 MHz is also desirable.
The previous studies contained in ECC Report 240 showed that LTE based BB-PPDR systems operating in
the 420-430 MHz band would cause severe desensitization of radars. Indeed, on co-channel case, to avoid
the radar receiver desensitization, calculations lead to an exclusion zone larger than 400 km for airborne
radars considering free space propagation loss and the range of 230 km for ground radars considering a
statistic propagation model (EPM73) (see section 3.7 of ECC Report 240).
Additional studies showed that for LTE based BB-PPDR base stations operating in the 420-430 MHz band
with a deterministic propagation model (taking into account terrain and clutter data) and at a determined
location, calculations lead to a needed exclusion zone of about 120 km on co-channel situation and 40 km
on adjacent channel situation around ground radar. However, these analysis performed rely on some
deterministic propagation models using parameters which depend on the location considered, therefore the
resulting conclusions may not be valid for all places and environments and therefore the results should not
be generalized so as to derive conclusions which apply everywhere and in all circumstances. In any case,
these results confirm those already recorded in ECC Report 240 if no mitigation technique is applied on LTE
systems, i.e. the separation distance needed between LTE systems and radars would not be negligible.
The adjacent channel configuration between LTE system (BB-PPDR) and radars (ground and airborne)
showed that, if the 410-430 MHz band is used for BB-PPDR as an alternative band to 450-470 MHz, the
OOBE falling into the adjacent band (430-440 MHz) interfere with applications of the radiolocation service.
To avoid causing harmful interference on stations of the primary service in the adjacent band, LTE based
BB-PPDR systems should implement the following options:
1) Maintaining a frequency guard band below 430 MHz (9.95MHz was assessed as minimizing interference
from OOBE of the BS but reveals impractible for operation of LTE), and/or
2) Reducing the OOBE of the BS as currently specified in the 3GPP standard (Release 12 considered).
The combination of the two options is an optimal solution, and in order to provide sufficient bandwidth for the
operation of LTE systems in the band 410-430 MHz, it is proposed to consider a reduced guard band
between the LTE channel upper limit and the radar lower limit. However, it should be noted that a sufficient
frequency separation is needed in order to facilitate the filter roll-off to the required value of attenuation. A
configuration that would permit to facilitate the adjacent channel coexistence between radar and LTE while
permitting to design LTE filters of acceptable complexity and cost is as follows:
To limit of the upper limit of the LTE BS channel to 427.5 MHz so that a frequency guard band of 2.5
MHz is observed from 430 MHz, and
ECC REPORT <No> - Page 63
To add an additional filter to the BS TX that provides further 40 dB of attenuation to the levels
of OOBE specified by the 3GPP standard
Radiolocation systems in the 420-430 MHz band extend over a very long range, and uses include object
identification, tracking, and cataloguing. Radar characteristics are defined in Recommendation ITU-R M.1462
[15][15]. This band is also listed in the NATO Joint Civil/Military Frequency Agreement and NATO
harmonised for these radiolocation systems.
The 420-450 MHz band is the tuning range of airborne, shipborne and ground based radars, as described in
ITU-R Recommendation M.1462. There is no information about transportable radars in Recommendation
ITU-R M.1462, but these radars are likely to have lower sensitivity (lower antenna gain). This report focuses
on fixed ground and airborne radars, because these are the most sensitive to the interference.
Radiolocation applications operating in 420-450 MHz are not limited to French and UK territories as as it
could be concluded based on EFIS database. Indeed, usage of systems includes at least the areas over
international waters (i.e. generally 12 NM -nautical miles- away from the coasts considering low water marks)
in particular for aeronautical radars. With this regard, radiolocation applications operating from around 22 km
away from the coasts of any country need to be protected from any possible harmful OOBE falling in 430-
440 MHz as radiolocation is a primary service in this band. For the analysis, only radar desensitization is
evaluated as this is the condition that determines the most the cohabitation with LTE systems (e.g.
separation distances) compared to saturation or the radar receiver. To avoid radar desensitization, a power
level of -114 dBm/MHz or below is permitted for the interferer, in accordance with ECC Report 240.
In order to evaluate co-existence, the following assumptions were used in the calculations for ground and
airborne radars:
ECC REPORT <No> - Page 64
several LTE carrier bandwidths (1.4 MHz, 3 MHz and 5 MHz) were applied;
conditions for the co-existence of LTE based BB-PPDR and radars were based on the impact of LTE on
radars and not focused on the impact of radars on LTE networks.
the calculations focused mainly on the LTE downlink frequency band (420-430 MHz), due to the fact that
radar operating band is overlapping within this sub-band;
the mobile terminals in the LTE uplink band (410-420 MHz) do not impact on desensitization of the radar
receiver, because the low power (23 dBm) UEs and UL band does not overlap with radars.
LTE parameters were derived from existing 3GPP LTE specifications (Release 12 [6]).
In order to evaluate co-existence, the following assumptions were used in the calculations for ground radars
in adjacent scenario:
real environment (terrain and topographical conditions) was used to investigate the interference
situations (e.g. hilly and flat rural areas);
more sophisticated and calibrated radio wave propagation models such as General Okumura-Hata
(COST-231) and MYRIAD were used in the selected areas;
virtual LTE network was planned with real antenna parameters (Kathrein for LTE BSs and omni-
directional for UEs) and real LTE parameters (3GPP Release 12).
The planning consideration for the traffic was to build a network with a 50 % static load inside of each cell
and a 50% load at the cell border. This means it is not created a location based traffic map and not
calculated with dynamic Monte-Carlo simulation in the territory, which is good assumption to assess the
interfering signal to the ground radar receiver. Taking into consideration our method to assess the
ground radar interfering signal to the downlink of the LTE based BB-PPDR network (420-430 MHz), it is
perfectly appropriate to use static steady load, set to 50 %, as a general average load (this is a normal
assumption of radio planners).
The following parameters are considered for LTE and radar systems. The LTE parameters are similar to
those provided in ECC Report 240, but take into account the amendments considered to enhance the results
of the previous study and facilitate the compatibility. With respect to radar parameters, they are similar to
those of ECC Report 240 and compliant to Recommendation ITU-R M.1462.
(Bth+I/N)
e.i.r.p.LTE [dBm] 56 dBm / 5MHz
54 dBm / 3MHz GRADAR [dBi] 22 38.5
50.5 dBm / 1.4MHz
Emission mask Annex 13
Saturation level -15 -10
[dBm]
hLTE [m] 30 hRADAR [m] > 9000 8
ECC REPORT <No> - Page 65
B
L prop e.i.r. p.PPDR.LTE Pps GRADAR 10 log RADAR DEC pol DECant
BLTE
The worst-case scenario assumes coupled main lobes, DECpol = 1.5 dB, DECant = 0 dB.
The necessary propagation loss is around 160 dB for the three cases.
Assuming non-coupled main lobes, the side lobe of the radar antenna can be considered. The side lobe is
attenuated by a minimum of 30 dB, and the pathloss will be modified with DEC ant = 30 dB within this range.
For ground radar, the Free-space model, which requires clear Fresnel zone, would not be the most realistic
propagation model to consider. For instance on the Earth, with distances of over 400 km (referring to ECC
Report 240) it demands more than 14 000 m BS antenna height, if the radar antenna height is 8 m, which is
clearly unrealistic. That is the consequence of the curvature of the Earth.
B
L prop e.i.r. p.PPDR.LTE Pps GRADAR 10 log RADAR DEC pol DECant
LTE
B
Free space model is generally considered for MCL analysis and reveals adequate for the study implying the
airborne radar. However, for ground radars it represents the worst case for compatibility as it does not take
into account other phenomena except the frequency and distance losses. To evaluate necessary separation
distances between LTE and ground radar systems, different propagation models can be considered.
Interference from LTE systems could be reduced in areas where the propagation environment is subject to
obstacles between the LTE transmitting system and the radar receiver, given that those obstacles may
provide additional power reduction to the interfering signal. In such case, adequate channel propagation
model with corresponding terrain model are required. As an illustration, the figure 1 below illustrates the
distance separation corresponding to a given attenuation in free space. It shows how much distance
separation can be saved depending on a given power loss reduction. For initial high propagation loss (zone
1), a 10 dB propagation loss reduction from 150 dB down to 140 dB (e.g. from obstacle) leads to a reduction
of 1200 km of the separation distance, where an additional 10 dB of propagation loss (zone 2) gives a
supplementary reduction of 400 km for the separation distance.
For a generic study, propagation model must permit to evaluate separation distances for most of cases, so
the chosen model(s) must be adapted to the configuration of the study (frequency, nature of link), and
must be well known and recognized. To illustrate some cases of the study, examples with models using
digital terrain and clutter data can be used but given the potential impact of the terrain model on the
required separation distance, the applicability of the results and their conclusions would need to be
considered with caution. Furthermore, they must be well identified as particular cases that are valid only in
the location where calculation has been done. In particular, deterministic propagation models are more
appropriate when coordination needs to be done or when a national context needs to be clarified (Annex 7).
Most propagation models need to be tuned (calibrated) by being compared to measured propagation data;
otherwise accurate pathloss predictions will be impossible to obtain. Tuned propagation models can be
calibrated to the GIS (Geographic Information System) data that takes into account the clutter data and
coverage.
The accuracy of the models that predict pathloss depends on the input parameters and the utilised
propagation mechanism. For these reasons tuned propagation models provide more realistic and accurate
result because these models use digital maps tuned by measurements. These models can be well applicable
for ground radars in adjacent channel scenario. However, it should be noted that the results obtained are
ECC REPORT <No> - Page 68
valid essentially for places where calculations have been performed. For this reason, results obtained with
deterministic models cannot be used to build a conclusion for a general case. Statistic models give a right
idea of result independently of the area (A13.8)
EPM73 and ITU-R P.526-13 models can be used to calculate propagation path loss, for both co-channel and
adjacent scenarios. These models are not valid for an air-to-ground link (Hmax: 3000 m), so free space
propagation model is used in this case.
Considering the difference of polarisation between radars and BB-PPDR base stations, a mitigation factor of
1.5 dB is taken into account.
In the case of cohabitation between radars and BB-PPDR base stations, the antenna tilt of the base station
(3) leads to the need of using a decoupling antenna factor (DECant), in the link budget:
In a first step, a value of 3 dB can be used in the budget link, as it has been done in ECC Report 240.
In a second step, a value of 10 dB is used, which may reflect much more decoupling effect due to site
configuration or environmental context.
For airborne radars the decoupling antenna factor depends on the radar positions. The decoupling
antenna factor also depends on the LTE BS antenna mask, and the airborne radar pattern. Assuming the
LTE BS antenna is not tilted, and the airborne radar antenna main lobe can scan 60 elevation and
360azimuth. The attenuation of LTE BS antenna (Kathrein, K742 242) has been calculated for several
airborne positions and angles which could be considered as DEC ant factor. These values can be found in
Annex 13.
These values were considered for both the co-channel and the adjacent channel scenarios.
Using proper mitigation techniques on LTE network, the separation distance (exclusion zone) between LTE
BSs and radars can be reduced in both co- and adjacent channel scenarios for ground and airborne radars.
Reducing the output power of LTE BS decreases the separation distance. The effect of power reduction is
significant in co-channel scenario. In adjacent channel scenario the 3GPP mask defines the absolute level of
unwanted emission, independently by the output power. For example, if the LTE bandwidth is 5 MHz in co-
channel scenario the next table represents the decrease of exclusion zone depending on reduction of power:
d (ITU-R
P.526-13) (km)
PPDR (dBm) PPDR (W) EIRP (dBm) Lprop (dB)
h1=40 m,
h2=8 m, f=420
MHz
46 40 60 202.9 120
43 20 57 199.9 115
ECC REPORT <No> - Page 69
d (ITU-R
P.526-13) (km)
PPDR (dBm) PPDR (W) EIRP (dBm) Lprop (dB)
h1=40 m,
h2=8 m, f=420
MHz
40 10 54 196.9 110
37 5 51 193.9 105
Reducing the antenna height of LTE BS decreases the separation distance. The effect of the height
reduction in co-channel scenario is slight.
In adjacent channel scenario the antenna tilting and rotation is more significant. That is why, these
modifications are preferred.
The LTE BS antenna tilt decreases the separation distance in both co- and adjacent channel scenarios. It
can be calculated with the following formulas:
In co-channel scenario:
B
L prop EIRPPPDR.LTE Pps GRADAR 10 log RADAR DEC pol DECant LMASK (, )
LTE
B
L (, )
Where MASK is the antenna mask loss value for elevation and azimuth angles. The elevation angle
represents antenna tilt, the azimuth angle represents antenna direction.
Example:
B
L prop e.i.r. p.PPDR.LTE Pps GRADAR 10 log RADAR DEC pol DECant LMASK (358,45)
LTE
B
1
L prop 56 (115.9) 38.5 10 log 1.5 3 6.6 192.3 dB
5 (non-coupled main lobes)
L prop e.i.r. p.unwantedemPPDR Pps GRADAR DEC pol DEC ant LMASK (358,45)
6.3.4 Filtering
Using additional LTE BS filtering (10-60 dB) will reduce the OOBE and therefore the separation distances
can be reduced significantly between the LTE BS and radars in adjacent channel scenario. In the case of
interference occurring due to intermodulation products within radar receiver, additional filtering for better
radar selectivity could be deployed.
ECC REPORT <No> - Page 71
According to the simulations, the required separation distance is around 120 km in the co-channel scenario
(between LTE BSs and ground radars), depending on the bandwidth of the LTE.
For any adjacent channel, the required separation distance (for ground radars) is less than 40 km over
smooth Earth (EPM73 and ITU-R P.526-13).
In this case, it can be assumed that radars will always be at a distance of 9000 m or above from the base
stations, which corresponds to its minimum flight altitude.
9000 m corresponds to a free space loss of 104 dB, at 420 MHz. This means that about 12 dB are missing in
the budget link for spurious emissions and about 35 dB for OOBE (see table A6: adjacent channel necessary
propagation losses of 140.9 dB for OOBE and 115.9 dB for spurious).
The 430-440 MHz frequency band is allocated to radiolocation on a primary basis. Therefore, to keep the
airborne radar free of interference in this frequency band, three possible solutions could be considered:
With a lower OOBE level, a frequency guard band of 100 kHz could be sufficient and the minimum frequency
separation ( F) of the LTE center frequency to the 430 MHz boundary are:
PPDR 1.4 MHz: F = 0.8 MHz
PPDR 3 MHz: F = 1.6 MHz
PPDR 5 MHz: F = 2.6 MHz
2) Use of a frequency guard band between 430 MHz and LTE center frequency (Annex 8):
This solution insures that OOBE remain essentially in the 420-430 MHz frequency band and only
spurious emissions fall above 430 MHz.
To complete this solution, spurious levels needs also be reduced by at least 11 dB.
Figure 2425: LTE emission mask with legacy OOBE and increased frequency separation to 430MHz
Without changing OOBE level, frequency guard band should be as follows ( F to the 430MHz boundary):
PPDR 1.4 MHz: F = 9.95 MHz
PPDR 3 MHz: F = 9.95 MHz
PPDR 5 MHz: F = 9.95 MHz
3) A combination of the two previous approaches to insure that OOBE remains in the 420-430 MHz
frequency band.
a frequency margin between 430 MHz and LTE center frequency, and
a reduction of the OOB level which could also be obtained with a reduction of the LTE transmitted power
(e.i.r.p), assuming that OOBE are reduced by the same value.
For airborne radars, the separation distance remains more than 400 km required if no particular mitigation
technique is applied. It should be noted that the radio horinzon basically limits the maximum value of
ECC REPORT <No> - Page 73
separation distance between LTE BS and radars considering free space propagation and line of sight
regarding flight level.
6.5 CONCLUSIONS
The analysis focused on the impact of LTE based BB-PPDR into radar systems and investigated several
propagation models and scenarios for co-channel (420-430 MHz) and adjacent channel (430-440MHz)
operation of the two systems.
According to the simulations results, it can be concluded that LTE BSs should be excluded within a radius of
about 120 km around ground radars in the co-channel scenario, the radius depending on the bandwidth of
the LTE system.
For airborne radars, the separation distances are greater than 120 km.
Coordination between radars and LTE based BB-PPDR applications can be done only in the case of fixed
ground radars;
The use of 420-430 MHz by transportable ground radars or airborne radars will most likely become difficult in
areas where LTE networks will be deployed due to large separation distances required to avoid radar
desensitization;
Although airborne radars could continue to operate (with degraded performances) in remaining radiolocation
bands (430-450 MHz), the saturation of transportable ground radar could lead to make them not usable in
areas where LTE networks are deployed (depending on the BS density, e.i.r.p, etc).
For any adjacent channel, the required separation distance to protect radars from LTE OOBE is less than
40 km or 50 km depending on propagation model used for ground radar.
ECC REPORT <No> - Page 74
Applying digital terrain based propagation models (General 450 and MYRIAD), the minimum required
separation distance (for ground radars) could be varied from 1.5 km to 28 km (depending on the channel
bandwidth and the distance of the carrier frequencies). It should be noted that this is a much more realistic
situation because the ground radar is directed towards the air and the LTE BS antenna is directed to its UEs
(downtilted) so the main lobes cannot be coupled. However, these results (obtained with General 450 and
MYRIAD) are valid only for the location where calculation has been performed.
For airborne radars, the separation distance remains more than 400 km required if no particular mitigation
technique is applied.
Using additional filtering the co-existence can be granted between LTE BS and airborne radars. In case of 5
MHz bandwidth the minimum required separation distance could be varied from 2.2 to 56.8 km with 30 dB
filtering, from 0.7 to 18 km with 40 dB filtering, and from 0.07 to 1.8 km with 60 dB depending on fc.
Based on this result, it could be concluded that without appropriate technical measures applied on LTE base
stations, ground transportable and airborne radars will suffer from an unacceptable degradation of their
performances6 as access to the core radiolocation band (430-440 MHz, radiolocation on a primary basis) will
be compromised.
Additionally, even though a significant improvement in term of a reduction of the separation distance required
to protect radars could be achieved (compared to the figures reported in ECC Report 240), deployment of
LTE networks without any mitigation techniques will lead to prevent radar operations in large areas and
would harmfully impact them in neighbouring countries and over international waters.
6.6.1 Considerations
Considering that co-channel scenarios need large separation distances, a coordination procedure will be
needed for both ground and airborne radars, but in practice any coordination could be implemented only for
ground radars with fixed location. Such coordination procedure may need mitigation techniques (such as
deterministic propagation model, BS power reduction, antenna discrimination through height or tilt and
azimuth) to facilitate coexistence possibilities on a case by case basis.
In addition, in order to ease the protection of radars in co-channel situation but also to ensure that
radiodetermination applications could continue to operate in the adjacent channel (430-440 MHz band where
radiolocation is a primary service), one of the following solutions has to be applied:
maintaining a frequency guard band below 430 MHz,
LTE systems have to reduce their OOBE levels of around 35 dB and then a 100 kHz guard band could
be sufficient between PPDR LTE and radars. This reduction of OOBE could be made by means of an
external filter or by reducing the BS transmitted power.
It should be noted that a filter achieving a deep attenuation of at least 35 dB within a 1 MHz guard band
would be very challenging to design and could be too expensive. Experience in the 800 MHz band provides
filters achieving an attenuation of 17 dB 1 MHz after its cut-off frequency. In general it is expected that filters
providing 50 to 60 dB of attenuation at 5 MHz further the cut-off frequency could be found on the market at
fair rate. Based on the later filter in can be realistically extrapolated that the dB = 40 dB attenuation is
achievable after f = 2.5 MHz from the filter cut-off frequency.
6 Taking into account that 420-430 MHz is unusable and that 440-450 MHz band (radiolocation on secondary) status, is already
interfered by a heavy use of applications under mobile service.
ECC REPORT <No> - Page 75
Figure 2627: Required frequency for the filter to meet the required attenuation
Annex 13 illustrates how the values of dB were derived, considering different antenna decoupling
attenuation. The 2.5 MHz corresponds to a band edge to band edge guard band as illustrated below,
meaning that the LTE BS center frequency should not be higher than 430 2.5 BLTE/2 MHz.
Figure 2728: Proposed guard band between LTE and radar systems
In order to protect the radars it is suggested that 2.5 MHz guard band should be applied at upper ends of the
410-420/420-430 MHz, assuming an additionnal reduction of 40dB is provided for OOBE of the LTE BS (on
the basis of the value specified in the 3GPP standard)
Any LTE channel with 1.4 MHz, 3 MHz, and 5 MHz bandwidth should be entirely placed in the tuning range
of 412-417.5/422-427.5 MHz applying 100 kHz channel spacing starting at 412.7/422.7 MHz and finishing at
416.8/426.8 MHz (for 1.4 MHz bandwidth).
For the protection of ground radar systems, in accordance with the calculations provided in this contribution
and existing regulatory practice (UK), it is suggested that:
a general exclusion zone of 40 km around radar sites should be required,
co-ordination of LTE base stations should be required within the range of 40-120 km from radar sites,
no co-ordination of LTE base stations should be required above the distance of 120 km from radar sites.
a maximum OOBE level of -47 dBm/100 kHz be set for LTE BS, assuming a minimum 10 dB of antenna
decoupling attenuation with radar receiver (see Table 157 in section A13.9 of Annex 13).
ECC REPORT <No> - Page 76
For airborne radars it is proposed that the 430-440 MHz band be assigned for their operation in locations
where their protection is required.
The impact of LTEin400 activities in the 410-430MHz band on RAS in the 406.1-410 MHz band is studied
with the consideration of emission from base stations (BS) and user equipment (UE). In order to protect the
radio astronomy stations it is suggested that 1 MHz guard band should be applied at the lower end of the
410-420 MHz band. The simulations for the UE therefore assume the transmission band edge at 411 MHz.
The LTEin400 parameters used for this study, as listed in Table 22Table 23, are obtained from Annex 1 of
this Report.
For the analysis, the BAKOM (Switzerland) implementation of the propagation model P.452-16 including free
space, smooth earth diffraction, tropospheric scatter, ducting, and ground clutter attenuations was used. The
atmospheric attenuation was approximated to be 0.0 dB/km. The MCL calculations are performed for single
and aggregate emitters, assuming a flat terrain for radio astronomy stations, and operational bandwidth of 5
MHz for both base stations (BS) and user equipment (UE) with duty cycles of 100%. For indoor usage a wall
attenuation of 11 dB and a general body loss of 4 dB were considered for the user equipment in the
calculations. The deployment of UE is split to 75 % outdoor and 25 % indoor use. Recommendation ITU-R
RA.769-2 provides threshold levels of -203 dBW (or -173 dBm) for interference detrimental to the RAS for
the band 406.1- 410 MHz. The receiver antenna gain is assumed 0 dBi and the typical height of the receiver
is taken to be 50 m.
The results of the compatibility study for flat terrain profile are summarised in Table 23Table 24. In the case
of a single base station emitting at a direct line of sight of a RAS station (i.e., the worst case scenario), the
obtained MCL is 106 dB, which translates to a separation distance of 17 km. In the aggregation study with
statistical Monte-Carlo simulation using 15000 trials with 500 devices and a deployment density of 0.0057
ECC REPORT <No> - Page 78
Aggregate
Single Interferer
Single Aggregate Interference
Bandwith of 5 MHz
Interferer Interference indoo
outdoor indoor outdoor
r
Single emitter
emissions in 406.1 - -97.1 -41.1
-52.1
410 MHz (dBW)
Aggregate power
received by RAS -200.7 -165.6 -149.5
station in 406.1 -
410 MHz (dBW)
Protection Level -
-202.9 -202.9 -202.9 -202.9 -202.9
(dBW) 202.9
MCL (dB) 105.8 92.8 150.8 161.8 153.8 164.8
Separation distance
17 1 34 78 190 326
(km)
Required reduction
in spurious 30.0 56.3 67.3 37.2 53.3
2.2
emissions (dB)
single emitter
emission limits in -103.0 -84.3 -84.3 -74.4 -79.3
RAS band -77.2
(dBm/MHz)
The path-loss profiles for single emitter base stations and user equipment computed with different
propagation models in ITU-R P.452-14, P.452-16, P.525-2, and P.1546 are shown in Figure 28Figure 29.
The separation distances are obtained by intercepting the MCL plot with that of the path-loss profiles. In
conclusion, despite a 1 MHz guard band, compatibility will be very difficult to maintain with the UE due to
high levels of out-of-band emissions. The mobile usage of the UE will make it particularly difficult to protect
the RAS stations from interference.
ECC REPORT <No> - Page 79
Figure 2829: Path-loss profiles for single emitter base stations (left) and user equipment (right)
considering different propagation models ITU-R P.452-14, P.452-16, P.525-2, and P.1546. The reported
calculations results are based on P.452-16
ECC/DEC/(16)02 [28][28] is addressing the harmonised conditions for the implementation of Broadband
Public Protection and Disaster Relief (BB-PPDR) radio systems. Paragraph 3 of Decides section of this
Decision presents several frequency allocation options for introducing BB PPDR in 450-470 MHz band. This
approach of several frequency allocation options could also be used for 410-430 MHz frequency band.
Shifting the operating frequencies of LTE UE could reduce interference on RAS service operating in the
406.1-410 MHz band. Calculations for guard bands larger than 1 MHz were done in order to estimate the
effect of different guard bands between the RAS and possible LTE systems in the 410-430 MHz range.
The studies were done using MCL approach calculating interference from single and aggregate interferer
scenarios for the UE into the RAS band.
The interference from single and aggregate LTE UE Tx having 5 MHz bandwidth was calculated. The guard
band was set from the 411 MHz band edge. The results of the compatibility studies are summarised in the
following Table:
Table 2425: LTEin400-RAS separation distances (km) assuming different guard bands
1 34 78 190 326
2 33 75 185 322
3 31 67 171 312
4 29 57 148 296
5 23 41 103 261
The path loss attenuation between LTE UE and RAS station was estimated using simplified propagation
model provided by CRAF in a document SE7(16)072 and used in ECC ECC Report 240. This document
depicts that for the analysis the propagation model P.452-15, including free space, smooth earth diffraction,
tropo-scatter, and ground clutter attenuations were selected. The atmospheric attenuation was assumed 0.0
dB/km.
ECC REPORT <No> - Page 80
Guard band values between 1 MHz and 5 MHz for single and aggregate LTE UE having 5 MHz bandwidth
and the RAS allocation were studied. Calculations showed that the separation distances needed to protect
RAS stations do not vary drastically over the range of the guard bands, and that compatibility with the
aggregate interference from the UE, particularly outdoor usage, will require very large separation distances.
A more detailed study using SEAMCAT is provided in the next section.
7.3.1 Introduction
ECC Decision (16)02 is addressing the harmonised conditions for the implementation of Broadband Public
Protection and Disaster Relief (BB-PPDR) radio systems. Paragraph 3 of Decides section of this Decision
presents several frequency allocation options for introducing BB PPDR in 450-470 MHz band. These several
frequency allocation options could also be used for 410-430 MHz frequency band. Shifting the operating
frequencies on LTE user equipment (UE) and LTE base station (BS) could reduce interference on RAS
service operating in 406.1-410 MHz band. Following studies were done in order to estimate the effect of the
guard band between the RAS and possible LTE systems in 410-430 MHz frequency band.
In the studies, interference into RAS receiver (RAS Rx) from multiple LTE UE transmitters (UE Tx) and LTE
BS transmitters (BS Tx) was evaluated. The studies were carried out for the maximum bandwidth (5 MHz) for
both base stations and user equipment using SEAMCAT tool v.5.1.0.
The LTEin400 UE and BS 5 MHz bandwidth parameters, as well as RAS parameters used for these studies
are listed in Table 22Table 23 UE density is 0.0342 km2
The parameters were taken from SE7(17)013A1 draft ECC Report 338323 [33][33]. According to this
document, BS deployment density for RAS scenario is defined as 0.0057 km2. This leads to an area of
175,857 km2 occupied by one BS. In SEAMCAT, Tri-sector hexagon type was chosen. Following equation is
used to calculate Cell range - 2R. (In SEAMCAT, this R is defined as Cell radius. Cell range = 2R):
1 1
Deployment density = = 33 2
. (1)
Cell area 3( R )
2
According to the ECC Report 240, average density of active UE is 0.027 km2 for RAS scenario, while
deployment density of BS is 0.0057 km2. This leads to the average of 4 UE per BS. For the studies, Tri-
sector BSs were taken and UEs density was increased to 6 UE per BS or 2 UE per 1 sector. Therefore, UE
density in the study was increased to 0.0341 km2.
Usually LTE networks are not fully loaded, so 10 resource blocks (RB) were taken per UE. UE emission
mask was based on 3GPP TS 36.101 V9.2.0 and adjusted using approach described in Report 240 (Table
62) to correspond with 2 UE per sector using 10 RB.
ECC REPORT <No> - Page 81
BS emission mask is based on 3GPP TS 36.104. Spurious emission level is adjusted to -139 so that it
corresponds to the BS spurious power equal to -96 dBm and BS transmitting power as in Table 1.
A "static" model, also referred to as a "snapshot" Monte-Carlo model, was employed by means of
SEAMCAT. This model sets up a random distribution of users based on one time instant correspondent with
a network configuration and considered service characteristics. A set of statistics which accurately reflects
these scenarios is derived by simulating several such snapshots. Investigation of the coexistence of a BB
PPDR (LTE) network with Radio Astronomy Station (RAS) was performed by using SEAMCAT v.5.1.0.
During the simulations, unwanted signal levels from all BSs and UEs were included and then compared with
the protection level of RAS. RAS protection level used in this study is -173 dBm as in Recommendation ITU-
R RA.769-2 with time percentage of 2 %. ITU-R P.1546-4 and P.452 propagation models were used in ECC
Report 240. Following that SEAMCAT does not take into account terrain profile, therefore propagation model
ITU-R P.1546-4 was used in the studies.
A cellular environment (hexagonal clutter) was used for modelling in SEAMCAT consisting of 19 cells (2-tiers
tri-sector 3GPP layout) and wrap-around technique to reduce the number of cells required in the simulations
of endless network and consequently to enable faster simulation run times. The geographical separation of
the victim receiver and interferer LTE network was modelled by specifying the modelled cellular cell as laying
at the edge of the network and thus as if the cellular system extending to one side only. For this simulation,
left-hand side and right-hand side networks where applied simultaneously. This is illustrated in the Figure 11,
where DE is RAS distance from the network edge, DC is distance from hexagonal clutter centre, RC is clutter
maximum range, and is central (network visibility) angle.
The set of simulations was done using UE power control settings showed in Figure 32Figure 33.
The simulation was performed using 100000 snapshots in each scenario. Interference probability
dependence on the distance from the network edge for various values of guard band is presented in Figure
33Figure 34. Interference criterion I/N was applied for the victim receiver.
Figure 3334: Interference probability dependence on distance from the LTEin400 network edge
The SEAMCAT translation mode was used, to calculate the probability of interference as a function of the
total output power of interferer transmitters. The 5% interference probability level as a trigger was chosen to
evaluate the power supplied value to victim receiver at various RAS distances from the interferer LTE
network edge. The results are presented in Figure 35Figure 36.
Figure 3435: Power supplied dependence on RAS distance from the interferer LTE network edge for
various guard bands to keep 5% interference probability
The increase of power supplied at bigger distances means that the interfering signal probability at victim
receiver is less than the 5 % level, and therefore, the number of interferers can be increased to keep the
ECC REPORT <No> - Page 84
same 5 % probability level of interference. And on the contrary, the less distances can be assured by
decreasing the interferers number.
This way, the maximum number of interferers (2x114 UE + 2x19 BS) used in SEAMCAT simulations can be
easily recalculated to another number by introducing the effective clutter of interferers in the form of the part
of the circular ring of the same width as the hexagonal clutter and having the same area, interferers number
(density) and the distance from the RAS receiver to the ring centre, Figure XXQ.
Figure 3536: Effective area of part of ring to generate the same interfering signal at RAS receiver
360 Seff
eff deg
2 2 RC DC , (2)
3 3 2
S eff 19 R
2 , (3)
K dB 10 log
eff , (4)
where K is the interferer signal power gain factor, is required central (network visibility) angle.
For example, given guard band value 2.5 MHz, two clutters (left and right) area 2Seff = 6682.566 km2, clutter
range RC = 38 km, distance from RAS receiver to clutter centre DC = 64.5 km (distance from clutter edge DE
= DC RC = 27 km), interference probability 5 % and required angle = 360 deg (closed circular ring around
the RAS placeded in the ring centre) power gain factor is K = 6.64 dB. Equating the value of K to the same
value of the power supplied (q.v. Figure 35) gives a recalculated distance from the interferer network edge
that is increased to about 40 km to keep the same 5 % interference probability.
It should be noted that the simulation of interference in SEAMCAT is not confined to the case of two identical
LTE clutters located in diametrically opposite sides of the victim receiver. Their sizes, distances from the
victim's receiver, guard bands and propagation models can be different. Also, the decrease in the number of
interferers (or interferers density) by half will reduce the power supplied by 3 dB, and increasing the number
of interferers twice, will raise the value of the power by 3 dB. Thus, the intersections of the power supplied
ECC REPORT <No> - Page 85
curves with the straight lines at 3 dB or +3 dB will give values of distances with interference at the level of
5% for other values of the number of interferers.
The interference from all LTE BSs and LTE UEs Tx into RAS Rx station was calculated. RAS stations
operate in 406.1-410 MHz frequency band. Therefore, the guard band was estimated from the 410 MHz
point. LTE UE 412.5 MHz centre frequency is considered to use 0 MHz guard band. LTE BS 422.5 MHz
centre frequency is shifted 10 MHz away from the UE frequency.
Separation distance is considered to be the radius of the 1st (closest to RAS) ring of cells. The results of the
compatibility studies are summarised in Table 25Table 26 and Table 26Table 27.
In this contribution, several options of the guard bands between aggregated impact from BB PPDR (LTE)
UEs and BSs into RAS stations were analysed. Performed calculations showed that the separation distance
needed to protect RAS station given its full circular environment are as follows:
at least 43 km using 4 MHz guard band;
at least 41 km using 2.5 MHz guard band;
at least 69 km using 0 MHz guard band.
The calculated separation from one LTE hexagonal clutter consisting of 19 cells are as follows:
at least 25,5 km using 4 MHz guard band;
at least 24,2 km using 2.5 MHz guard band;
at least 37,5 km using 0 MHz guard band.
Studies were done in rural environment with ITU-R P.1546-4 land propagation model. In real life situations,
separation distances could be lower because of the specifics of the propagation environment, e.g. terrain
constraints, clutter obstacles.
7.4 CONCLUSION
Two studies by using different calculation methods were used for evaluation of interference from LTE PPDR
systems operating in 410-430 MHz band into radio astronomy stations in the 406.1-410 MHz frequency
band. One study was done by using SEAMCAT and propagation model ITU-R P.1546-4 with different
network layout when aggregated effect of BSs and UEs were taken into account; another one - using
MATLAB program and the BAKOM (Switzerland) implementation of the propagation model P.452-16. The
previous methodology used in ECC Report 240 involved simplified propagation model ITU-R P.452-14. The
new study with MATLAB also considered a guard band of 1 MHz for base stations and larger guard bands of
up to 5 MHz for user equipment, both indoor and outdoor usage.
Generic compatibility calculations for LTE systems in the 410-430 MHz band and radio astronomy operating
in the 406.1-410 MHz band showed that compatibility may be achievable by implementing emission-free
zones around RAS stations.
Analysis by using SEAMCAT showed that for BB PPDR LTE network completely surrounding RAS station,
exclusion zone extended up to 70 km around RAS when no guard band was used and up to 45 km with 2.5
MHz guard band. Separation distances became smaller when networks layout comprise a part of the ring
placed on one side of RAS. It may shrink down to 38 km without guard band between systems and to 25 km
with 2.5 MHz guard band. Such case could be met when coordination of different systems between two
countries occurs.
Larger protection distance may be introduced around RAS stations, which are deployed in nearly flat terrain
environment. In analysis with MATLAB for the outdoor UE, considering a 1 MHz guard band, the separation
ECC REPORT <No> - Page 87
distances for single emitter and aggregate cases become 78 km and 326 km, respectively. For indoor usage
and additional wall attenuation of 11 dB the separation distances for single emitter and aggregate cases are
reduced to 34 km and 190 km, respectively. The separation distances decrease with larger guard bands; for
example, with a guard band of 5 MHz the separation distances for single emitter and aggregate cases of
outdoor UE become 23 km and 261km, respectively. Moreover if the BS filter is applied to the UE frequency
range and not further, the elevated spurious emissions of -26 dBm/MHz in the RAS band will result in
increased separation distances of more than 500 km between a RAS station and the base stations. List of
RAS stations in Europe operating in the 400 MHz band
Table 2728: RAS stations in Europe operating in the 400 MHz band
Elevation
Observatory Administration Coordinates
(m AMSL)
Effelsberg Germany 06o 5300 E, 50o 3132 N 369
e-callisto solar Germany, Italy, Belgium, Ireland,
network Finland, Czech Republic
Thermopiles Greece 22o 4100 E, 38o 49 00 N
Westerbork Netherlands 06o 3615 E, 52o 5501 N 16
Lustbhel Austria 15o 2934 E, 47o 0403 N 483
Humain Belgium 05o 1519 E, 50o 1131 N 293
Metshovi Finland 24o 2335 E, 60o1304 N 61
Nanay France 02o 1200 E, 47o 2300 N 150
Medicina Italy 11o 3849 E, 44o 3114 N 28
Noto Italy 14o 5920 E, 36o 5233 N 90
Sardinia Italy 09o 1442 E, 39o 2934 N 600
Kayseri Turkey 35 32 43E, 38 42 37 N 1054
Bleien Switzerland 08o 0644 E, 47o 2026 N 469
Cambridge United Kingdom 00o 0220 E, 52o 0959 N 24
Jodrell Bank United Kingdom -02o 1826 E, 53o 1410 N 78
ECC REPORT <No> - Page 88
8.1.2 More realistic estimation Worst-case estimation of sharing and compatibility of LTE and FS
This estimation aims at a more realistic estimation and implies the ITU-R P.452-16 propagation model
between the LTE BS and the FS station. Between the LTE MS and the FS station the
extended HATA propagation model is used. An antenna discrimination as described in 8.3.48.3.4 was
applied.
If more realistic investigation options are used a sharing of a common frequency range will be possible
between LTE BS and the FS if protection distances of about 85 km are kept. Their compatibility if used in
adjacent frequency ranges can be expected, if protection distances of about 35 km are respected.
LTE MS satisfy sharing requirements for operation distances larger than 4 km to the FS station. If used in an
adjacent frequency range, no interference for operational distances larger than 0.5 km is expected.
Table 2930: More realistic estimation of sharing and compatibility of LTE and FS
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 3637: Example result array with continuous transfer of regions of mCL values
Since the interesting 0 cannot be tracked easily red/green arrays were produced for an easier
interpretation. These allow clearly to distinguish between combinations of the frequency separations (f) and
distances providing sufficient coupling loss (in green) from those which do not (in red, see Figure 37Figure
38)
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 3738: Example result array with discrete regions9 of mCL values
7 Considered between ILT and VLR to ensure the operation of the victim system link without interference
9 Note that the scale for this display option still informs about the range of mCL values
ECC REPORT <No> - Page 90
19 eq 1
= 10 ( ) = 12,77
360
In case of the low gain antenna with a gain of 5 dBi and providing an azimuthal half power beamwidth of
150
150) the random probability of pointing in a certain direction will be = 41,66%, or if to be taken into
360
account in a link level equation
150 eq 2
= 10 ( ) = 3,8
360
For the BS the probability to be directed towards a FS station was assumed to be 100%. This assumption
was made because it has to provide a comprehensive coverage.
Remark: These decoupling factors express a statistical unlikelihood and are strictly spoken bound to the
condition of an equal distribution in the azimuthal antenna pointing what might be disturbed by the limitations
of the radio site acquisition. Nevertheless, this method is used to assess the statistical unlikelihood of a main
beam coupling.
Remark: The values for the antenna discrimination according to eq 1 and eq 2 are quite close to the main
beam gain of the related antenna. It can therefore be expected that the impact of the main beam gain on the
result will nearly vanish in case of the antenna mode option antenna discrimination.
It shall be noted that the antenna heights of the LTE BS (hBS = 30 m) and the FS (hFS = 26 m) are not that
different. Taking into account the range of simulation distances l (0.25...100 km) between the stations this
effect will be negligible in all scenarios involving BSs and FS.
Even for the scenarios introducing a LTE MS antenna height of 1.5 m the effect is just faintly to be noticed in
the very first steps of the scenario distances. However it is taken into account.
ECC REPORT <No> - Page 91
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 3839: Influence of the LTE system bandwidth 1.4 MHz and 5 MHz
Figure 38: Influence of the LTE system bandwidth 1.4 MHz and 5 MHzFigure 39: Influence of the LTE
system bandwidth 1.4 MHz and 5 MHz confirms the result formulated in clause 8.4.18.4.1 and shows that an
increase of the LTE system bandwidth will enlarge the frequency range with interference to be expected.
Remark: It shall be noted that the transmitter output power values of the LTE BS serve for a constant power
spectral density with respect to the system bandwidth. This means that for scenarios where the LTE BS is
the ILT the results will be independent from the bandwidth.
Remark: For the LTE MS that is different. Since its transmitter output power will always be 23 dBm the
system with the smallest bandwidth will provide the largest power spectral density. Therefore, if scenarios
involve LTE MS the system bandwidth considered is preferably 1.4 MHz.
ECC REPORT <No> - Page 92
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 3940: Influence of the main beam gain of the FS Antenna in case of main beam coupling
Figure 39: Influence of the main beam gain of the FS Antenna in case of main beam couplingFigure 40:
Influence of the main beam gain of the FS Antenna in case of main beam coupling indicates that the
protection distance will change remarkably if a main beam coupling is assumed. This change exactly equals
the difference in the propagation loss of the model selected.
As expected the influence of the antenna main beam gain nearly vanishes if the antenna discrimination as
described in clause 8.3.48.3.4 is considered (see Figure 40Figure 41). This results from the fact that the
antenna discrimination values are in the range of the main beam gain of the antenna considered.
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 4041: Influence of the main beam gain of the FS Antenna in case of antenna discrimination
ECC REPORT <No> - Page 93
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 4142: Influence of the antenna mode in case of a higher gain antenna
Figure 41Figure 42 shows a difference in the minimum protection distances depending on the option whether
the antenna discrimination as described in 8.3.48.3.4 is used or not.
This effect depends on the main beam gain of the antenna (compare Figure 41Figure 42 and Figure
42Figure 43) and becomes more decisive as the gain increases.
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 4243: Influence of the antenna mode in case of a low gain antenna
ECC REPORT <No> - Page 94
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
The free space propagation model always delivers the most critical results for the evaluation of a scenario. It
might be considered as a worst-case scenario and was therefore also used in clause 8.4.18.4.1. For the
scenario shown in Figure 43Figure 44 no co-frequency use will be possible. Depending on the frequency
separation between the systems the protection distance is 10.25 km (f 3.2 MHz) or 1.75 km
(f > 3.2 MHz).
Figure 44Figure 45 displays the results for the same scenario but using the extended HATA propagation
model. Even for the LTE MS wanted frequency range only a protection distance of about 4 km is necessary.
For frequency separations of f > 0.7 MHz a protection distance of only 250 m is already sufficient to prevent
an interference in the FS receiver caused by the LTE MS.
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 45Figure 46 finally uses Recommendation ITU-R P.452-16 for the prediction of the propagation path
loss. It suggests protection distances of 45 km in the co-frequency case (f = 0), 11 km for 0.8 MHz f
3.2 MHz and 2 km for f > 3.2 MHz.
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
= (f) eq 3
In the fixed service (FS) the use of the NFD approach is very common (see ref 1).
It shall be noted that the masks used in the investigation were extended in their frequency range to cover at
least 10 MHz from their centre frequencies to produce NFDmc results. For that purpose the attenuation of
the outermost frequency points of the mask were considered to be constant over the larger frequency range.
This is certainly a worst-case approach compared to real system behaviour which of course will show an
increasing attenuation considering larger frequency separations f.
The step size in frequency10 used in the study was chosen to be 0.1 MHz.
An example of the frequency decoupling between a LTE-MS using a bandwidth of 1.4 MHz and a FS system
is given in Figure 46Figure 47.
For the scenario MS FS the NFDmc(f=0) = -18,45 dB because the ratio of bandwidths involved:
0.02 eq 4
( = 0) = 10 ( ) = 18,45
1.4
The extreme values of decoupling ( -120 dB) result from a f where half of the FS system mask still
overlaps with the LTE-MS system mask. This is an undefined state of calculation which however doesnt
affect the results since f > 10 MHz are neither considered.
Figure 4647: Decoupling loss between a LTE-MS 1.4 MHz and a FS station as a
function of frequency
Remark: Please note that the NFDmc as used in this document covers an addition compared to the definition
of the NFD in clause 4.2.4 of ref 1: The shape and bandwidth of both masks (Tx and Rx) is taken into
account. According to clause 4.2.4 in ref 1 the NFDmc is always =0 in the co-frequency case. As it can be
seen in Figure 46Figure 47 this is not the case for the NFDmc as used in this document. This issue is also
taken into account and described in ref 4. In its technical meaning the NFDmc can be considered as equal to
the ACLR. The difference is that in case of ACLR an ideal receiver filter (rectangular shaped) is assumed. In
case of NFDmc a realistic filter is used.
In total, 144 scenarios have been produced. Following cases have been chosen to derive the main findings
of the study:
environ
gain co- adjacent
No - propagatio antMod considere
Scenario Figure [dBi frequenc frequency
. n model e d as
] y range
ment
Figure
not
1 BS FS 47Figur any FreeSpace 15 mB worst-case not possible
possible
e 48
ECC REPORT <No> - Page 97
Figure
not
2 FS BS 47Figur any FreeSpace 15 mB worst-case 28.25
possible
e 48
Figure
FS not
3 48Figur any FreeSpace 15 mB worst-case 1
MS possible
e 49
Figure
MS F not
4 48Figur any FreeSpace 15 mB worst-case 8.25
S possible
e 49
Figure
5 BS FS 49Figur rural P.452-16 15 AD use11 case 86 33.75
e 50
Figure
6 FS BS 49Figur rural P.452-16 15 AD use case 50.5 6.5
e 50
Figure
7 BS FS 50Figur rural P.452-16 5 AD use case 83.25 32
e 51
Figure
8 FS BS 50Figur rural P.452-16 5 AD use case 62.25 15.25
e 51
Figure
FS extended
9 51Figur urban 15 AD use case 1.25 -
MS HATA
e 52
Figure
MS F extended
10 51Figur urban 15 AD use case 4 -
S HATA
e 52
Figure
FS extended
11 52Figur urban 5 AD use case 2 -
MS HATA
e 53
Figure
MS F extended
12 52Figur urban 5 AD use case 3.5 -
S HATA
e 53
FS Figure extended
13 urban 15 mB critical case 2.75 -
MS 25 HATA
MS F Figure extended
14 urban 15 mB critical case 9 0.5
S 53Figur HATA
e 54
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
109 100 90 81 72 62 53 43 34 25 15
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
83 74 64 55 46 36 27 17 8 -2 -11
Figure 4748: Worst case assessment of compatibility and sharing between LTE BS and FS using
the freespace propagation model and pointing the higher gain antennas towards each other
Although the worst-case scenarios involving the LTE MS show a significant relaxed estimation, a sharing
between the LTE system and the FS an interference-free operation will also not be possible, see Figure
48Figure 49.
ECC REPORT <No> - Page 99
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
78 69 59 48 38 28 18 8 -2 -12 -22
Figure 4849: Worst case assessment of compatibility and sharing between LTE MS and FS using
the freespace propagation model and pointing the higher gain antennas towards each other
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 4950: More realistic assessment of compatibility and sharing between LTE BS and FS using
the high gain antenna the P.452-16 as propagation model and assuming an antenna discrimination
ECC REPORT <No> - Page 100
As already mentioned in clause 8.3.48.3.4, the impact of the antenna gain performance nearly vanishes if
the antenna discrimination is used (compare scenarios BS FS in Figure 49Figure 50 and Figure
50Figure 51).
The reason for the lower protection distance for the scenario FS BS in Figure 49Figure 50 compared to
Figure 50Figure 51) is the national legislation on the maximum e.i.r.p. for the FS. The output power of the FS
transmitter has to be decreased remarkably compared to the use case with a low gain antenna.
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 5051: More realistic estimation of compatibility and sharing between LTE-BS and FS using
P.452-16 as propagation model, the low gain12 antenna and assuming an antenna discrimination
From the following results, it can be drawn that for use cases of LTE MS and FS stations:
a sharing of a common frequency range is possible with a low risk of interference;
a compatible use in adjacent frequency ranges will doubtlessly be possible without any risk of
interference.
12 When additionally a statistical antenna discrimination is taken into account, please remember that in this case the difference in
antenna gain performance nearly vanishes (as described in clause 8.3.44.9.3.4).
ECC REPORT <No> - Page 101
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 5152: More realistic estimation of compatibility and sharing between LTE MS and FS using
the extended HATA propagation model, the higher gain antenna and assuming an antenna
discrimination
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 5253: More realistic estimation of compatibility and sharing between LTE MS and FS using
the extended HATA propagation model, the low gain antenna and assuming an antenna
discrimination
In the last result array shown in Figure 53Figure 54 a main beam coupling is assumed between the FS
station and the LTE-MS. Remaining the extended HATA propagation model in use this can be considered as
a mixed form (worst-case/use case) of scenario. Even so a noticeable risk might be seen in the sharing.
ECC REPORT <No> - Page 102
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
missing coupling loss versus shift of centre frequencies and distance of stations
distance [km]
0 10 20 30 40 50 60 70 80 90 100
frequency separation [MHz]
1
2
3
4
5
6
7
8
9
10
missing coupling loss scale [dB]
Figure 5354: More realistic estimation of compatibility and sharing between LTE MS and FS using
the extended HATA propagation model, the higher gain antenna in the main beam coupling mode
8.5 REFERENCES
ref 1 ETSI TR 101 854 V1.3.1 (2005-01): Fixed Radio Systems; Point-to-point equipment; Derivation
of receiver interference parameters useful for planning fixed service point-to-point systems
operating different equipment classes and/or capacities
ref 2 ETSI EN 301 390 V1.3.1 (2013-08): Fixed Radio Systems; Point-to-point and Multipoint Systems;
Unwanted emissions in the spurious domain and receiver immunity limits at equipment/antenna
port of Digital Fixed Radio Systems
ref 3 http://confluence.SEAMCAT.org/display/SH/A17.3.2+Indoor-outdoor+propagation
ref 4 Annex 3B Determination of the Masks Discrimination and the Net Filter Discrimination in the
Fixed Service to the AGREEMENT between the Administrations of Austria, Belgium, the Czech
Republic, Germany, France, Hungary, the Netherlands, Croatia, Italy, Liechtenstein, Lithuania,
Luxembourg, Poland, Romania, the Slovak Republic, Slovenia and Switzerland on the co-
ordination of frequencies between 29.7 MHz and 43.5 GHz for the fixed service and the land
mobile service. (HCM Agreement), signed Budapest, 6 November Commented [JC29]: These references should be added
at the end of the document in the last annex of the
report
ECC REPORT <No> - Page 103
This section provides information about a specific compatibility analysis between Analogue PMR and LTE
systems. It illustrates the situation of mobile PMR networks where the BS of the networks can be moved
depending on the need. This is the case for PMSE PMR links used in the Audio-Visual production which
could be located nearby LTE BS such as those used by a PPDR network.
The studies contained in ECC Report 240 [1][1][1][1]considered that there is a limited correlation between
the locations of PMSE and PPDR equipment. During major events, a variety of equipment for wireless
communications is used. Not only for course communications and audio connections, but also
communication equipment for television, police and ambulance authorities. All these applications need
transmitting frequencies. By nature, PPDR equipment are going to be deployed in the same areas where
PMSE operations are going to be deployed. In particular:
PPDR MS/BS equipment could be deployed in the production studios for security reasons, quite close to
the production equipment, and
PPDR MS/BS equipment are expected to be used at the same locations where PMSE will be deployed to
cover some events (sport, concerts), therefore, if there is any potential of interference, interference will
occur each time a PPDR MS/BS is located nearby a PMSE MS/BS. In such a situation, the equipment
will be operated in the same areas or along the same routes, which was not considered in the current
simulations given in ECC Report 240 for Analogue PMR.
It should be noted that PMSE are constrained in terms of location, since they would be located in the vicinity
of the events they are covering or within a studio. In case of events, this implies that there are limited
possibilities for coordination with PPDR equipment deployed to ensure the security during those events.
With regard to the interference resulting from the BS LTEin400 on MS Analogue FM, the studies relating to
Analogue PMR (see ECC Report 240) considered terrestrial MS, while in the case of Audio-Visual
production, the MS may also be airborne in helicopters or in airplanes. The propagation model which was
considered in the existing compatibility study given in ECC Report 240 is Extended Hata (Urban). This
propagation model does not fit in the case where the potential victim is airborne. It should be noted that the
frequencies operated in such cases are internationally coordinated and that the deployment of LTE
equipment in one country may have an impact on the neighbouring countries.
9.3 TDD
The overall channel plan given in Rec. TR 25-08 for the band 406.1 - 430 MHz and 440 - 470 MHz is
provided below:
ECC REPORT <No> - Page 104
3 MHz 2 MHz
Figure 5556: LTEin400 on Analogue FM in compatibility studies [Source: Figure 7 of ECC Report 240]
This leads to a guard band of more than 10 MHz between the MS considered in the studies, and 5 MHz
while investigating the impact of PPDR BS on PMSE BS. However, it should be noted that analogue FM BS
and MS could be deployed not in conformity with the CEPT arrangement given in Rec. TR 25-08 resulting
in a much smaller frequency separation offset. In particular, frequencies might be used by T DD systems
which mean that a given frequency could be operated by both MS and BS for links of production.
The PMSE equipment uses a raster of 12.5 kHz and/or 25 kHz. The corresponding characteristics as given
in ECC Report 240 are considered also taking into account of the following:
The BS of the PMSE network used in Audio-Visual production is connected with various receivers such as
crew, motorbikes, helicopters, airplanes This implies that the directivity of the antenna is limited. Typical
BS antennas are omnidirectional with a maximum antenna gain of 3 dBi. The maximum e.i.r.p. for the BS is
13 dBW.
In typical scenarios, to reduce the potential interference to other users of the spectrum, the power may not
exceed 0 dBW for the MS.
ECC Report 240 considered the scenario as given in Figure 55Figure 56.
In the case of PMR links used for Audio-Visual production and PPDR equipment, it is expected that the
equipment is located nearby. Therefore, this section provides MCL calculations to derive the necessary
separation distances.
MCL calculations are provided in this section for 12.5 kHz PMSE in Annex 4.
ECC REPORT <No> - Page 105
The studies in ECC Report 240 considered a frequency offset of more than 10 MHz between the PPDR MS
and the PMSE MS. It implies that the transmitter (Tx) power of the system is going to be -36 dBm in 100 kHz.
The following table provides the results of calculations to assess the separation distances considering a
frequency offset of 10 MHz. It should be noted that considering the Extended Hata (Urban) model, the
distance is more than 64 meters. Additional propagation models are also considered for the purpose of
comparison.
Assuming 4 dB body loss, it can be seen that considering a frequency offset of 10 MHz, the separation
distances between the PPDR MS and the PMSE MS would be at least 60 m.
For this case, ECC Report 240 considered the two systems operating in adjacent blocks of spectrum. This is
further considered in the following tables where an offset of 500 kHz is considered between the frequency
used by the PMSE BS and the edge of the PPDR UE band.
Table 3233: LTEin400 MS impact on 12.5 kHz analogue FM BS offset 500 kHz 3GPP TS 36.101 and
3/5 RB
ECC REPORT <No> - Page 106
Gain -3 dB -3 dB -3 dB -3 dB
Body loss 4 dB 4 dB 4 dB 4 dB
e.i.r.p. in the
direction of the -22.74 dBm in 8 kHz -25.74 dBm in 8 kHz -27.74 dBm in 8 kHz -38 dBm in 8 kHz
PMSE equipment
PMSE sensitivity -120 dBm -120 dBm -120 dBm -120 dBm
PMSE sensitivity +
-117 dBm -117 dBm -117 dBm -117 dBm
3 dB
C/I 21 dB 21 dB 21 dB 21 dB
Attenuation to
118.26 dB 115.26 dB 113.26 dB 103 dB
meet the criterion
The separation distances are ranging from 350 m to about 1 km assuming 4 dB body loss.
At a frequency offset of 1 MHz, the Tx power will be -10 dBm in 1 MHz, therefore a distance of 0.554 km and
0.35 km considering 4 dB body loss.
In this case, the systems are assumed to operate in adjacent block. The following table provides results for
an offset of 500 kHz.
Table 3334: LTEin400 BS impact on 12.5 kHz analogue FM MS 500 kHz frequency offset
e.i.r.p. in 8 kHz -2.54 dBm in 8 kHz -4.64 dBm in 8 kHz -5.67 dBm in 8 kHz
PMSE sensitivity -117 dBm -117 dBm -117 dBm
PMSE sensitivity +
-114 dBm -114 dBm -114 dBm
3 dB
C/I 21 dB 21 dB 21 dB
ECC REPORT <No> - Page 107
Table 3435: LTEin400 BS impact on 12.5 kHz analogue FM MS 1 MHz frequency offset
Offset of 1 MHz
Parameter 1.4 MHz 3 MHz 5 MHz
(ECC DEC(16)02
e.i.r.p. -43 dBm in 100 kHz
Tx power -8.14 dBm in 100 kHz -8.3 dBm in 100 kHz -8.4 dBm in 100 kHz
e.i.r.p. in 8 kHz -53.97 dBm in 8 kHz -6.11 dBm in 8 kHz -6.3 dBm in 8 kHz -6.37 dBm in 8 kHz
PMSE sensitivity -117 dBm -117 dBm -117 dBm -117 dBm
PMSE sensitivity
-114 dBm -114 dBm -114 dBm -114 dBm
+ 3 dB
C/I 21 dB 21 dB 21 dB 21 dB
Attenuation to
81 dB 128.89 dB 128.7 dB 128.6 dB
meet the criterion
At 1 MHz, the separation distances are more than 1 kilometre except if the e.i.r.p. limit of -43 dBm in 100 kHz
given in ECC DEC(16)04 is implemented.
The following table provides the separation distances assuming an offset of at least 5 MHz as in ECC Report
240 [1][1] below.
Table 3536: LTEin400 BS impact on 12.5 kHz analogue FM BS 5 MHz frequency offset
Tx power -16 dBm in 100 kHz -15 dBm in 100 kHz -14 dBm in 100 kHz
e.i.r.p. in 8 kHz -54 dBm in 8 kHz -14 dBm in 8 kHz -13 dBm in 8 kHz -12 dBm in 8 kHz
PMSE sensitivity -120 dBm -120 dBm -120 dBm -120 dBm
PMSE sensitivity
-117 dBm -117 dBm -117 dBm -117 dBm
+ 3 dB
C/I 21 dB 21 dB 21 dB 21 dB
Attenuation to
87 dB 124 dB 125 dB 126 dB
meet the criterion
It can be seen that the separation distances achieved using as a reference the level given in ECC
DEC(06)02 (e.i.r.p. limit of -43 dBm in 100 kHz) are much smaller than when considering the levels given in
the 3GPP reference document (GPP TS 36.104 see Annex 2).
9.5.3 TDD
In case TDD links are considered, then both the MS and the BS will be operating on the same frequency,
resulting in frequency offsets smaller than those considered in ECC Report 240. This section considers the
case of PPDR MS on PMSE MS and PPDR BS on PMSE BS.
The following table provides the results for an offset of 500 kHz.
Table 3637: LTEin400 MS impact on 12.5 kHz analogue FM MS 3GPP TS 36.101 and 3/5 RB 500
kHz offset
Gain -3 dB -3 dB -3 dB -3 dB
Body loss 4 dB 4 dB 4 dB 4 dB
e.i.r.p. in the
direction of the
-20 dBm in 8 kHz -23 dBm in 8 kHz -25 dBm in 8 kHz -38 dBm in 8 kHz
PMSE
equipment
ECC REPORT <No> - Page 109
PMSE sensitivity
-114 dBm -114 dBm -114 dBm -114 dBm
+ 3 dB
C/I 21 dB 21 dB 21 dB 21 dB
Attenuation to
meet the 112.26 dB 109.26 dB 107.26 dB 97 dB
criterion
The following table provides the results for an offset of 500 kHz.
Table 3738: LTEin400 BS impact on 12.5 kHz analogue FM BS 500 kHz frequency offset
e.i.r.p. in 15 kHz -2.54 dBm in 8 kHz -4.64 dBm in 8 kHz -5.67 dBm in 8 kHz
PMSE sensitivity -120 dBm -120 dBm -120 dBm
PMSE sensitivity +
-117 dBm -117 dBm -117 dBm
3 dB
C/I 21 dB 21 dB 21 dB
I -138 dBm -138 dBm -138 dBm
Gain MS PMSE 3 dBi 3 dBi 3 dBi
Attenuation to
138.46 dB 136.36 dB 135.33 dB
meet the criterion
Distances 22.4 km 20 km 18.8 km
Considering scenarios similar to those considered in ECC Report 240, the separation distances are quite
large. The implementation of the limit given in ECC DEC(16)02 (e.i.r.p. of -43 dBm) at 1 MHz allows to
significantly reduce those separation distances.
For scenarios where the PMSE and the PPDR systems are operated in adjacent blocks, the separation
distances are quite high (for example about 2 km for PPDR BS on PMSE MS with 500 kHz frequency offset).
The level for the unwanted emissions in the 1 MHz outside from the LTEin400 band should be clarified since
ECC REPORT <No> - Page 110
the operation of the PMSE may be limited in this frequency range due to the risk of interference resulting
from LTE systems.
Due to possible co-location of PPDR equipment and PMSE, the potential of interference is quite high, this is
further considered in the following section.
The scenarios considered in this section are based on those considered in ECC Report 240. It is noted that
the frequency ranges identified for PPDR in ECC DEC(06)02 are slightly different.
In ECC Report 240, the situation described in the following figure was considered.
The range can offer national flexibility, e.g. in the context of additional spectrum beside the 700 MHz range.
1.4 MHz, 3 MHz, and 5 MHz LTE FDD channeling arrangements could be implemented in the paired
frequency arrangements in 450.5-456.0 MHz / 460.5-466.0 MHz and 452.0-457.5 MHz / 462.0-467.5 MHz.
These options are based on a set duplex spacing of 10 MHz.
In addition, in the framework of Work Items SE7_25 and SE7_26, the frequency range 406.1-430 MHz is
under consideration.
However, the conclusions will still be valid considering the frequency offset between the possible interferer
and the possible victim.
The three-sector LTEin400 BS transmits at 461.5 MHz whereas the analogue FM MSs receives signals
coming from 25 kHz Analogue FM BS transmitting between 463 and 465 MHz. The victim frequency is
randomly chosen (discrete distribution option) in SEAMCAT.
The limit for a frequency offset of -43 dBm at 1 MHz for the e.i.r.p. of the BS is also implemented in the
simulations.
In order to consider the possible co-location of the PPDR BS and the PMSE BS, simulations are run
considering separation distances lower than 100 m.
The following table provides the interference probability as calculated with SEAMCAT.
Analogue FM
Interference Probability Reference
bandwidth
ECC Report 240 + -43 dBm e.i.r.p. at 1
12.5 kHz 1.2 % (PPDR BR radius: 3.75 km) MHz
Victim in adjacent block (2 MHz)
-43 dBm e.i.r.p. at 1 MHz
49.2 % Victim in adjacent block (2 MHz)
Distance less than 100 m
ECC REPORT <No> - Page 111
Analogue FM
Interference Probability Reference
bandwidth
-43 dBm e.i.r.p. at 1 MHz
94.4 % Victim in adjacent block (1 MHz)
Distance less than 100 m
-43 dBm e.i.r.p. at 1 MHz
1 MHz guard band (victim in 1 MHz
4.7 %
block)
Distance less than 100 m
ECC Report 240 + -43 dBm e.i.r.p.
1 % (PPDR BR radius: 3.75 km) at 1 MHz
Victim in adjacent block (2 MHz)
-43 dBm e.i.r.p. at 1 MHz
49.1 % Victim in adjacent block (2 MHz)
25 kHz
Distance less than 100 m
-43 dBm e.i.r.p. at 1 MHz
94 % Victim in adjacent block (1 MHz)
Distance less than 100 m
-43 dBm e.i.r.p. at 1 MHz
1 MHz guard band (victim in 1 MHz
4%
block)
Distance less than 100 m
The three-sector LTEin400 BS transmits at 461.5 MHz whereas the analogue FM MSs receives signals
coming from 25 kHz Analogue FM MS transmitting between 453 and 455 MHz. The victim frequency is
randomly chosen (discrete distribution option) in SEAMCAT.
The limit for a frequency offset of -43 dBm at 1 MHz for the e.i.r.p. of the BS is also implemented in the
simulations.
In order to consider the possible co-location of the PPDR BS and the PMSE BS, simulations are run
considering separation distances lower than 100 m.
The following table provides the interference probability as calculated with SEAMCAT.
Table 3940: LTEin400 BS impact on Analogue FM BS (12.5 kHz and 25 kHz) - 5 MHz guard band
Analogue FM
Interference Probability Reference
bandwidth
Analogue FM
Interference Probability Reference
bandwidth
Victim in 2 MHz block with 5 MHz
guard band
Distance less than 100 m (Victim in
453-455 MHz)
ECC Report 240 + -43 dBm e.i.r.p. at
0.55 % (PPDR BS radius: 3.75 1 MHz
km) Victim in 2 MHz block with 5 MHz
25 kHz guard band
-43 dBm e.i.r.p. at 1 MHz
Victim in 2 MHz block with 5 MHz
66 % guard band
Distance less than 100 m (Victim in
453-455 MHz)
The following table provides the TDD case, i.e. the PMSE MS could be transmitting in the 2 MHz adjacent to
the block where the PPDR BS is transmitting.
Table 4041: LTEin400 BS impact on Analogue FM BS (12.5 kHz and 25 kHz) - adjacent block
Analogue FM
Interference Probability Reference
bandwidth
It is not clear which coordination threshold is applied at the border in case of coordination between PMR
Analogue systems and LTE systems. This should be clarified.
In case of frequencies used by airborne PMSE (airplane, helicopters), it is not clear how the coordination
process will be managed. In such a case, the attenuation to protect PMSE equipment will be more than 180
dB resulting in separation distances of thousands of kilometres. Therefore, the deployment of LTE equipment
in one country will have an impact on the possibility for PMSE airborne equipment to operate on the same
frequencies in the neighbouring countries.
9.8 CONCLUSIONS
Coordination:
It is not clear which coordination threshold are applied at the border in case of coordination between
Analogue PMR systems and LTE systems. This should be clarified.
The deployment of LTE equipment in one country will have an impact on the possibility for PMSE
airborne equipment to operate on the same frequencies in the neighbouring countries, therefore, the
coordination process for such cases should be clarified.
ECC REPORT <No> - Page 114
In Germany and France, e*Message operates paging service applications in the 430 470 MHz band as
Narrow-band Point-to-Multipoint (nP2M) systems. nP2M systems are unidirectional radio systems for digital
data that pages all or groups of appropriately equipped receivers in a predefined area and delivers short
messages. They provide a solution for the delivery of short messages to large receiver populations, and are
optimized to guarantee very high levels of reliability and latency as required by safety services, to be cost -
efficient, and to support energy-efficient operation of low-cost receivers. A typical application of nP2M
systems is cost-efficient alerting services for European citizens.
Applications based on nP2M radio systems comprise but are not limited to:
alerting services for first responders and/or service personal for critical infrastructure in case of disasters or
any other kind of events of relevance;
unidirectional information services supporting applications in the area of smart energy management e.g.
secure and reliable control of distributed small load and production facilities in the low voltage grid level
update or maintenance of information provided to industry and consumer products (weather stations,
cognitive pilot channel for radio equipment without bidirectional connectivity (e.g. PMSE equipment)).
nP2M can provide better coverage and higher availability than competing alerting systems (e.g. sirens, send
emails or SMS, TETRA with call out service), and the limited complexity of the receivers allows an
economically reasonable application of nP2M in applications than other systems. In addition, battery stand -
by-times in the order of month or years can be realized that are often required by fire departments, utility
companies, etc.
A recent ETSI report [1] lists the societal benefits of nP2M systems and provides an overview on the market
potential for the European market.
Some million fixed-frequency receiver terminals13 are using radio paging service networks in Germany and
France. In the whole national territory of Germany approximately 800 paging base stations send information
in the simplex mode to many mobile and fixed receivers. In France about 400 paging base stations send
information in the simplex mode to many mobile and fixed receivers.
Among a wide range of other applications, todays paging applications in Germany cover e.g.
Alerting services for fireman, ambulances, medical staff, service personal of utility companies, some of which
are safety critical and require a very high level of availability, outdoor and indoor coverage, and support of
paging receivers carried at the body,
Remote control for secure smart energy grid operation where the paging receivers are often located in
basements.
13 receivers
ECC REPORT <No> - Page 115
Paging services use 20 and 25 kHz carriers in the 430 MHz 470 MHz band as a tuning range (where the
sub-bands 448,4125 MHz 448,4375 MHz, 448,4625 MHz 448,4875 MHz, 465.96 MHz - 465.98 MHz,
466.0375 MHz - 466.0625 MHz, 466.0625 MHz - 466.0875 MHz, 466.1625 MHz 466.1875 MHz, 466.22
MHz - 466.24 MHz and 466.19375 MHz 466,21875 MHz are identified for paging systems) and are thus
potentially affected by LTE systems.
The following tables provide an overview of the characteristics for paging systems.
Bandwidth 20 / 25 kHz
Germany
448,4125 MHz
448,4375 MHz
430 MHz 470 MHz
448,4625 MHz
Carrier (specification/operation in Germany) frequency
448,4875 MHz
(Tuning range)
France
466.19375 MHz
466,21875 MHz
Modulation 2FSK
Parameter Value
ECC REPORT <No> - Page 117
Parameter Value
Blocking immunity, PS 60 dB
Intermodulation immunity, PS 55 dB
Possible spectrum arrangements are covered in Section 2.3. Options 4-6 are considered here.
To reflect the typical characteristics of LTE networks, the working document considered mentioned in section
2.1 above, other than those deployed for PPDR purposes and described in ECC scenario where specific
parameters are applied for compatibility studies. This is reflected in the following table.
43 dBm/5 MHz
Transmit Power
41 dBm/3 MHz
(Report ITU-R M.2292) [5]
37.5 dBm/1.4 MHz
Antenna gain
15 dBi (Note 2)
(Report ITU-R M.2292)
Antenna height
30 m
(Report ITU-R M.2292)
(3GPP specification TS
Spurious power
36.104)
Antenna pattern/Number of
Directional/3
sectors (Note 2)
ACLR
Note 1: Considering two power amplifiers transmit power is up to up to 47 dBm. (For one amplifier, the power
is 44 dBm.)
Note 2: Antenna gains between 6.5 to 13 dBi are considered in this Report. In ECC Report 240, 13 dBi was
used, typical directive antenna (15 dBi) including cable loss (2 dB). Kathrein product datasheet (742 242)
used to created SEAMCAT antenna patterns.
Note 3: Report ITU-R M.2292 defines 3 dB, but in ECC Report 240 value of 2 dB was used, since according
to manufacturers it is more realistic for the 400 MHz range
Note 4: -96dBm/100kHz is the 3GPP requirement for protection of own UL band. BS emissions are expected
to be equal or lower than this level at the RAS frequencies due to rejection from the TX filter.
For the unwanted emission of the BS, the following tables are considered.
1.4 MHz
2.1 to 3.5 -11 dBm 100 kHz
3 MHz
4.5 to 7.5 -15 dBm 100 kHz
The BS spurious emission are -36 dBm in 100 kHz from Table 6.6.4.1.2.1-1. in TS 36.104.
In the SEAMCAT simulations only the impact of the LTE450 BS on a pager receiver is considered.
The characteristics of the 3 MHz / 41 dBm stations given in table 3 are considered.
The BS is assumed to operate at 463.5 MHz while the paging link is operated at 465.97 MHz (offset of 970
kHz at the edge of the LTE band).
Table 454: Simulation results for Paging Receiver and LTE BS with 3 MHz bandwidth
It should be noted that the dominant effect is resulting from the unwanted emissions.
10.5 CONCLUSIONS
The results of SEAMCAT simulation considering the impact of BS on paging receiver indicate a high level of
interference. In particular, a decrease of the level of unwanted emissions would be needed to improve the
compatibility or outage due to interference may be reduced by repeating the messages. The simulations
considered an offset of 970 kHz, if the offset between the paging frequency and the edge of the LTE system
is smaller, then, the probability of interference will be higher.
Additional simulations may be needed to consider the impact of 5 MHz and 1.4 MHz systems and the impact
of the MS.
ECC REPORT <No> - Page 121
Automotive wireless systems on the bases of non-specific short range devices are according to ETSI EN 300
220-1 and -2 [8][8]. These SRDs operate at 433 MHz and are used to enable remote keyless entry systems
and tire pressure monitoring systems. The purpose of this study is to evaluate the impact of the LTE
operating below 430 MHz on the automotive systems operating in the 433 MHz SRD band (ERC/REC 70-03
[7][7]).
The most critical scenario by far is the definition for permitted spurious emissions of -36 dBm @ 100 kHz.
A limit for spurious emissions of -96 dBm @ 100 kHz would solve the problem (see also chapter
11.4.611.4.6).
As such, the scenario described in chapter 11.4.511.4.5 that involves an interference probability (= engine
does not start) of approx. 83 % is reduced to 0.001 % (chapter 11.4.611.4.6)!
With regard to the SRDs, there is no technical possibility of becoming significantly more robust in the future
to resist the impact of spurious emissions attributed to LTE 400 (a smaller bandwidth would also result in a
lower data rate and at best would give rise to just a few dB of improvement).
whereby:
MCINT = 0 dB
GINT = -3 dBi
whereby:
MCINT = 0 dB
GINT = 11 dBi
whereby:
f (dBcINT, PINT) = -33.8 dBm =-36 dBm @ 100 kHz + 10*log (165 kHz/100 kHz)
GINT = 0 dBi
(power above is emitted from antenna -> antenna gain is not required here)
ECC REPORT <No> - Page 123
C/IVICT = 13 dB
-> This interference can also not be counteracted with a better blocking response by the receiver! This
measure would only bring about an improvement for future wireless concepts in the aforementioned
scenarios in chapter 11.111.1.
-> The permissible spurious emissions limit would have to be reduced by 60 dB to ensure that the above
scenario does not lead to any problems:
Permissible spurious emissions: -96 dBm @ 100 kHz
Adding to this is the fact that every individual has different bodily dimensions, creating even more variations
of possible locations of a key in the vehicle. The preferred driving position also differs from person to person,
giving rise to still greater variance.
If one considers the possible propagation paths of an electromagnetic wave emitted by a key, additional
dispersions also arise:
different vehicle geometries;
different load states in the vehicle (1 individual, several individuals, with luggage, etc.);
driving position (height);
distance between front and rear seats;
variation in vehicle bodies, size of glass roof, steel roof, convertible;
...
ECC REPORT <No> - Page 124
All aforementioned items lead to the situation whereby one is not in a position to take reproducible
measurements in the vehicle.
This is why at Daimler, a decision was made to use the level distribution for evaluation purposes.
For in-vehicle measurements in a defined vehicle area (e.g. seat surface), the key is placed in different
positions as a permanent transmitter while at the same time continually moved in its temporal coordinates.
The following excerpt from the measurement description illustrates the approach by example.
At the same time, the power received by the vehicle antenna (50 ohms) is recorded. The following graphic
includes a practical measurement example.
Transmitter: Key with a transmission power of -24 dBm e.i.r.p. (worst case)
Readings: Reception level at 50 ohms directly at the antenna base
Key moved in the vicinity of the driver's seat
Receiving antenna: rear window
ECC REPORT <No> - Page 125
Messwerte values
real measurement
-40 dBm
-50 dBm
-60 dBm
-70 dBm
-80 dBm
-90 dBm
-100 dBm
0,0s 2,0s 4,0s 6,0s 8,0s 10,0s 12,0s 14,0s 16,0s 18,0s 20,0s
The probability density of the received levels for the above measurement is:
ECC REPORT <No> - Page 126
Probability density of
Wahrscheinlichkeit frreal measurement
geringere Pegel
1000 %
100 %
10 %
1%
0,1 %
0,01 %
-110 dBm -100 dBm -90 dBm -80 dBm -70 dBm -60 dBm -50 dBm
It states that e.g. 1 % of all received levels quantified were below -80 dBm.
Previous experience shows that the basic progression of the probability density is always the same and is
independent of the measurement location in the vehicle. It is described sufficiently in mathematical form
1 mn
Messwert _ m
1
P dBm PMittelwert dBm1, 5 9d B
PMittelwert
1 e 10
10
p A P n m 1
whereby
Pmean value represents the averaged received level across all "dBms" (this does not include the mean
physical power. This type of "averaging" has different reasons that will not be discussed in further detail
here).
Due to the good conformity with the prevailing circumstances, the following assumes the above relationship
for the probability density.
11.3.1 Example: Key (e.g. -20 dBm) and interferer (e.g. -36 dBm)
As discussed in the previous chapter, the key can assume almost any conceivable position or location in the
vehicle. The same must also be assumed for a cellular phone, however.
The key and cellular phone therefore have the same progressional curve for the probability density function.
It is also assumed that the spurious emissions can be compared to white noise, which leads to a
necessary noise level signal for the useful signal receiver (vehicle) of
S/N =+13 dB;
The key itself has a transmission power of
-20 dBm e.i.r.p.
Reception S/N = 13 dB
Successful
Figure 62: Example of Probabiltiy density for key (-20 dBm) and LTE mobile (-36 dBm@100 kHz
spurious emission), while MS and key are located in the same area
As the key and cellular phone do not have a defined position with respect to each other and are accepted at
random, the key and cellular phone levels received by the vehicle antennas are also distributed at random.
Let us assume that the key generates a level of -80 dBm at the vehicle antenna. The vehicle receiver can
only successfully receive this signal if
Its sensitivity is sufficient (typically approx. -105 dBm)
The minimum required distance between the useful signal and interference signal (here: 13 dB) is
fulfilled. (see red arrow at top)
If one now carries out a projection with respect to the probability density for the cellular phone (green line
from tip of arrow to red curve), there is a probability of approx. 1.3 %. I.e. 1.3 % of all base noises from the
ECC REPORT <No> - Page 129
cellular phone are sufficiently low and do not cause any interference. HOWEVER, all remaining levels cause
a failure consistent with
A quantitative statement, however, can only be made if one also takes into account the density distribution of
the respective received level. The previous example only uses a single received level. The frequency wit h
which this takes place is therefore not yet known.
This gap can be closed by leveraging the density distribution. Afterwards (bottom graphic, in reference to
1 dB), the frequency with which the key signal is received at a level of -80 dBm (-79.5 dBm to -80.5 dBm) is
approx. 0.15 %; by contrast, approx. 8 % of all received levels lie within the range of -59.5 dBm to -60.5
dBm.
Figure 63: Example of received level density for key (-20 dBm) and LTE mobile (-36 dBm@100 kHz
spurios emission), while MS and key are located in the same area
Refer to the red plotted line for the circumstances surrounding the interference signal.
f A P e 10
10
10
If one also takes into account the frequency/density distribution, this gives rise to a relationship that can be
used to estimate the failure probability.
The Keyless-Go system is also involved in an engine start. If the radio signal path between the key and
vehicle is compromised due to interference, the engine cannot be started.
11.4.1 Worst case (minimum permissible useful signal, maximum permissible interference signal)
Further assumption: The key and cellular phone are in almost the exact same location (e.g. the driver is
carrying the key and cellular phone on his person). As such, it does not make a difference whether the
cellular phone is in the left pants pocket and the key in the right pocket, or the devices are stored in similar
locations (e.g. chest pocket).
Table 4644: Calculation with worst case levels (minimum permissible useful signal, maximum
permissible interference signal)
Useful
Key Mean reception performance -62.00 dBm
signal
Interference
LTEin400 Mean reception performance -76.00 dBm
signal
ECC REPORT <No> - Page 131
44.269%
Remaining/residual as above.
Table 4745: Calculation for typical useful signal, maximum permissible interference signal
Useful
Key Mean reception performance -60.00 dBm
signal
Interference
LTEin400 Mean reception performance -76.00 dBm
signal
33.386%
Remaining/residual as above.
Table 4846: Calculation for useful signal best case, maximum permissible interference signal
Useful
Key Mean reception performance -55.00 dBm
signal
Interference
LTEin400 Mean reception performance -76.00 dBm
signal
13.681%
Remaining/residual as above.
Table 4947: Calculation for typical useful signal, interference signal 10 dB below limit
Useful
Key Mean reception performance -60.00 dBm
signal
Interference
LTEin400 Mean reception performance -86.00 dBm
signal
4.773%
Unlike the previous assumptions, here the LTEin400 device is NOT in the immediate vicinity of the key, but is
e.g. on the rear seat and therefore has an up to 10 dB lower path attenuation:
Table 5048: Calculation for typical useful signal, maximum permissible interference signal
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
bandwidth. As such, the value for 100 kHz represents a good estimate.
83.366%
ECC REPORT <No> - Page 134
11.4.6 Typical useful signal, spurious emission interference signal: -96 dBm
Same scenario as in the previous chapter; only the permissible spurious emissions are reduced
from -36 dBm @ 100 kHz to -96 dBm @ 100 kHz.
Table 5149: Calculation for typical useful signal, spurious emission interference signal: -96 dBm
0.001%
ECC REPORT <No> - Page 135
12 CONCLUSION
[The studies carried out for PPDR base stations in 400 MHz in ECC Report 240 concluded on a set of out-of-
band emissions from base stations. The scenario for this report is similar and therefore the same limits
should apply. However, additional mitigation measures may be required to solve possible residual
interference from LTEin400 BSs on a case by case basis in a manner similar to the situation between
LTE800 and DTT (see also ANNEX 2:ANNEX 2: (list of mitigation measures)).
LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz are given in Table 52Table
50 below.
Table 5250: LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz
The maximum power of the LTE 400 user equipment was assumed to be 23 dBm e.i.r.p.
Compatibility between the LTEin400 UE and DTT above 470 MHz has been studied. It is concluded that
compatibility can be achieved by limiting the unwanted emissions as shown in the following table:
Table 5351: LTE 450 MHz UE maximum mean unwanted emissions for protection of DTT
]
ECC REPORT <No> - Page 136
A1.1 LTEIN400
The following tables provide technical information regarding LTEin400 mobile and base stations. Those
values are derived from existing 3GPP LTE frequency bands leveraging the recent introduction of band 31
(452.5-457.5 / 462.5-467.5 MHz). Reference documents are:
3GPP TS 36.101 v12.5.0 (2014-09) [12][12]; 3rd Generation Partnership Project; Technical Specification
Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA); User Equipment
(UE) radio transmission and reception (Release 12);
3GPP TS 36.104 v12.5.0 (2014-09) [13][13]; 3rd Generation Partnership Project; Technical Specification
Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA); Base Station (BS)
radio transmission and reception (Release 12)
Table 5452:
Table 5554: LTEin400 MS emission limits (From Table 6.6.2.1.1-1 in TS 36.101 [12][12])
Table 5655: LTEin400-3 MHz MS emission limits when transmission is limited to 3 or 5 RBs
-55
-60
-65
-70
-75 3GPP Min. Requirements
-80
-85
-90
0 1 2 3 4 5 6
Delta fOOB (MHz)
Table 5756: LTEin400 MS spurious emissions limits (From Table 6.6.3.1-2 in TS 36.101 [12][12])
Table 5857: UE Spurious emissions limits for protection of own UE receiver for 3GPP Band 31
(From Table 6.6.3.2-1 in TS 36.101 [12][12])
Spurious emission
E-
UTRA Maximum
Frequency range MBW
Band Protected band Level Note
(MHz) (MHz)
(dBm)
FDL_hig
E-UTRA Band 3 FDL_low - -50 1 2
h
Note: Band 31 UL emissions towards Band 31 own Rx is -50dBm/MHz. The same requirement applies for UE UL emissions to
protect own Rx
ECC REPORT <No> - Page 138
Channel width
Channel width
A1.2 TETRA
The ETSI standard ETS 300 392-2 [16][16] has been used to obtain most of the TETRA system parameters.
This standard is titled Radio Equipment and Systems (RES); Trans-European Trunked Radio (TETRA);
Voice plus Data (V+D); Part 2: Air Interface (AI). Those parameters which cannot be obtained from the
standard are assumed values believed to accurately model operational TETRA systems. Following Tables
list all of the parameters required by the Monte Carlo simulation to model a TETRA system.
Table 6564: Unwanted emissions for the TETRA System (measurement bandwidth of 18 kHz)
Frequency Offset 30 dBm Mobile 35 dBm Mobile 40 dBm Mobile 40 dBm Base
Station Station Station Station
Frequency Offset 30, 35, 40 dBm Mobile Station 40 dBm Base Station
A1.3 TETRAPOL
TETRAPOL parameters have been obtained from the TETRAPOL's publicly available specification and are
reflecting ETS 300-113 [20][20] related radio systems.
Table 6867 : Receiver blocking for the TETRAPOL system (MS and BS)
Table 6968 : Unwanted emissions (dBm) for the TETRAPOL system (in 8 kHz bandwidth)
The ETSI standards ETS 300 086 [19][19] and ETS 300 113 [20][20] have been used to obtain information
regarding 25 kHz FM system parameters. Other parameters are assumed values believed to accurately
model operational FM systems. Following Tables list all of the parameters required by the Monte Carlo
simulation to model a 25 kHz FM system.
(1): The first values were taken from ECC Report 099 [14][14]; the second ones were in this report following guidance received from
ETSI ERM.
ECC REPORT <No> - Page 142
Table 7170: Unwanted emissions for the 25 kHz FM PMR systems (measurement bandwidth of 18
kHz)
Table 7271: Receiver blocking values for the 25 kHz FM PMR systems
The ETSI standards ETS 300 086 and ETS 300 113 have been used to obtain information regarding 20
kHz FM system parameters. Other parameters are assumed values believed to accurately model
operational FM systems. Table 73Table 72,
Table 74
Table 73 and
Table 75
Table 74 list all of the parameters required by the Monte Carlo simulation to model a 20 kHz FM system.
Table 7473: Unwanted emissions for the 20 kHz FM PMR systems (measurement bandwidth of 12
kHz)
Linear interpolation (in dB) is used between 20 kHz and 100 kHz
Table 7574: Receiver blocking values for the 20 kHz FM PMR systems
The ETSI standards ETS 300 086 and ETS 300 113 have been used to obtain information regarding 12.5
kHz FM system parameters. Other parameters are assumed values believed to accurately model operational
FM systems. Following Tables list all of the parameters required by the Monte Carlo simulation to model a
12.5 kHz FM system.
Table 7675: Parameters assumed for the 12.5 kHz FM PMR systems
Table 7776: Unwanted emissions for the 12.5 kHz FM PMR systems (measurement bandwidth of 8
kHz)
Table 7877: Receiver blocking values for the 12.5 kHz FM PMR systems
A1.5 CDMA-PAMR
Table 8079 : Receiver blocking values for the CDMA PAMR systems
4 MHz p 82 p 75 30 kHz
6 MHz p 74 p 76 100 kHz
12 MHz p 74 p 85 100 kHz
(1): where p represents the transmission power expressed in dBm.
A1.6 DTT
Reference broadcast transmitter configurations in Table 82Table 81 are provided that are representative of
actual deployments.
For compatibility calculations a directional low power DTT antenna based on a cardioid should be used. The
attenuation (A) of the cardioids horizontal pattern with azimuth angle () is given by
= 2010 (2 2 ) dB
where:
1 + + ()
=
2+
where:
= Azimuth angle
k = 0.4187 for 10 dB pattern minima
Parameter Value
Where no terrain information is available when propagation predictions are being made, the height h1 (m) of
the antenna above ground is calculated according to path length, d (km), as follows:
h1 ha
(m), for d 3 km
h1 ha (heff ha )( d 3) / 12
(m), for 3 km < d < 15 km
where:
ha is the antenna height above ground level;
Heff is the effective antenna height.
The field strength in the vicinity of the broadcast transmitting station is a function of the vertical radiation
pattern of the transmitting antenna. The equation below is an approximation to be used for sharing studies.
Sin
E abs
where
A Sin ;
and
A = the antenna vertical aperture in wavelengths;
= the beam tilt below the horizontal.
To allow for null fill the value of E() should not go below the value shown in Table 83Table 82.
Limit on E()
For the third null and at all angles of beyond the third null the value of E() should not fall below 0.05, see
Figure 66Figure 66.
Out-of-band (OOB) emission limits of DTT transmitters to use in compatibility studies are based on ETSI EN
302 29614.
Out-of-band emissions limits are given as mean power level measured at the antenna port in a 3 kHz
bandwidth for the non-critical (Table 84Table 83 & Figure 67Figure 67) and critical (Table 85Table 84 &
Figure 67Figure 67). It should be noted that these are the same as the GE06 masks but referenced to a 3
kHz not 4 kHz measurement bandwidth and they extend to +/- 20 MHz not +/- 12 MHz.
14 Draft ETSI EN 302 296v1 (2016), Electromagnetic compatibility and Radio spectrum Matters (ERM); Transmitting equipment for the
digital television broadcast service, Terrestrial (DVB-T and DVB-T2); Harmonized EN covering the essential requirements of article
3.2 of the Directive 2014/53/EU
ECC REPORT <No> - Page 148
3,9 -34
4,2 -74
12 -111
20 -111
Figure 67: Non-critical OOB limits for DVB-T/T2 transmitters 8 MHz channels
ECC REPORT <No> - Page 149
3,9 -34
4,2 -84
12 -121
20 -121
Figure 68: Critical OOB limits for DVB-T/T2 transmitters 8 MHz channels
The reference receiving antenna height considered to be representative for fixed reception is 10 m above
ground level.
Standard radiation patterns for receiving antennas for Bands IV and V (see Figure 69Figure 69), are given in
Recommendation ITU-R BT.419.
Figure 69: Discrimination obtained by the use of directional receiving antennas in broadcasting
The antenna gain values (relative to a half-wave dipole) to be used are given in Table 86Table 85.
It is possible to take advantage of polarization discrimination for fixed reception. However, in the case of
orthogonal polarization, the combined discrimination provided by directivity and orthogonality cannot be
calculated by adding together the separate discrimination values. A combined discrimination value of 16 dB
shall be applied for all angles of azimuth in Bands IV and V.
The reference receiving antenna height considered to be representative for portable reception is 1.5 m
above ground level.
Table 88Table 87 contains the mean values for building entry loss and the corresponding standard deviation
at UHF derived from table 14 in Recommendation ITU-R 1812.
10.4 dB 5 dB
Recommendation ITU-R BT.1368-12 gives in its Annex 5, 4.1, information on antennas for portable
reception. For portable reception, an omnidirectional antenna shall be applied. The antenna gain (relative to
a half-wave dipole) is as given in Table 89Table 88.
Band IV 0
Band V 0
For portable indoor and outdoor reception of DTT, a location probability of 95% shall be used.
Polarization discrimination shall not be taken into account in frequency planning for portable reception.
Thermal noise
dBm/Hz -173.98
density (kT0)
Receiver noise
dB 7
figure
The studies should consider two reception modes, one mode for fixed reception and one mode for portable
reception.
21 dB 19 dB
Thermal noise
dBm/Hz 173.98
density (kT0)
Receiver noise
dB 6
figure
The studies should consider two reception modes, one mode for fixed reception and one mode for portable
reception.
ECC REPORT <No> - Page 153
21 dB 19 dB
For more details on the ACS values see ECC Report 240 Annex 1.7.
Radio paging services are listed over the whole frequency range from 450-470 MHz in the EFIS. These
applications are usually narrow band (20 and 25 kHz), wide area, simplex 15, P2P16 or P2MP17 (see
ETSI TR 103 102 and ETS 300 719). They use a 2-FSK modulation and provide a throughput of 512 baud or
1200 baud. The receivers are tuned to a dedicated receive frequency which cannot be changed after put into
operation. An essential feature of the paging receivers is their very low power consumption implying not to
offer a transmission function.
16 point to point
17 point to multipoint
ECC REPORT <No> - Page 154
Frequency/MHz 460,0 460,5 461,0 461,5 462,0 462,5 463,0 463,5 464,0 464,5 465,0 465,5 466,0 466,5 467,0 467,5 468,0 468,5 469,0 469,5 470,0
Assignment 248064 460 - 465,74 MHz
LTE Band 31 462,5 - 467,5 MHz
Guard Band UHF 467 -470 MHz
Assignment 248073 465,96 - 465,98 MHz
Assignment 248078 466,0625 - 466,0875 MHz
Assignment 248085 466,22 - 466,24 MHz Paging
Parameter Value
Minimum required C/(N+I) -4 dB (-5dB+1dB)
Sensitivity 20 dBV/m (-110dBm)
Blocking immunity, PS 56 dB
Intermodulation immunity, PS 42 dB
When the paging receiver is used by a person there is an additional path loss of 10 dB because of the
attenuation caused by the body.
To compensate fading effects, a margin of at least 10 dB (above sensitivity) is needed to get the needed
reliability of 99,5 %.
The additional path loss for indoor use is 10 dB. Indoor use will reflect the usage inside buildings in rooms
with windows at the ground floor and above.
The additional path loss for deep indoor use is 20 dB. Deep indoor use will reflect the usage inside buildings
in rooms without windows in the basement.
In Germany, approximately 500 fixed service (FS) narrowband applications are used e.g. by utility
companies. These applications are not covered by the harmonized standard ETSI EN 302 217-2-2 [24][24]
and might be considered as specific. The technical parameters of the point-to-point-service applications in
the 400 MHz frequency range are the following:
ECC REPORT <No> - Page 156
Two antenna specifications were agreed to be representative examples of the antennas used for fixed
service applications in this frequency range.
PROCOM
France 2 elements
S.A.R.L. 380-430 5 dBi
Yagi
R 70-3/
directional
KATHREIN antenna
K733421 and half power 400-470 13.5 dBd
K733427 beamwidth
19
As the relevant HN EN 302 217-2-2 (04/2014) [24][24] does not contain any equipment specification in this
frequency range, the following generic data were derived from specific systems in use and drawn from the
ECC REPORT <No> - Page 157
German national licensing database. A protection ratio of (N+I)/N = 1 dB equal to an I/N = -5.87 dB is
assumed. The parameter T represents the receiver threshold and is very common in protection
considerations of the FS. It is basically the same as the sensitivity and describes the lowest possible signal
level which can be processed by the receiver.
The following table specifies the both the spectral transmission mask and the receiver selectivity which can
be used as information on the blocking performance of the receiver.
frequency relative
offset (MHz) PSD (dB)
-0.05 -40
-0.013 -40
-0.01 0
0 0
0.01 0
0.013 -40
0.05 -40
Table 103102: Further technical information on the fixed link applications in the 400 MHz frequency
range
The following technical parameters of the Paging applications in the 400 MHz frequency range are provided
additionally to the information available in ETS 300719-1 (see [22][22]) and ETSI TR 103102 (see [23][23]):
ECC REPORT <No> - Page 158
Parameter Value
Frequency band 465.960 MHz-466.240 MHz (in Germany)
Channel spacing 20 kHz, 25 kHz
Polarisation Vertical
Link mode Simplex
e.i.r.p. (max) 0.10 kW (22.15 dBW e.i.r.p. or 20 dBW ERP)
Designation of emission F1D, G1D
Table 105104: Limits for adjacent channel power according to clause 7.3.3, table 3, [22][22] Commented [BP30]: Was 16 in old document
Table 106105: Limits for spurious emission power19 according to clause 7.4.3, table 4, [22][22]
19 Conducted or radiated
20 In the standard it is indeed formulated that the ACPR is smaller or equal 70 dBc. This is probably a misprint since only a power ratio
of the limit or LARGER makes sense.
ECC REPORT <No> - Page 159
Figure 72: Paging transmitter spectra for a single channel 25 kHz emission (see figure 9 in [23][23]),
and for three neighboured 25 kHz channels emission (see figure 10 in [23][23])
Note: In sub clause 7.2.3 of [23][23] two classes of receivers (standard receiver and simple receiver) are
introduced. In case of the simple receiver three channels are aggregated and a operation is considered
than for only the centre channel. The outer channels are not supposed to be operated which results in
relaxed conditions.
ECC REPORT <No> - Page 162
locking systems
Parameters (keyless-go OR
remote keyless entry)
Modulation FSK
Txon <250 ms
Txoff >25 ms
longest transmission period (worst
60 s
case)
manual triggered
duty cycle
Similar to DAA (keyless-go):
The following behaviour is every time
Mitigation active:
After each 50 ms, all used channel will be
checked. That channel with the lowest
occupancy is stored to memory. If keyless-
go is triggered (e.g. a person touch the door
handle), the communication starts
150 kHz 30 dB
225 kHz 40 dB
450 kHz 45 dB
800 kHz 50 dB
1500 kHz 60 dB
2500 kHz 70 dB
5000 kHz 80 dB
10000 kHz 95 dB
at image frequency 40 dB
150 kHz 30 dB
225 kHz 40 dB
450 kHz 45 dB
800 kHz 50 dB
1500 kHz 50 dB
ECC REPORT <No> - Page 164
2500 kHz 55 dB
5000 kHz 65 dB
10000 kHz 75 dB
at image frequency 40 dB
Channel spacing no
< -15 dBm e.i.r.p. (key);
Transmit Power26
<+10dBm e.i.r.p. (car)
600 kHz (remote keyless entry)
Receiver Bandwidth
150 kHz (keyless-go)
-20 dBi (key)
Antenna gain
-5 dBi (car)
no specific direction
Antenna pattern
(for simulations: Omnidirectional)
Receiver Sensitivity -99 dBm
Modulation FSK
Txon <250 ms
Txoff >20 ms
longest transmission period (worst
60 s
case)
Mitigation manual triggered
duty cycle
ECE
f in MHz
ECC REPORT <No> - Page 166
Blocking
90.00
80.00
70.00
60.00
50.00
40.00
30.00
20.00
10.00
0.00
-10.00
0.001 0.01 0.1 1 10 100 1000
MHz
90 dB
80 dB
70 dB
60 dB
50 dB
40 dB
30 dB
20 dB
10 dB
0 dB
-10 dB
420 430 440 450
MHz
25+5.57+5.57+5.57ms = 41.71ms
Txon:
(Mode dependant worst case)
Rx CORAX3
Center frequency: 433.92 MHz
Receiver Bandwidth:
Antenna Gain:
no specific direction
Antenna pattern:
(for simulations: Omnidirectional)
Rx CORAX3
Minimum Signal to Noise ratio: 33 % FER
Modulation: AM OOK
SAW Filter (TDK B3743) to reject sidebands
Mitigation:
Omnidirectional antenna design
Rx MFR
Center frequency: 433.92 MHz
Receiver Bandwidth: 360 kHz
Antenna Gain:
no specific direction
Antenna pattern:
(for simulations: Omnidirectional)
TDK B3743
NDK WFC68K0433CJ
ATTENUATION(DB)MIN
60
50
40
30
20
10
0
405 410 415 420 425 430 435
A tool was constructed to examine the interference effects of a network of LTE cells on the mobile receivers
in a single narrowband PMR (PPDR or commercial) cell. The tool operates as described in the following
sections.
A service area is defined by a 200x200 pixel grid, providing 40 000 possible simulation locations. Within this
area, the coverage area for a narrowband cell is defined to provide 95 % area coverage for UHF handheld
portable coverage. At each point defined by a pixel within the grid within the narrowband cell, the received
signal strength was simulated using the following parameters, derived from TETRA receiver specifications:
Characteristics Value
This gave a range for the narrowband cell of 4.2 km, and using a 6 km pixel grid, 15361 locations for
measurement.
The propagation model is the SEAMCAT suburban extended HATA model, as specified in ECC Report 252,
Annex 17 [26][26], which includes variable propagation loss with range, in conjunction with the range
dependent standard deviation specified in the same Annex. An omnidirectional transmitter antenna pattern
was assumed.
A typical simulation result without interference provides a following coverage result as illustrated in Figure
99Figure 99.
ECC REPORT <No> - Page 174
Within the service area, a network of LTE cells was overlaid at defined positions on the pixel grid. Different
positions and different cell radius and cell centre spacings were simulated, using the cell geometry defined in
ITU-R M.2290-0 [27][27], as shown in Figure 82Figure 82 below.
Cell spacing
3R/2
Cell radius R
A cell radius of 2 km, corresponding to a cell centre spacing of 3 km, and a cell radius of 3km corresponding
to cell centre spacing of 4.5 km were simulated. The overlay positions chosen were the limiting positions,
where in one case one LTE cell site was co-located with the narrowband cell site, and in the other case the
closest LTE cell sites were distributed equidistantly around the narrowband cell site see Figure 83Figure 83
below. The cell radii were chosen using recommended suburban deployment parameters from ITU-
R M.2290-0, which recommends typical 2 km cell radius below 1 GHz, and then increasing by 50 %
recognising the probable lower density of cells in a PPDR or commercial deployment in the 400 MHz band.
Figure 83Figure 83 below illustrates the two overlay patterns.
ECC REPORT <No> - Page 175
Each simulation then derived combined signal strength from the LTE transmitters at each pixel location, also
using the SEAMCAT propagation model and range dependent path loss and standard deviations. The
antenna pattern was derived from a production 450 MHz panel antenna, using the relative gain figures at the
relevant azimuth and elevation offset for each pixel.
Each pixel point within the defined range of the narrowband cell was then checked for coverage in the
presence of interference caused by:
Out-of-band emissions from the LTE transmitters;
Third and fifth order intermodulation in the narrowband receiver;
Blocking in the narrowband receiver.
The out-of-band emissions are taken from 3GPP TS 36.104 [13][13] 'operating band emissions' figures,
which apply within 10 MHz of the transmitter frequency.
Third order intermodulation products were derived by calculating an interference spectrum (in dBm/Hz) at a
reference level, and calculating the actual interference at the simulated received signal level against that at
the reference level.
ECC REPORT <No> - Page 176
Blocking effects made use of the narrowband receiver's blocking specification at the relevant offset from the
wanted received frequency.
Sets of simulations were then run where the wanted narrowband received frequency was set at offsets from
the edge of the LTE transmitter wanted bandwidth of 250 kHz, 500 kHz, 1 MHz and 2 MHz, allowing for a
duplexer on the LTE transmitter increasing attenuation of out-of-band emissions by 0 dB (no duplexer), 25
dB and 30 dB. Ten simulations were made of each scenario to combine and average the results.
NOTE: It may be impractical to obtain as much as 30 dB attenuation at frequencies less than 500 kHz from
the edge of the LTE transmitter wanted bandwidth.
Each simulation generated a new estimate of both narrowband and LTE signal strengths at each pixel, and
multiple simulations (up to ten) were run at each frequency offset and with both cell patterns to obtain mean
results and also the spread of results between simulation runs. Thus as single simulation estimated the
coverage or outage for 15 000 points, ten simulations gave 150 000 result points.
The degradation within a radius of an LTE transmitter site is of particular importance where the narrowband
system is providing safety critical service, i.e. PPDR, or to a critical national infrastructure (transport, utilities
etc).
In Figure 84Figure 84, the smaller circles represent a 500 m radius around the interfering LTE transmitters.
Following the simulations of the network patterns, a further study was undertaken to look at the outage within
the 500 m radius of an LTE transmitter located anywhere within the narrowband system's coverage area. To
do this, the simulation measured the outage within the 500 m radius of a single LTE transmitter site with the
transmitter placed at intervals of 0.2 km along a radius from the centre of the narrowband transmitter
coverage area (i.e. from its transmitter site) to the edge of the area. Then a 0.2x0.2 km pixel grid was
overlaid and each pixel assigned to the closest 0.2 km radius. Thus the number of locations for each
interference result at each radius could be used to generate an overall average of the level of interference
degradation with a random placement of an LTE transmitter within a narrowband transmitter cell. Ten
simulations were run at each radius, using the same models as the previous simulations, and assessing the
effect of intermodulation, out-of-band emissions and blocking. A single frequency offset and duplexer
attenuation was chosen as a result of the first set of simulations.
In this network layout, the LTE cells have 2 km radius, leading to a site spacing of 3km. A typical result for
the simulations is illustrated in Figure 85Figure 85 below.
The table below shows the outage within the narrowband cell due to the combined interference from the LTE
transmitters.
ECC REPORT <No> - Page 178
Table 121120: Simulation results with 3 km spacing, centred, over narrowband transmitter coverage
area
The 'total outage' column illustrates the outage over the coverage area of the narrowband transmitter from
both signal to noise and interference. The 'Increase in outage compared to S/N only' indicates the increase in
outage compared to the non-interference case, as the narrowband coverage without interference was 95.21
%, giving an outage of 4.79 %.
The table below however shows the same figures for outage within 500 m radius of an LTE transmitter.
Table 122121: Simulation results with 3 km spacing, centred, within 500 m of LTE transmitter
'Increase in outage compared with S/N' compares the outage with interference with that without interference;
in this case due to the placement of the LTE cells within the coverage area of the narrowband transmitter,
the outage without interference averaged across the 500 m radius circles of interest was 3.84 %.
In this scenario, the positions of the LTE cells were offset, such that the three closest transmitters were an
equidistant distance from the narrowband site. A typical simulation result is shown in Figure 86Figure 86
below.
ECC REPORT <No> - Page 179
Figure 86: LTE network, 3 km site spacing, offset from narrowband transmitter
The table below shows the outage within the narrowband cell due to the combined interference from the LTE
transmitters in this network configuration. The outage in non-interference conditions was 4.79 %; the
increase above this over the total coverage area of the narrowband transmitter is shown in the right hand
columns.
Table 123122: Simulation results with 3 km spacing, offset, over narrowband transmitter coverage
area
The table below shows the outage within 500 m of an LTE transmitter in this network configuration. The
outage in non-interference conditions was 4.38 % averaged over the 500 m radius circles around the LTE
transmitters.
ECC REPORT <No> - Page 180
Table 124123: Simulation results with 3 km spacing, offset, within 500 m of LTE transmitter
In this scenario, the LTE network is centred on the narrowband transmitter, with site spacing increased to 4.5
km (3 km cell radius). A typical simulation result is shown in Figure 87Figure 87 below.
Figure 87: LTE network, 4.5 km site spacing, centred on narrowband transmitter
The table below shows the outage within the narrowband cell due to the combined interference from the LTE
transmitters in this network configuration. The outage in non-interference conditions was 4.79 %; the
increase above this over the total coverage area of the narrowband transmitter is shown in the right hand
columns.
ECC REPORT <No> - Page 181
Table 125124: Simulation results with 4.5 km spacing, centred, over narrowband transmitter
coverage area
The table below shows the outage within 500 m of an LTE transmitter in this network configuration. The
outage in non-interference conditions was 4.38 % averaged over the 500 m radius circles around the LTE
transmitters.
Table 126125: Simulation results with 4.5 km spacing, centred, within 500 m of LTE transmitter
In this scenario, the positions of the LTE cells was offset such that the three closest transmitters were an
equidistant distance from the narrowband site, with the larger 3 km cell radius providing 4.5 km spacing
between LTE transmitters. A typical simulation result is shown in Figure 88Figure 88 below.
ECC REPORT <No> - Page 182
Figure 88: LTE network, 4.5 km site spacing, offset from narrowband transmitter
The table below shows the outage within the narrowband cell due to the combined interference from the LTE
transmitters in this network configuration. The outage in non-interference conditions was 4.79 %; the
increase above this over the total coverage area of the narrowband transmitter is shown in the right hand
columns.
Table 127126: Simulation results with 4.5 km spacing, centred, within 500 m of LTE transmitter
The table below shows the outage within 500 m of an LTE transmitter in this network configuration. The
outage in non-interference conditions averaged 2.6 % over the 500 m radius circles around the LTE
transmitters.
ECC REPORT <No> - Page 183
Table 128127: Simulation results with 4.5 km spacing, offset, within 500 m of LTE transmitter
From the results, it is apparent that the density and placement of the LTE transmitter network is critical to the
interference results for the narrowband network. If an acceptable degradation in coverage for a narrowband
network is 1%, the higher density LTE network exceeds this in either of the two network layouts.
The LTE transmitter requires a duplexer to reduce out-of-band emissions in the narrowband mobile receive
band at all separations. Of the two values used in simulations, the increase in out-of-band emission
attenuation from 25 dB to 30 dB only has marginal effect; this is because the intermodulation performance in
the receiver is already the dominant effect when 25 dB OOBE attenuation is provided. During the
simulations, some experiments were carried out with higher attenuations (40 and 50 dB), but the results
showed no improvement.
The degradation within 500 m of the LTE transmitter site is only under 10% in the case that the LTE
transmitter is co-sited with the narrowband transmitter in the 4.5 km LTE site spacing simulation. Because of
this, a further simulation was carried out to look at placement of the LTE transmitter within the narrowband
transmitter coverage area.
As described in section 5.1, interference simulations from an LTE transmitter were carried out at 0.2 km
radius intervals from the narrowband transmitter, and the results mapped to a 0.2 km x 0.2 km pixel grid
within the circular 4.2 km radius of the narrowband coverage area. An illustration of the mapping over one
quarter of the area is shown in Figure 89Figure 89 below.
ECC REPORT <No> - Page 184
The simulations were carried out at 500 kHz offset between the edge of the LTE wanted transmitter
bandwidth and the narrowband mobile receiver frequency, using a 30 dB duplexer attenuation.
Table 129128: Simulation results for outage within 500 m of LTE transmitter at varying distance from
narrowband transmitter site
Using the number of locations at each radius as a weighting factor, the average increase in interference
within 500 m of an LTE transmitter at any point in a cell is 12.08 %.
The results show that the placement of LTE cells with respect to the narrowband transmitter coverage area is
important when considering the results. They also show that within a 500 m radius of an LTE site, significant
degradation of the narrowband coverage will result (from approximately 4 % to 14 %) and the acceptability of
this to the narrowband service will have to be taken into account.
The LTE transmitter will require a duplexer to reduce out-of-band emissions to the level that intermodulation
in the victim receiver becomes dominant. This appears to be at about 25 dB attenuation.
Editor's note: further investigation of the cross over point would be useful.
Further attenuation does not add significant improvement because of the intermodulation effect. The results
are relatively little affected by increased frequency spacing from the edge of the LTE transmitter (because of
the bandwidth and therefore the spread of frequencies over which intermodulation is caused). However the
practical limit for attenuation curves in duplexers will limit the minimum spacing between LTE transmitter and
victim receiver.
ECC REPORT <No> - Page 186
A3.1 INTRODUCTION
The ECC 240 report [1] presents methods for analyzing the impact of introducing LTE technology within the
400 MHz band. The interaction of these new systems with legacy systems such as TETRA already operating
in this band has to be evaluated.
Monte Carlo coexistence simulations are used to investigate the mutual interference impact between LTE450
and TETRA-PPDR. The simulations are based on snapshots where users are randomly placed in a
predefined deployment scenario. It is a kind of static simulation which doesnt take into account the small
scale fading, however it considers the large scale fading caused by shadowing of obstacles modelled by
adding a log-normal distribution to the median path loss, see Tables 1 and 2 . By Monte Carlo approach the
aggregate interference generated by the LTE networks can be simulated.
In the LTE network, the transmit power of user equipment (UE) has to be adapted to channel conditions and
traffic needs. Thus analyzing an LTE uplink scenario (UL) comprises modelling admission control, power
control, and scheduling algorithms.
In downlink (DL) case, the Base Stations (BS) of both LTE and TETRA transmit without any power control.
Hence, a TETRA BS transmits with a constant power. In contrast, the total transmit power of an LTE BS is a
function of the number of resource blocks (RBs) simultaneously transmitted. This number is a measure of
utilization of channel frequency resources depending on traffic burstiness and is captured as a percentage
by Channel Utilization Factor (CUF). For a 3 MHZ bandwidth, up to 15 RBs can be allocated and therefore,
the transmit power can vary in a range of 11.76 dB.
The methodology introduced in this contribution for modeling interference between radio systems is applied
in order to achieve more accurate results for interference scenarios where aggregation of interference from
multiple users needs to be taken into account, and where positions and transmit power of interfering
equipment needs to be reflected in detail. It also provides means for realistically reflecting the interaction
between radio links in the interfering and/or interfered systems, e.g. through power control and scheduling.
LTE is OFDMA based in DL and SC-FDMA based in UL, whereas TETRA is TDMA based. The frequency
reuse factor in LTE is 1, and in TETRA generally in the order of 20. This leads to a quite different cell layout
and modelling for both systems.
A proposed frequency band allocation for LTE is shown in Figure 90Figure 90. The legacy TETRA system
uses 2 MHz bandwidth. A 3 MHz band could be assigned to the LTE system. The intention of the
coexistence analysis is to verify whether the complete 3 MHz band could be occupied by a LTE system
without the need for a guard band between the LTE and TETRA bands.
UL UL DL DL
A macro network layout of a TETRA system and an LTE system is depicted in Figure 91Figure 91. Base
stations are denoted as triangles, and the randomly deployed TETRA MSs or LTE UEs respectively are
indicated as squares.
In a TETRA network deployment, a set of frequencies is assigned to each of these cells. Neighbouring cells
are using different sets of frequencies for avoiding co-channel interference. In distant cells the frequencies
can be reused. We assume in the present interference analyses a sufficiently high reuse-factor so that co-
channel interference can be neglected.
The LTE network layout shows a 1-tier cell layout with 21 sectors. This reflects a compromise for an
acceptable simulation time and a realistic interference situation.
For the simulation of packet-switched LTE systems, there is need for a traffic model for the services
supported by the system and a scheduler algorithm for the allocation of RBs to different users. The selection
of the traffic model and the scheduler in the aggressor network has a crucial impact on modelling the amount
of interference caused to the victim network. Therefore, realistic assumptions on traffic model and scheduler
in the LTE network are the key for a fair assessment of coexistence between LTE and other services.
An essential characteristic of the LTE450 system is the type of applications operated in this frequency band
such as maintenance and operation of smart meters. Hereby a very high number of devices are periodically
or irregularly (partly once per month) transmit messages with low amount of data. Accordingly, a bursty
traffic load will be created.
Modelling such bursty traffic is achieved by applying appropriate scheduling algorithms. There exist different
scheduling algorithms. To prevent the problem of comparing arbitrary schemes, we introduce the channel
utilization factor, which is defined as the fraction of total RBs used. Due to the low amount of transmit data
per UE, it is assumed that only one RB is allocated to each active UE. The CUF determines the number of
scheduled UEs and is constant during an interference analysis.
The propagation model used for the interference analyses is the extended Hata model, as specified in the
SEAMCAT handbook (ECC Report 252 [2]). It supports the calculation of median path loss for urban,
suburban, and rural environments (see Table 130Table 129). Shadowing (variation in path loss) is achieved
by applying a log-normal distributed pathloss with variance according to Table 131: Description of variance in
path loss depending on the distance (SEAMCAT Handbook)
When the calculated pathloss is lower than the free space attenuation for the same distance, the free space
attenuation is taken instead. This free space attenuation is given by Eq.1 in Table 130Table 129
Table 130129: Description of the median path loss L depending on the distance for the 400 MHz band
(SEAMCAT Handbook)
( )2
= 32.4 + 2010() + 1010 [ 2 + ]
106
d < 0.04
(Eq. 1)
d > 0.04 urban 150 < f 1500 44.9 6.5510 max 30,H b 10 d
a Hm b Hb
(Eq. 2)
L L(urban)
= () 40.94
2
4.78{10[{max(150; )}; 2000]}
open area +18.33 10[{max(150; )}; 2000]
(Eq. 4)
[10 () + 10 (0.04)]
= (0.04) + [(0.1) (0.04)]
[10 (0.01) 10 (0.04)]
0.04 < d < 0.1
(Eq. 5)
ECC REPORT <No> - Page 189
With:
b Hb max 0, 2010 ( ))
30
Hm min(h1, h2);
Hb max(h1, h2)
h1 : tx antenna height
h2 : rx antenna height
1 20
= {
1 + (0.14 + 1.87 104 + 1.07 104 ) (10 )0.8 20 < < 100
20
Table 131130: Description of variance in path loss depending on the distance (SEAMCAT
Handbook)
d 0.04 = 3.5
12 3.5
0.04 < d 0.1 above roof = 3.5 + ( 0.04)
0.1 0.04
17 3.5
below roof = 3.5 + ( 0.04)
0.1 0.04
below roof = 17
9 12
0.2 < d 0.6 above roof = 12 + ( 0.2)
0.6 0.2
9 17
below roof = 17 + ( 0.2)
0.6 0.2
0.6 < d =9
In the case of indoor-outdoor conditions (LTE-UE indoor) the median loss and the shadowing have to be
increased with additional wall loss (
Wall loss 11 dB
A3.2.4 Downlink simulation methodology for LTE BS interferer and TETRA MS victim
In DL scenario, both LTE BS and TETRA BS transmit with constant power. Power control is not applied.
1. TETRA MS victim:
Distribute sufficiently many TETRA MSs uniformly across the centre cell of the TETRA system. For each
snapshot, one MS is randomly chosen to be served by the BS. Please note that because neither co-channel
interference nor adjacent channel interference from own system (TETRA) need to be considered, a simplified
ECC REPORT <No> - Page 191
TETRA system with only one centre cell is modelled. This simplified scenario matches well to the SEAMCAT
layout.
Allocate at each snapshots randomly one of the 80 TETRA frequency channels (80 channels with 25 kHz
bandwidth in the 2 MHz band).
The TETRA BS transmit power is constant: PBS,max
Calculate the desired received signal strength (dRSS) for the selected TETRA MS:
dRSS = CMS = PBS,max pathloss(MS,BS).
Tetra MSs which dont get the required S/N ratio (i.e. 19 dB, see Table 5) according the formula Eq.10 are
excluded from the calculation of interference probability
2. LTE BS interferer:
Create a 1-tier (7 BSs) or 2-tier (19 BSs) cell layout, each cell consisting of 3 sectors. There is no need to
consider the LTE UEs. In the DL case the interference is created only by the LTE BSs.
Create for each snapshot a random cell-shift between the LTE interferer network (i.e. the positions of all
BSs) and the TETRA MS, which receives the interference.
Note:
The cell-shift is a 2-dimensional vector and produces an equally distributed position. It is not a 1-
dimensional distance. The latter would create a higher probability of short distances compared to the first
method.
Calculate the pathloss from each (shifted) LTE BS to the TETRA MS. Include the directional antenna
gain, depending on the angle of line of sight between each sector antenna and the TETRA MS.
3. Calculate the Interference Probability as defined in ECC Report 252 [34][34] (SEAMCAT handbook),
Annex 2:
= 1 (
>
| > )
where (see Error! Reference source not found.Error! Reference source not found.)
A3.2.5 Uplink simulation methodology for LTE UE interferer and TETRA BS victim
ECC REPORT <No> - Page 192
1. TETRA victim:
Distribute sufficiently many TETRA MSs uniformly across the centre cell of the TETRA system. For each
snapshot one MS is randomly chosen to be served by the BS. Please note that because neither co-channel
interference nor adjacent channel interference from own system (TETRA) need to be considered, a simplified
TETRA system with only one centre cell is modelled. This simplified scenario matches well the SEAMCAT
layout.
Allocate at each snapshots randomly one of the 80 TETRA frequency channels (80 channels with 25 kHz
bandwidth in the 2 MHz band).
Perform UL power control, in the case power control is applicable (see Error! Reference source not
found.Error! Reference source not found.):
PMS,min PMS PMS,max
Calculate the desired received signal strength (dRSS) for the selected TETRA BS:
dRSS = CBS = PMS pathloss(MS,BS) .
Tetra BSs which dont get the required S/N ratio (i.e. 19 dB, see Table 5) according the formula Eq.18 are
excluded from the calculation of interference probability.
2. LTE interferer:
Create a 1-tier (7 BSs) or 2-tier (19 BSs) cell layout, each cell consists of 3 sectors. Distribute sufficiently
many LTE UEs randomly throughout the system area such that to each sector within the HO margin of 3 dB
the same number of users is allocated as active UEs.
Calculate the pathloss from each UE to all sectors and find the smallest pathloss including the directional
antenna gain (depending on the angle of line of sight between each sector antenna and the LTE UE)
Link the UEs randomly to a sector to which the pathloss is within the smallest pathloss plus the HO
margin of 3 dB.
In the 3 MHz band up to 15 RBs can be scheduled during one snapshot. Select K=15 UEs randomly
from all the UEs linked to one sector. This is the maximum number of UEs which could be scheduled
during this snapshot.
Set CUF in the range [1:1:15]/15. This number determines the portion of 15 UEs selected in the previous
step, which will be scheduled as active UEs and will be kept constant during an interference analysis. A
CUF = 30 % indicates that 5 UEs are scheduled. Consequently the number of scheduled UEs per sector
is
Error! Objects cannot be created from editing field codes.
Assign to each scheduled UE the appropriate channel frequency corresponding to the assigned RB
Combine for each UE the LTE UE emission mask (180 kHz bandwidth), placed at the RBs individual
carrier frequency, with the TETRA BS blocking mask at the randomly selected TETRA channel and
calculate the portion of interference attenuation amask(UEk,fTETRA).
Loop over all LTE cells from Error! Objects cannot be created from editing field codes. to Error! Objects
cannot be created from editing field codes.
(the number of cells in the system area e.g. 21 for 57 sites with tri-sector antennas)
Loop over all active and scheduled UEs from Error! Objects cannot be created from editing field codes. to
Error! Objects cannot be created from editing field codes.:
For the Error! Objects cannot be created from editing field codes.-th active UE in the Error! Objects cannot
be created from editing field codes.-th cell (i.e.Error! Objects cannot be created from editing field codes.) its
interference power is denoted by Error! Objects cannot be created from editing field codes.:
Create for each snapshot a random cell-shift between the LTE interferer network layout and the victim
TETRA BS, which gets the interference.
Note:
The cell-shift is a 2-dimensional vector and produces an equally distributed position. It is not a 1-
dimensional distance. The latter would create a higher probability of short distances compared to the first
method.
Calculate the pathloss from each (shifted) LTE-UE to the TETRA BS (the one in the center of the TETRA
cell layout). The TETRA BS and the LTE UEs have omni-directional antennas, hence only antenna gain
is contributing additionally to the overall pathloss calculation.
Error! Objects cannot be created from editing field codes. is the interference power from Error! Objects
cannot be created from editing field codes. at the TETRA victim BS
= 1 (
>
| > )
where (see Error! Reference source not found.Error! Reference source not found.)
The interferer LTE BS transmits with constant power. Power control is not applied. The TETRA MS can be
operated with power control enabled or disabled. The reference interference analysis, which has been
presented in the ECC Report 240 (Table 4 in [1]), has been performed by SEAMCAT simulations without
TETRA power control.
1. TETRA BS victim:
Distribute sufficiently many TETRA MSs uniformly across the centre cell of the TETRA system. For each
snapshot, one MS is randomly chosen to be served by the BS. Please note that because neither co-channel
interference nor adjacent channel interference from own system (TETRA) need to be considered, a simplified
TETRA system with only one centre cell is modelled. This simplified scenario matches well to the SEAMCAT
layout.
ECC REPORT <No> - Page 194
Allocate at each snapshots randomly one of the 80 TETRA frequency channels (80 channels with 25kHz
bandwidth in the 2MHz band).
Calculate the desired received signal strength (dRSS) for the selected TETRA BS:
Tetra MSs which dont get the required S/N ratio (i.e. 19 dB, seeError! Reference source not found.Table
5) according the formula Eq.22 are excluded from the calculation of interference probability
2. LTE BS interferer:
Create a 1-tier (7 BSs) or 2-tier (19 BSs) cell layout, each cell consisting of 3 sectors. There is no need to
consider the LTE UEs, hence the interference is created only by the LTE BSs.
Create for each snapshot a random cell-shift between the LTE interferer network (i.e. the positions of all
BSs) and the TETRA BS, which receives the interference.
Remark:
The cell-shift is a 2-dimensional vector and produces an equally distributed position. We shift the interferer
BS randomly against the victim BS in an area defined by the interferer cell range. The SEAMCAT workspace
for evaluating the BS-BS interference, which has been presented in ECC Report 240 Table 4 [1], uses a cell
shift of the LTE BS against the TETRA MS. Thus the effective cell shift is higher (sum of both cell ranges of
LTE and TETRA), and consequently the interference lower.
Calculate the pathloss from each (shifted) LTE BS to the TETRA BS. Include the directional antenna gain,
depending on the angle of line of sight between each sector antenna and the TETRA BS.
In the 3 MHz band up to 15 RBs can be scheduled for each LTE BS during one snapshot. Set a CUF
between [1/15:1/15:1]. In average, over some 105 or more snapshots, the CUF corresponds to an equally
reduction of BS transmit power.
Combine the LTE BS emission mask (3 MHz bandwidth) with the TETRA BS blocking mask at the randomly
selected TETRA channel and calculate the portion of interference attenuation amask(fTETRA). For defining
the interference attenuation factor see Annex Error! Reference source not found.9.1.
Calculate the received interference signal (interference signal strength iRSS) for the selected TETRA BS b y
summing up the interference from each LTE-sector:
Nsec tor
iRSS I cross P
k 1
LTE _ BS ,max PL( BS TETRA , BS kshift
, LTE ) amask ( f TETRA ) CUF
(Eq. 921)
3. Calculate the Interference Probability as defined in the SEAMCAT handbook Error! Reference source
not found.[2] Annex 2:
= 1 ( > | > ) (Eq. 1022)
where (see Table 131: Description of variance in path loss depending on the distance (SEAMCAT
Handbook)
Table 2)
The abovementioned simulation methodology has been implemented in a MATLAB based simulation tool,
which provides the full control over all algorithms, and parameter settings. Particularly, incorporating a bursty
traffic model has been facilitated. In the SEAMCAT simulator up to version 5.1.0 [2] such traffic model has
not been available. It provides only a full buffer traffic model.
The simulations use a parameter setting which has been defined basically in the ECC Report 240 [1]. LTE
specific parameters have been changed according to the proposal in [3]. A basic set of these parameters is
included in the tables below.
Feeder loss 0 dB 2 dB
5 dB steps to a minimum of
Power Control Characteristic Not used
15 dBm
Downlink interference from LTE BS to TETRA MSs is the more relevant case since TETRA handhelds
should not require an expensive technology for preventing adjacent channel interference.
For a basic overview of the capabilities of a TETRA network the coverage for urban environment has been
evaluated depending on the cell radius. At 5.2 km cell radius (highlighted below) a TETRA cell has 94 %
coverage (i.e. received signal strength > -103 dBm receiver sensitivity) in an urban environment.
The interference probability presented in the following indicates how many of these MSs which would get
coverage are blocked in the case of interference.
TETRA Coverage
8.5 79.368%
7.5 84.292%
6.2 90.227%
5.2 94.111%
ECC REPORT <No> - Page 197
TETRA Coverage
4.93 95.00 %
2.77 99.473%
1.96 99.907%
In ECC Report 240 [1] the need for adding additional duplex filter has been brought up. The results for 10 dB
and 25 dB additional attenuation to the adjacent band are presented in Error! Reference source not
found.Error! Reference source not found.. However, the resulting interference based on the updated
parameter setting according to [3] is much smaller compared to the results published in [1]. This is also partly
due to the SEAMCAT-based implementation of cell shift used in [1] (cf. chapter A3.2.4A3.2.4). As could be
recognized in the last row of Error! Reference source not found.Error! Reference source not found., the
highest probability of interference caused by LTE DL to TETRA MSs is in the order of 2 % corresponding to
the smallest LTE Cell Radius (1.6 km) considered.
The impact of CUF on the interference performance is depicted in Figure 92Figure 92Error! Reference
source not found.Error! Reference source not found. for cell radius 5.2 km. The value for CUF=1
corresponds to the result for the full buffer case in Error! Reference source not found.Error! Reference
source not found. and Error! Reference source not found.Error! Reference source not found.. Also, results
for interference to the most adjacent two TETRA channels are included in Figure 92Figure 92. In this case,
the interference probability increases from 0.5397 % to 0.7729 %. Figure 93Figure 93 shows the
dependency of interference probability on each TETRA channel.
Table 137136: DL Impact of LTE BS on TETRA MS; Full buffer (15 UE), Urban
Figure 93: Interference probability of LTEin400 BS on TETRA MS: Impact of each TETRA channel.
Full buffer. Urban environment
Results of uplink interference evaluation from LTE UEs to TETRA BS are collected in Figure 94Figure 94
and depicted in Error! Reference source not found.Error! Reference source not found.. The impact of the
channel utilization factor together with the environment is shown. The results show that the probability of
interference caused by LTE UL to the TETRA BS is negligible.
The total maximum transmit power Pmax = 23 dBm (see Error! Reference source not found.Error!
Reference source not found.) is shared by 15 RBs, consequently a LTE-UE using only one RB can transmit
a maximum power (32-11.76) dBm = 11.24 dBm.
(TETRA: Cell radius 5.2 km, LTE Cell radius 3.75 km, see [1], [3])
LTE Urban Urban Suburban
Channel utilization outdoor 25% indoor outdoor
factor [%] 75% outdoor
100 0.4362 % 0.3874 % 0.3854 %
The scenario with the highest impact on the TETRA system is the case of interference from LTE-BS to the
TETRA-BS. The reason for this phenomenon is the much lower path loss between the BSs as explained in
the following.
The term a(Hm) in Eq. 6 models the contribution of the antenna with the lower height to the median path
loss, where this height is put in relation to the clutter value of 10m. For the BS to BS scenario, the TETRA
MS antenna height Hm = 1.5m considered in the UL and DL scenarios has to be replaced by the BS antenna
height of 30m resulting in a decrease of the path loss by 28.5 dB.
The results of interference evaluation from LTE BS to the TETRA BS are presented in Table 139Table 10.
The probability of interference can be considerably reduced by adding 25dB attenuation to the LTE emission
mask, e.g. by applying a duplex filter. In addition, it could be observed from the results that a duplex filtering
beyond this attenuation doesnt result in much considerable reduction of the probability of interference.
Table 13910: DL Impact of LTE BS on TETRA BS; Full buffer (15 UE), Urban,
(TETRA: Cell radius 5.2 km),
Pathloss limited by free space attenuation and MCL=70dB
A3.8 CONCLUSION
This contribution presents the simulation methodology, parameters, assumptions and results for compatibility
and sharing studies in ECC PT SE7 on LTE-based M2M type communications in the bands 410-430MHz
and 450-470 MHz. The results are based on LTE parameters proposed in [3] and therefore a revision of
those submitted before to ECC PT SE7.
The UL results show that the probability of interference caused by LTE UL to the TETRA BS is negligible. In
addition, the UL results demonstrate that the highest probability of interference caused by LTE DL to TETRA
MSs is in the order of 2 % corresponding to the smallest LTE Cell Radius (1.6 km) considered. This appears
to be an acceptable performance degradation for the TETRA PPDR. Although the probability of interference
caused by LTE BS with a duplexer of 10 dB/25 dB attenuation to TETRA MS will be considerably reduced,
but it needs to be weighed up if the trade-off interference probability reduction / cost would be justifiable.
The probability of interference caused by an LTE BS to a TETRA BS is the highest due to the fact that path
loss between the BSs based on Hata model is much lower compared to the UL and DL scenarios. The
results presented demonstrate that the probability of interference can be considerably reduced by an
additional duplex filtering in the order of 25dB at the LTE BS. In addition, it could be observed that a duplex
filtering beyond this attenuation doesnt result in much considerable reduction of the probability of
interference. Such duplex filter would furthermore reduce the probability of interference in DL, i.e. from LTE
BS to TETRA MSs, as pointed out in the previous paragraph.
It is proposed to include in an appropriate manner the simulation methodology, parameters and assumptions
suggested in this contribution into the LTE450 report. This could be beneficial for those ECC PT SE7
members who wish to perform their own simulations with the aim reproducing and/or verifying the results. In
addition, it is proposed to include the results of this study into the report.
The LTE interferer emission mask as well the TETRA victim blocking masks are taken from the SEAMCAT
workspaces, which have been used for the reference interference evaluation presented in ECC Report 240
[1]. The respective workspaces for the reference analyses are for the DL case
The normalization of the LTE emission mask is performed in such a way that in the full 3 MHz transmission
bandwidth the maximum transmit power will be radiated. Thus in a resource block RB only a portion of 1/15
= -11.76 dB is transmitted.
Figure 95. DL
Figure 97. UL
A4.1 PARAMETERS
Parameter settings according to ECC Report 240 [1][1], A1.1 LTEin400 and A1.2 TETRA;
Calibrating Matlab simulation tool with SEAMCAT simulation results published in ECC Report 240,
chapter 3;
Extended HATA propagation model defined in SEAMCAT Handbook (new version ECC Report 252)
[26][26];
Assumption on the maximum transmit power: LTE BS 46 dBm, LTE UE 37 dBm.
TETRA: single cell considered (i.e. infinite reuse factor for TETRA, so that co-channel interference
vanishes);
LTE: 1 tier (19 cells), 3-sectors antenna;
TETRA cell randomly dropped in the LTE network.
where:
Depending on the Channel Utilization Factor, a set of [1,15] active UEs from the N UEs randomly dropped
are scheduled per snapshot.
Statistics:
Mean value: 27.177 dBm;
Median value: 29.297 dBm.
A4.6 CONCLUSION
= 1 > >
with
dRSS: desired received signal strength;
iRSS: unwanted received signal strength;
iRSScomp: composite interfering received signal strength;
sens: receiver sensitivity.
Time Slot
1 1 16 31 46 61 76 91 106 3646
3 3 18 33 48 63 78 93 3648
4 4 19 34 49 64 79 94 3634 3649
7 7 22 37 52 67 82 97 3622 3637
9 9 24 39 54 69 84 99 3654 3652
12 12 27 42 57 72 87 102 3642
13 13 28 0 43 58 73 88 3643 3627
Layout:
LTE-layout: 1 tier, 3-sectors. Power control target: -100 dBm at BS-RX;
TETRA-layout: 9 cells per cluster, omni-directional antenna. No power control (could be considered
included if required).
For SEAMCAT simulations, frequency allocation has been performed as presented in Figure 110Figure 110
below.
3 MHz 2 MHz
Figure 110: Illustrative Frequency Allocation of the 450 - 470 MHz Showing LTEin400 and TETRAPOL
In the simulations, the TETRAPOL Network is considered to be a Single Cell, centre of Infinite Network
with No Co-channel Interference. This scenario which does not consider any co-channel interference within
the NB system, is used for comparison with LTE PPDR purposes.
Table 142Table 140 below gives the interference probability as calculated with SEAMCAT when combining
unwanted emissions and blocking effects. Interference probabilities are given for the LTEin400 BS spectrum
emission mask minimum requirements, as well as with 25 dB additional attenuation to this mask for
protection of the BS receiver against own or different BS. The shaded fields correspond to interferer
densities that are not relevant for the M2M scenario, but are included to compare the results with the
corresponding scenarios presented in ECC Report 240.
Interference
Interference
Interferer Probability with
Probability with BS
Density Cell Radius 25 dB additional
emission mask
(LTEin400 BS / (km) attenuation
minimum
km) compared to BS
requirements [%]
emission mask
For comparison purposes, the corresponding figures of PPDR probability of interference on TETRAPOL are
given in the Table 143Table 141 below (extracted from table 8 in ECC Report 240):
ECC REPORT <No> - Page 215
Table 143141: LTEin400 PPDR BS impact on TETRAPOL BS (from ECC Report 240, for comparison)
Interference
Interference Probability
Probability with BS
Interferer Density Cell Radius with 25 dB additional
emission mask
(LTEin400 BS / km) (km) attenuation compared to
minimum
BS emission mask
requirements
As expected, interference probabilities of an eMTC LTE-system are significantly lower than in the
comparable case of a PPDR-LTE system.
The impact of the LTEin400 BS spectrum emission mask on TETRAPOL BS is nevertheless still high when
considering an isolated TETRAPOL BS. By introducing a 25 dB additional attenuation on the LTEin400 BS
emission mask within the considered TETRAPOL BS reception band, the interference probability is reduced
from up to 47 % to well below 4 % for even the worst conditions.
The three-sector LTEin400 BS transmits at 461.5 MHz whereas the TETRAPOL MS receive signals coming
from TETRAPOL BS between 463 and 465 MHz. The victim frequency is randomly chosen (discrete
distribution option). Table 3.3 below gives the interference probability as calculated with SEAMCAT for the
M2M scenario. For comparison purposes, Table 147Table 145 shows the corresponding results for the
PPDR scenario, as in table 9 of Report 240. The shaded fields correspond to interferer densities that are not
relevant to the M2M scenario, but are included to compare the results with the corresponding scenarios
presented in ECC Report 240.
Table 145143: LTEin400 PPDR BS impact on TETRAPOL MS (from ECC Report 240, for comparison)
ECC REPORT <No> - Page 216
Again, the M2M scenario leads to significant lower interference probabilities, as expected. The LTEin400 BS
minimum requirements are already offering a sufficient level of protection with regards to TETRAPOL MS.
Hence, there is no need for additional attenuation in the LTEin400 BS on TETRAPOL MS case.
0.089 0.067 %
0.179 0.100 %
0.268 0.101 %
0.358 0.129 %
0.447 0.115 %
0.537 0.095 %
The impact of LTEin400 MS on TETRAPOL BS is negligible and absolutely acceptable from an operational
point of view. As expected, the interference levels are lower than the corresponding figures from ECC Report
240 (extracted from table 10), for reference given in Table 147Table 145:
Table 147145: LTEin400 PPDR MS impact on TETRAPOL BS (from ECC Report 240, for comparison)
0.027 0.80 %
0.082 0.13 %
0.137 0.33 %
The LTE MS transmit at 451.5 MHz whereas the TETRAPOL MS receive signals coming from TETRAPOL
BS between 463 and 465 MHz. The victim frequency is randomly chosen (discrete distribution option) in
SEAMCAT
Table 148Table 146 below gives the interference probability as calculated with SEAMCAT when combining
unwanted emissions and blocking effects. The assumed MS emissions are considering 3GPP minimum
requirements.
0.089 0.067 %
0.179 0.100 %
0.268 0.101 %
0.358 0.129 %
0.447 0.115 %
0.537 0.095 %
The impact of LTEin400 MS on TETRAPOL MS is negligible and absolutely acceptable from an operational
point of view. As expected, the interference levels are lower than the corresponding figures from ECC Report
240 (extracted from table 11), for reference given in Table 149Table 147:
Table 149147: LTEin400 PPDR MS impact on TETRAPOL MS (from ECC Report 240, for comparison)
0.027 0.05 %
0.082 0.04 %
0.137 0.16 %
ECC REPORT <No> - Page 218
A5.5 CONCLUSIONS
Simulations for the LTE scenario in the 400 MHz band have been performed using SEAMCAT with eMTC
Parameters. The interference calculated is from an eMTC LTE Network into a TETRAPOL System.
In both the downlink and uplink, the interference percentages calculated for eMTC LTE on TETRAPOL are
lower than the figures shown in ECC Report 240 for PPDR application.
The interference levels for UL for eMTC LTE on TETRAPOL are well below 1% and are acceptable from the
operational point of view with no additional attenuation.
Of the three cases that were studied, only the eMTC LTE BS on TETRAPOL BS case may need a 25 dB
attenuation additional to the BS emission mask minimum requirement.
ECC REPORT <No> - Page 219
3GPP MS spectrum mask (following a ITU-R recommendation), is used for the compatibility studies between
LTE and other systems. The specific levels of the mask are included in ANNEX 1:ANNEX 1:.
A6.1.1 Introduction
In a situation where LTE transmitters are close in frequency to narrowband PMR mobile receivers,
interference from the LTE BS can cause outages in the coverage from a wanted PMR BS.
The overall effect can be estimated across the PMR BS coverage range. However close to the LTE BS,
outage is significantly increased when compared to the average across a coverage area, and whether this is
acceptable may depend on the type of user served by the PMR system.
This simulation provides results for interference within a 500m radius of an LTE BS at various offset
frequencies from the LTE TX, and where the BS is placed at various distances from the PMR BS.
For the simulations, an LTE BS was placed at increasing distance from a PMR BS within the coverage of a
PMR cell. Each simulation moved the LTE BS 200 m further from the PMR BS. Simulations were carried on
at a set of locations 100 m apart in a grid within a 500 m radius of the LTE BS location to assess the effect of
interference within the area close to the LTE BS. The simulations were repeated for PMR frequencies that
were close to the edge of the LTE transmitter bandwidth (500 kHz from the edge of the LTE transmitter
bandwidth), at a midpoint in a 5 MHz PMR band adjacent to the 5 MHz LTE transmitter (2.5 MHz from the
edge of the LTE transmitter bandwidth); and close to the edge of a 5 MHz PMR band furthest from the LTE
transmitter (4.5 MHz from the edge of the LTE transmitter bandwidth).
For each simulation, the PMR signal strength was simulated using the SEAMCAT urban extended Hata
propagation model with range dependent standard deviation, as specified in Draft ECC Report 252, Annex
17, and the signal/noise ratio calculated. The LTE signal strength at the PMR receiver was also estimated
using the same model; and then the effects of out-of-band emissions (attenuated by a duplexer filter at the
LTE BS) and intermodulation performance calculated, to provide a signal/(interference + noise) calculation
for the simulation. A point that failed the signal/(interference + noise) ratio requirement was counted as an
outage only if the signal/noise ratio in the absence of interference otherwise met the criteria for the receiver.
Thus within the 500 m radius of the LTE BS, the outages due to interference could be counted, and
presented as a percentage of all locations within that 500m radius. Ten simulations were conducted for each
measurement point within the stated 500 m radius of the LTE cell.
LTE B/W:
5 MHz
The out-of-band emissions for the LTE BS are taken from 3GPP TS 36.104, Table 6.6.3.2.1-3 operating
band emissions within 5 MHz of the edge of the carrier for a Category B wide area BS <1 GHz. Out-of-band
emissions were then attenuated by 25 dB to allow the effect of a duplexer on the LTE BS in improving the
performance of the LTE BS. Other results have shown that with more than approximately 20 dB of duplexer
attenuation of the out-of-band emissions the interference to the PMR MS does not improve, because the
intermodulation performance of the PMR receiver becomes the dominant effect.
The PMR system parameters were as shown in Table 151Table 149 below:
EiRP (dBm): 49
The receiver specifications were those from TETRA, taken from ETSI EN 300392-2. Channel Performance
Criteria equates to the required signal/noise or signal/interference ratio specifications from that standard.
The BS specifications are those from SE7(17)011.
ECC REPORT <No> - Page 221
16.00%
14.00%
12.00%
10.00%
+0.5MHz
8.00%
+2.5MHz
+4.5MHz
6.00%
4.00%
2.00%
0.00%
0 0.20.40.60.8 1 1.21.41.6 2 2.22.42.62.8 3 3.23.43.63.8 4 4.24.44.64.8 5
Figure 111: Simulation results for interference to PMR MS within 500m of LTE BS, with 25dB duplexer
attenuation, at varying ranges from PMR BS
Figure 111Figure 111 shows the percentage outage due to interference (after outage locations due to poor
signal to noise ratio of the PMR system were discounted) where the LTE BS is placed at ranges up to 5 km
from the PMR BS, at the stated offset frequencies from the edge of the LTE transmitter bandwidth. The
curves show the percentage of locations within the 500 m radius of the LTE BS that an outage due to
interference occurs.
For reference, Figure 112Figure 112shows the signal strength received at the PMR MS from the PMR BS.
ECC REPORT <No> - Page 222
-50
-55
-60
-65
-70
-75
-80
-85
-90
-95
-100
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8 5
Figure 112: Signal strength received at PMR MS from PMR BS at various ranges from PMR BS
From Figure 111Figure 111 and Figure 112Figure 112, it can be seen that the interference outages close to
an interfering LTE BS suffered by a victim PMR MS are dependent on how far the LTE BS is from the PMR
BS, and so how weak the wanted signal level is at the PMR MS. The effect is somewhat reduced when the
PMR frequency is at increased frequency separation from the LTE BS frequency.
The interference is the least when the LTE BS is close to or co-sited with the PMR receiver, and worst when
the LTE BS is close to the PMR cell edge. The interference within 500m of the LTE site varied from
approximately 2.5 % of locations to approximately 13 % of locations depending on position and on PMR
frequency.
A6.2 INTERMODULATION
The design applied for this study is a real system developed for a 900 MHz client. We shifted frequency and
changed antennas for the UHF range and then re-run the Hydra model to get the new map configuration as
illustrated in this paper.
There is no specific LTE site constellation; the idea is to look at the PMR signal strength at various locations
and model the outage around the LTE site due to the IM generated in the PMR receiver by presence of the
LTE signal.
Thus, we are not looking at the composite outage but rather what the holes in coverage surface look like.
We have been studying broadband interference this way for some time now since it is our experience that
professional license holders mostly do not accept the argument that their overall coverage is not affected
very much by such interference. They take the view that a half-kilometre outage right in middle of the
coverage area is highly unacceptable regardless of the overall percentage.
ECC REPORT <No> - Page 223
We observe that the major differences between the UHF and 900 MHz scenarios are found within the
antenna patterns/gain/path loss, with UHF having much less vertical gain so excess interference at close
distances should be expected.
As we mentioned earlier, Motorola Solutions will be happy to run new calculations including the scatter -plots
based on different parameter assumptions for subsequent meetings/discussions. It will be possible to further
study how receiver specifications affect the IM performance, as well as the signal strength of the wanted
signal, the channel performance criterion and RF AGC, (if the receiver design include this) based on LTE site
antenna height, TxP and spectrum options.
Parameter Value
A6.2.2 Results
Parameter Value
A6.2.3 Discussion
Motorola Solutions have created two models to investigate and further study the interference impact from
LTE emissions into PMR operating in adjacent spectrum.
The first model estimates the spectrum due to inter-modulation products that results inside a PMR receiver
due to the presence of strong LTE signals such as will occur near LTE sites. It calculates the coefficients of a
non-linear model based on the ETSI EN 300 392-2 V3.8.1 (2016-08) [16][16] and ETSI EN 300 113 V2.1.0
(2015-12) [25][25] for TETRA and DMR radios, respectively.
Firstly, the receiver sensitivity, co-channel rejection and inter-modulation attenuation specifications are
applied to determine the signal strength of two narrowband signals that results in a third-order inter-
modulation product at the receive frequency with the same signal strength as the thermal noi se floor. The
same way as was discussed at the most recent SE 7 meeting for 900 MHz. The model coefficients are now
derived to produce this condition. Once the coefficients are determined, the model can be used with any
signal, which includes an LTE component.
Next, the model then generates an LTE signal of a user-input signal strength/frequency and generates the
resulting spectrum shape using standard spectrum estimation techniques.
The resulting spectra are shown in Option 0 (Slide 1). This is run for 3 MHz LTE at 461.5 MHz and PMR
receiver at 463.5 MHz. For completeness, see also Figure 1-12. Figures 1-6 are for a receiver that just meets Commented [BP31]: starts at figure 84 in the new
the TETRA standards of ETSI EN 300 392-2 V3.8.1 (2016-08) and Figure 7-12 are for a DMR receiver that document... you refer to figure 1-12, but figure 3-4 are
just meets the standards of ETSI EN 300 113 V2.1.0 (2015-12). out of the old document. Hence this update is for a later
stage.
-90
TETRA
-100
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-40 dBm
-100
-45 dBm
Power Spectral Density (dBm/Hz)
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-40 dBm
-100
-45 dBm
Power Spectral Density (dBm/Hz)
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
-110
-120
-130
-140
-150
-160
-170
-180
460 461 462 463 464 465 466 467 468 469 470
Frequency (MHz)
The figures above illustrate the power spectral density due to receiver-generated IM products. The LTE
transmitter will of course also contribute with sideband noise and transmitter IM, but we assume here that the
transmitter filters out those components and we examine only the effects of the LTE signals on the PMR
receivers. The cyan-coloured line is the approximate thermal noise floor of the receiver based on the
sensitivity and co-channel rejection specifications.
The amount that the inter-modulation spectrum exceeds the thermal noise floor is the approximate receiver
sensitivity degradation that results from the inter-modulation products generated by the LTE signals of the
given signal strength.
A second set of algorithms is used to explore the impact of inter-modulation interference that LTE may have
on PMR coverage. In this case, the HATA path loss model is applied with a model of the base station as
given in ECC Report 240 Table 14 [1][1], including the horizontal and vertical patterns of the base station
antenna, to find the strength of the LTE signal at a large number of locations around the LTE site.
The HATA model includes range-dependent and random components. The random component is a
Gaussian random variable.
A standard deviation of 5.5 dB has been applied for our simulation. Meanwhile, the average signal strength
of the PMR signal is entered from the dashboard and we assume that the separation between the PMR and
LTE sites is sufficient that the average signal strength is substantially constant across the test area. An
independent Gaussian random loss component is drawn for each location and applied to the PMR signal.
ECC REPORT <No> - Page 231
The dashboard furthermore includes inputs for the spectrum arrangement option and PMR frequency, as
well as the parameters needed to determine the inter-modulation power spectral density at the PMR
frequency. The inter-modulation power for the received LTE power level at each location is calculated from a
prototype spectrum for the selected spectrum arrangement option. The difference in the received LTE power
and the sensitivity, inter-modulation rejection, co-channel rejection specification values entered in the
dashboard from those used to develop the prototype spectrums are used in a system of equations to derive
the inter-modulation power level. This power is added to the thermal noise floor and subtracted from the
received LMR power to obtain a signal-to-noise ratio (SNR). If the SNR exceeds the SNR value entered in
the dashboard, the location is coloured green; if not, the location is coloured red. The model offers different
ways to calculate the outage.
The model was used to make outage maps at four PMR average signal strength levels: -62 dBm, -74.5 dBm,
-83 dBm and -90 dBm +/-1.5 dB. Maps were made for PMR receivers meeting TETRA and DMR
specifications for a total of eight maps. The maps are overlaid on an actual UHF system design that shows
approximate contours for the four signal strengths listed above. This is to give the reader an idea of how
much area each of the signal strengths occupy in the coverage area. Note that the LTE site causes outage
even in locations close the LMR base sites. Imagine how many LTE sites it would take to cover the area
shown here. The large number of LTE sites will result in a large number of holes in coverage across the
LMR coverage area, which will greatly reduce the dependability of the PMR systems and result in complaints
and mitigation costs for the LTE operator.
A6.3.2 Conclusion
This study illustrates the severe difficulties encountered, when the obvious incompatibilities between LTE
and narrowband technologies, analogue as well as digital, are surfacing. The study is looking entirely on
PMR as a victim, as the protection of LTE from narrow band carriers can be handled by many advanced
mitigation techniques such as MIMO selectivity arrangements even though this is less effective at such low
frequencies as 400 MHz.
ECC REPORT <No> - Page 234
A7.1 INTRODUCTION
In a situation where LTE transmitters are close in frequency to narrowband receivers, there are three forms
of interference which affect the performance of the narrowband PMR system. These are:
Out-of-band emissions from the LTE BS TX
Blocking in the PMR receiver
Intermodulation performance of the PMR receiver
The first of these can be reduced by a filter on the LTE BS TX, which can also be the same filter that is used
for duplex operation of the LTE BS TX and RX. The second and third are due to non-linearity in the MS
receiver, and so cannot be affected by improvements at the BS.
For the simulations, an LTE BS was placed at three spot locations within the coverage of a PMR cell.
Simulations were carried on at a set of locations 200 m apart within a 500 m radius of the LTE BS location to
assess the effect of interference within the area close to the LTE BS. The spot locations for the LTE BS
were chosen at 1km distance from the PMR BS, for strong signal effects, 2.6 km from the PMR BS (half the
PMR cell radius of 5.2 km) and 4.6 km from the PMR BS (so that the 500 m radius circle around the LTE BS
was fully inside the PMR BS coverage area).
For each simulation, the PMR signal strength was simulated using the SEAMCAT urban extended Hata
propagation model with range dependent standard deviation, as specified in Draft ECC Report 252, Annex
17, and the signal/noise ratio calculated. The LTE signal strength at the PMR receiver was also estimated
using the same model; and then the effects of out-of-band emissions and intermodulation performance
calculated, to provide a signal/(interference + noise) calculation for the simulation. A point that failed the
signal/(interference + noise) ratio requirement was counted as an outage only if the signal/noise ratio in the
absence of interference met the criteria for the receiver. Thus within the 500 m radius of the LTE BS, the
additional outages due to interference could be counted, and presented as a percentage of all locations
within that 500 m radius. Five simulations were conducted for each measurement point within the stated 500
m radius of the LTE cell.
The PMR frequency was selected to be 2.5 MHz from the edge of the LTE transmission.
The out-of-band emissions for the LTE BS are taken from 3GPP TS 36.104, Table 6.6.3.2.1-3 operating
band emissions within 5 MHz of the edge of the carrier for a Category B wide area BS <1 GHz.
The PMR system parameters were as shown in Table 155Table 153 below:
EiRP (dBm): 49
The receiver specifications were those from TETRA, taken from ETSI EN 300392-2. Channel Performance
Criteria equates to the required signal/noise or signal/interference ratio specifications from that standard.
The BS specifications are those from SE7(17)011.
To show the effects of duplexer attenuation, the out-of-band emissions were reduced in 5 dB steps from 0
dB to 35 dB and the sets of simulations repeated. The sets of simulations were repeated at different offset
frequencies from the edge of the wanted bandwidth of the LTE transmitter, specifically at 500 kHz, 2.5 MHz
and 4.5 MHz offset to look at results at edges and middle of an adjacent band containing narrowband
receivers.
60.00%
50.00%
40.00%
4.6km
30.00%
2.6km
1km
20.00%
10.00%
0.00%
0dB 5dB 10dB 15dB 20dB 25dB 30dB 35dB
Figure 129: Simulation results at different duplexer attenuations of OBE from LTE BS at 500 kHz
offset from edge of LTE transmitter bandwidth
50.00%
45.00%
40.00%
35.00%
30.00%
4.6km
25.00%
2.6km
20.00% 1km
15.00%
10.00%
5.00%
0.00%
0dB 5dB 10dB 15dB 20dB 25dB 30dB 35dB
Figure 130: Simulation results at different duplexer attenuations of OBE from LTE BS at 2.5 MHz from
edge of LTE transmitter bandwidth
ECC REPORT <No> - Page 237
45.00%
40.00%
35.00%
30.00%
25.00% 4.6km
2.6km
20.00%
1km
15.00%
10.00%
5.00%
0.00%
0dB 5dB 10dB 15dB 20dB 25dB 30dB 35dB
Figure 131: Simulation results at different duplexer attenuations of OBE from LTE BS at 4.5 MHz from
edge of LTE transmitter bandwidth
Figure 129Figure 129 to Figure 131Figure 131 show the variation in interference within a 500 m radius of an
LTE BS with duplexer attenuation of OBE with the LTE BE placed at 1 km, 2.6 km and 4.6 km from the PMR
BS, using TETRA as the BS and victim MS technology.
It is apparent that increasing duplexer attenuation up to 20 dB reduces the level of interference as the out-of-
band emissions from the LTE BS are attenuated. However beyond 20 dB, there is no further reduction,
which implies that intermodulation within the victim PMR RX becomes the dominant effect.
A7.5 CONCLUSION
Where interference is incurred to a victim PMR receiver in proximity of an LTE BS, this interference can be
mitigated by filtering at the LTE BS, such as from the duplexer filter, until approximately 20 dB attenuation at
the wanted frequency is reached. Beyond this, little further improvement can be seen as RX intermodulation
effects become dominant.
ECC REPORT <No> - Page 238
A8.1 INTRODUCTION
This paper describes the coverage and interference models used in contribution SE7(17)003 in more detail,
and provides some additional results using the same model when system parameters are set equal to those
used in ECC Report 240.
The coverage model uses a 200 x 200 pixel grid offering 40 000 coverage locations as described in
SE7(17)003. The LMR site is located at the centre of this grid. LTE interferers are added at point locations
within the grid also as described in SE7(17)003. Results are taken by summation of the coverage and
outage pixels in each simulation.
For each point within the grid, the RF signal strength from both the LMR site and the LTE sites is simulated
in order to calculate the signal/(noise+interference) ratio and to determine whether that pixel in that run of the
simulation is in coverage or has an outage. The defined range of the LMR site is used to draw a circular
boundary, and measurements outside this boundary are excluded from the results. Summation of the pixels
with an outage within the area of the circle of the LMR coverage compared with the total number of pixels
enclosed within the circle provides the percentage outage within the site coverage. The same mechanism is
used for smaller circles of 500 m radius around the LTE sites, to look at outage close to the LTE sites.
To obtain the RF coverage from both LTE and LMR sites, the Extended Hata propagation model defined in
SEAMCAT, as described in draft ECC Report 252, Annex 17 is used. This means according to Table 82 of
that document:
Table 156154: Propagation equations from SEAMCAT Commented [JC32]: This model is not exclusively
defined in SEAMCAT. Therefore, a referece to the
relevant specification of the propagation model would
d<0.04km (eq.182): be preferred
0.04km<d<0.1km (eq.189):
In SE7(17)003, a suburban propagation model was used; further results in this document have been
obtained using an urban propagation model according to that used in ECC Report 240.
For each pixel point in each simulation, a log-normal fading distribution is applied according to that specified
in Annex 17. Above roof propagation is assumed. Specifically:
ECC REPORT <No> - Page 239
Thus each point in each simulation run will have the actual calculated received signal strength of both
wanted LMR signal and interfering LTE signal varied according to a normal distribution with standard
deviation according to Annex 17.
For the LMR site, an omnidirectional coverage is assumed using the antenna gains outlined with the results.
As the gains are relatively low, no allowance is made for any variation in gain due to elevation, and so the
wanted received LMR signal will not be reduced close to the site due to elevation.
For the LTE site, the antenna pattern from a 'real' panel antenna, namely the Sinclair SP304C-
SF2P65LDF(D00), is used, with the forward gain scaled by the desired gain of the simulation. Both
azimuthal and elevation gain are considered, such that the level of the interfering signal is reduced when
close to the LTE site due to the vertical beamwidth of the antenna. For reference, the antenna patterns of
this antenna is shown compared with the Kathrein 742 242, the pattern from which is used in ECC Report
240.
ECC REPORT <No> - Page 240
Sinclair SP304C-
Kathrein 742 242
Kathrein 742 242 SF2P65LDF(D00)
The subscriber is assumed to have an omnidirectional antenna with the gain specified for the simulation.
At each pixel point, the calculated received signal strength of the wanted LMR signal and unwanted LTE
signal are calculated.
To calculate the level of the out-of-band emissions from the LTE signal, the emissions given in 3GPP 36.104
are used. There are two possible specifications which could apply:
a) ACLR, as given in Table 6.6.2.1-1, which is 45 dB measured in the wanted transmitter bandwidth (5
MHz as used for these simulations)
The operating band emissions are taken from Table 6.6.3.2.1-3 for a Category B wide area BS <1 GHz.
Category B has lower emissions that Category A. The specific specifications from this table are:
Frequency offset of
Frequency offset of
measurement filter Minimum requirement (Note Measurement
measurement
centre frequency, 1, 2) bandwidth (Note 8)
filter -3dB point, f
f_offset
5 MHz f < 5.05 MHz f_offset < -14 dBm 100 kHz
ECC REPORT <No> - Page 241
Frequency offset of
Frequency offset of
measurement filter Minimum requirement (Note Measurement
measurement
centre frequency, 1, 2) bandwidth (Note 8)
filter -3dB point, f
f_offset
Then at the offset frequency of the wanted LMR receiver, the results are scaled down according to the
receiver wanted bandwidth (e.g. by 18/100 for OBE in an 18 kHz TETRA receiver) to produce the signal
power in each band.
The resulting OBE emissions level is added to thermal noise within the receiver bandwidth, and the
signal/noise ratio of the resulting simulated LMR signal strength to OBE+thermal noise measured, and an
outage is noted for each point that the result fails to satisfy the criterion for the receiver concerned (e.g. 19
dB for TETRA).
A duplexer on the LTE transmitter will attenuate the OBE as part of its role in protecting the LTE receiver. To
allow for this effect and the reduction of OBE and therefore increased protection of the LMR receiver, an
attenuation figure has been added. This simply reduces the level of the OBE input to the
signal/(interference+noise) calculation without any regard for whether a practical duplexer curve is likely to
achieve the required level of attenuation.
The PSD produced by the intermodulation model is used with a set of transforms to determine the
intermodulation power produced in the receiver for the received LTE power and compares this to the signal
strength of the desired signal to determine if coverage exists at a particular point.
(1)
The complex conjugations implement the model as a baseband model, meaning that the nonlinear
operation does not produce third and fifth harmonics of the carrier frequencies but all the intermodulation
components that form around the carrier frequencies are represented accurately.
The key to the model is determining the values for A0, a3 and a5 in such a way to represent actual receiver
performance. This has been done based on standard receiver specifications, which can be deconstructed to
ECC REPORT <No> - Page 242
determine the physical values behind the specifications. The results than then be scaled to the
specifications of any receiver.
Intermodulation rejection specification (IMR3) is usually measured using two carriers of equal strength with
frequency offsets of + and +2, where is the desired frequency. In this way, the third-order
intermodulation product lands on the desired frequency . We begin by expanding (1) using these
frequencies and grouping by frequency:
(2)
(3)
(4)
where V0 represents the voltage of the input signals and V3 and V5 are the third-order and fifth-order
intermodulation products, respectively. We now have three equations and we can determine the values of
V0, V3 and V5 from published specifications, allowing us to solve for the unknown values of A0, a3 and a5.
The necessary specifications are the Reference Sensitivity (SensRef), Intermodulation Rejection (IMR3) and
Co-channel Rejection (CCR).
SensRef is the signal strength at which a channel performance criterion (CPC) is met. For analog receivers
the CPC is typically 12 dB SINAD; for digital receivers the CPC is typically 5 % bit error rate. IMR3 and CCR
are defined with respect to SensRef.
In the IMR3 test the strength of a desired signal is set to 3 dB stronger than the SensRef and the two-carrier
unwanted signal adjusted until the CPC is obtained again. The unwanted signal strength is typically 70 to 80
dB greater than SensRef at this point and the IMR3 value is equal to the unwanted signal strength minus
SensRef. With the unwanted signal strength at this level, the intermodulation power in the receiver bandwidth
is equal to the thermal noise floor in the receiver. Therefore, the unwanted signal strength is equal to
SensRef+IMR3.
CCR is determined by degrading the performance of a desired signal 3 dB stronger than SensRef back to
the CPC using an on-channel unwanted signal. Therefore when the unwanted power equals SensRef+IMR3,
the power third-order intermodulation product is SensRef-CCR.
Therefore, the voltage values for (2), (3) and (4) are:
Now (5), (6) and (7) can be substituted into (2), (3) and (4), respectively, and solved simultaneously for the
model parameters. As a practical matter, we have found that the equations must be solved sequentially as
follows or the finite precision of most computers will result in errors:
(8)
(9)
(10)
(11)
For example, for a receiver with Reference Sensitivity of -120 dBm, Co-channel rejection ratio of 8 dB,
Intermodulation Rejection (IMR3) of 70 dB and IMR5 of 80 dB, a3 = 20.8076, a5 = 8022.79, and A0 =
0.99953.
The nonlinear model is used in a simulation program that incorporates a spectrum estimation routine. An
LTE symbol is generated by populating a buffer with 72, 180, 300 or 600 16-QAM symbols depending of
whether 1.4, 3, 5 or 10 MHz LTE is selected. The QAM symbols are arranged in the buffer such that the
middle resource element is blank, according to LTE specifications. The voltage of the QAM symbols is
scaled so that the average transmitter power is the same for all bandwidths.
The Modified Partial Transmit Sequence 28 method of crest factor reduction (CFR) is applied to the LTE
signal. This method does not result in the best possible CFR nor is it used in practice. The advantage of it,
however, is that it does not affect the average power or spectrum at all. We felt that this is a worthwhile
trade-off since we are interested specifically in the spectrum impact of intermodulation and not base station
implementation issues. Figure 133Figure 133-Figure 134Figure 134 shows graphs of peak-above-average
probability (1a) and an example of the power spectral density after operating on the LTE signal with the
nonlinear model (1b) with (w/) and without (w/o) CFR.
28 Neil Braithwaite, Crest Factor Reduction for Down-link LTE by Transmitting Phase Shifted Resource Blocks, Powerwave
Technologies
ECC REPORT <No> - Page 244
Figure 133: Detail of CDF of instantaneous power Figure 134: Example of LTE received spectrum
with respect to average power for a with and without CFR
The symbols are scaled again based on a nominated received power level. The samples are then passed
through the nonlinear model, windowed with a Hann window and transformed into a periodogram. Typically
200 periodograms are averaged. The average periodogram is scaled to units of mW/Hz and converted to
dBm for display. Figure 135Figure 135 shows the spectrum that results inside the receiver described in the
example above when a 5 MHz LTE signal is received at -30 dBm.
Figure 135: Example LTE spectrums as they exist in an LMR receiver based on -120 dBm sensitivity,
8 dB co-channel rejection ratio, 70 dB intermodulation rejection and 80 dB fifth-order intermodulation
rejection
ECC REPORT <No> - Page 245
The spectrum estimation model generates a spectrum given a specific configuration of received power (RxP)
and receiver IMR, sensitivity and CRR specifications. It is necessary that the receiver IMR, sensitivity and
CNR be configurable in the coverage model. Of course, it is also necessary that the IM spectrum is a
function of the received LTE power as determined by the propagation model. So a set of transformations
have been developed so that an IM spectrum developed for a specific IMR, sensitivity, CRR and received
power (RxP) can be transformed to that which would result from any given IMR, sensitivity, CRR and RxP.
This is done by separating (1) into three components representing the linearly scaled signal and the signal
operated on by the third-order and fifth-order nonlinearities, respectively, and estimating the spectrum of
each as shown here:
(12)
(13)
(14)
Figure 136: 3 MHz LTE signal (blue) with third-order (red) and fifth-order (green) components broken
out so that they can be scaled separately
The reason for separating the spectrum into these components is that the power of the IM0 component will
increase (or decrease) in power by 1 dB for every 1 dB increase (or decrease) in RxP with respect to the
ECC REPORT <No> - Page 246
power level used to generate the spectrum, which we call RxPREF. In addition, the power of IM3 will change
by 3 dB for every 1 dB change in RxP with respect to RxPREF and IM5 will change by 5 dB for every 1 dB
change in received power with respect to RxPREF. The IM spectrum is obtained by summing the RMS
voltages of the three spectra, IM0, IM3 and IM5, as required by (1) as shown in (12).
(15)
(16)
(17)
(18)
Once the spectrum has been calculated and scaled to the level of the simulated received signal and the
specifications of the victim receiver, it is scaled according to the receiver bandwidth (e.g. 18 kHz for TETRA)
to calculated to unwanted signal level within the receiver bandwidth. This is then added to the thermal noise
within the bandwidth, and the received signal to noise ratio calculated. A signal outage is noted if the
signal/(interference+noise) ratio is less than that demanded by the receiver (e.g. 19dB for TETRA).
A blocking model has also been applied. In this case, the far from carrier blocking performance is assumed
(e.g. -25 dBm at >500 kHz for TETRA). A scaling factor was applied to allow for cases where the edge of
the LTE signal was closer to the LMR frequency than this and so a fraction of the LTE signal bandwidth
impinged on the reduced blocking protection of the wanted receiver: in practice this reduces blocking
performance by 2 dB when the LTE signal is 50 kHz from the wanted frequency and 0.2 dB when 250 kHz
from the wanted frequency, and so can be ignored.
An outage was noted if the wanted LTE signal level exceeded the blocking threshold of the receiver, as there
are no known degradation curves for signal levels above this threshold. In practice, the locations where
blocking occurred were always also outage locations due to IMD and OBE, and so blocking could have been
ignored.
Assumption differences
SE7(17)003 used the same suburban model and handheld subscriber device as the previous submissions
SE7(16)073r1, as this is a very representative condition. However ECC Report 240 uses the following
parameters which differ from those in that contribution:
LTE transmitter power +47 dBm vs +43 dBm used in SE7(17)003
LTE site range of 3.2 km (1.6 km radius) and higher vs 2 km and 3 km used in SE7(17)003
Urban propagation environment vs suburban used in SE7(17)003
TETRA transmitter power +40 dBm plus 9dBi antenna gain, giving +49 dBm EiRP, vs +46 dBm EiRP
used in SE7(17)003
TETRA subscriber antenna gain as 0 dBi vs -6 dBi in SE7(17)003
Declared TETRA cell range as 5.2 km, vs 4.2 km calculated for 95 % coverage in SE7(17)003
ECC REPORT <No> - Page 247
The other difference of course is that the model in SE7(17)003 uses deterministic site placing rather than
random site placings of the SEAMCAT simulations.
Therefore, a further series of results have been run to directly compare with the SEAMCAT simulations in
Report 240 using exactly the same parameters.
A8.8 RESULTS
The model was re-run using the Report 240 parameters together with a 3.2 km site range (1.6 km radius)
and using LTE sites spaced equidistant from the LMR site, the following results have been obtained for
interference based outage due to the combined effects of OBE and IMD. The outages from OBE vs IMD
have been plotted separately with 0 dB and 25 dB duplexer attenuation to show the relative levels of the
results under each condition.
7.00%
6.00%
5.00%
1.00%
0.00%
465.5 466.5 467.5 468.5 469.5
Figure 137: OBE and IMD interference to TETRA MS with equidistant LTE site spacing from TETRA
site, 3.2km range with 0 dB and 25 dB duplexer attenuation
The average interference with this site layout across the TETRA coverage area is 4.1 % (compared with 10.4
% in Report 240) with no attenuation, and 0.45 % with 25 dB attenuation of the OBE from the LTE
transmitter.
As can be seen on the curves, with no duplexer attenuation OBE dominates, and with 25 dB duplexer
attenuation, the IMD results dominate.
ECC REPORT <No> - Page 248
The difference between these results and those in Report 240 will be due to the static placing of the LTE site
pattern in this model compared with the random placing of sites offered by SEAMCAT.
Results, for this LTE site pattern, are also shown below for the interference (combined OBE and IMD) within
500 m of an LTE site.
50.00%
45.00%
40.00%
35.00%
30.00%
25.00% 0dB
25dB
20.00%
15.00%
10.00%
5.00%
0.00%
465.5 466.5 467.5 468.5 469.5
Figure 138: Interference outage to TETRA within 500 m of LTE site with 0 dB and
25 dB duplexer attenuation
The average with no duplexer attenuation across all sampled frequencies is 39.0 %, and with 25 dB duplexer
attenuation the average is 16.2 %.
ECC REPORT <No> - Page 249
A9.1 CONTEXT
The study analyses the adjacent band compatibility between LTE PMAR/PAMR systems operating in the
450-470 MHz with DTT receivers operating at DTT channel 21 (470-478 MHz). Both BS and UE may create
interference into the DTT receiver system, therefore the two cases of interference from BS and UE into DTT
are evaluated. Although the study of the impact of DTT on LTE BS and UE has already been done and the
results presented in ECC Report 240 remain applicable, this new study considers a new set of parameters
different from those of LTE BB-PPDR in ECC report 240. In particular, in order to be aligned with 3GPP
channel 31, a guard band of 2.5MHz between the LTE BS channel upper limit and DTT channel 21 lower
limit is investigated.
A Monte Carlo analysis is performed to assess the probability of interference faced by the victim DTT
receiver. The Monte Carlo analysis for the UE case enables to compare the interference with the MCL
analysis where the UE is considered to be located at the position that maximizes the interference. Indeed,
this situation, although possible, is not representative of real life operation of the device which may be
located anywhere in the vicinity of the DTT Rx. Therefore, if only the conclusions of the MCL analysis are
considered, they may lead to undue constraints for the operation of the UE.
The DTT transmitter is placed at the centre of the coverage area as depicted in Figure 139.
DTT coverage radius for DTT high power transmitters (85.15 dB)
The LTE base station (BS) is placed at the centre of the cell. Each LTE cell is composed of three sectors as
depicted in Figure 140.
BS
This LTE cell is repeated to build up a perfectly homogeneous single frequency LTE cluster composed of 7
cells (BS) as depicted in Figure 141. A cluster of size 7 is composed of 21 (7 x 3) hexagonal-shaped sectors.
BS
PPDR cluster
The coexistence scenario for the analysis of potential interference from LTE BS or UE to DTT receiver uplink
is illustrated in Figure 142.
ECC REPORT <No> - Page 251
The DTT and LTE network parameters are summarised in Annex 1. The LTE network cluster is placed at a
separation distance D of the DTT receiver (D is measured between the DTT receiver and LTE cluster centre
reference cell site).
6 The DTT receiver is randomly positioned, following a uniform polar distribution, in the DTT cell or in a
pixel of 100 m x 100 m at the edge of the DTT cell as depicted in Figure 143. The azimuth orientation of
the TV receiver antenna is directed toward the DTT transmitter in case of fixed rooftop reception.
Around the DTT receiver within a radius of Dmax, a LTE cluster is randomly positioned following a
uniform angular distribution. The position of the cluster is defined by the position of the central cells
BS as depicted in
7 Figure 144.
8 The active LTE user equipment (UE) is randomly positioned, following a uniform distribution, within each
cell of the LTE cluster.
9 The probability of interference is calculated based on a high number of events generated for each
simulation.
ECC REPORT <No> - Page 252
DTT Tx
Position of the LTE cluster around the victim DTT receiver (a single Monte Carlo event):
TxDTT
rDTT
The considered DTT Rx protection criterion is C/(N+I) = 21 dB. For a good statistical accuracy, a number of
100 000 events is considered for each run of the simulation.
The DTT Rx at the DTT coverage edge, uniformly located in an area of 100m x 100m (one pixel). This
corresponds to the most sensitive case for DTT reception scenario.
The position of the interfering link (IL) is set to be dynamic relative to the victim DTT Rx so that the LTE
cluster reference cell is located in the vicinity of the DTT receiver, uniformly positioned within an area limited
by a circle of radius Dmax = 7.5 km whose center is the DTT Rx.
The reported received power (dRSS, iRSS) correspond to the median values. They are reported in order
to evaluate which contribution impacts more the DTT Rx. The calculated probability of interference takes into
account mutually the interference from the LTE unwanted emissions, the DTT Rx blocking and the
overloading.
The impact of the LTE BS transmitted power is first evaluated and the result is presented in Table 2.
DTT radius = 40.46 km, LTE Cell Radius = 750 m (ISD = 2.25 km), Dmax = 7.5km, ACLR =45 dB
LTE Bandwidth
1.4 3 5
(MHz)
LTE BS center
466.8 466 465
frequency (MHz)
BS Tx Power
37.5 29 41 37 28 43 37 30
(dBm)
dRSS (dBm) -68.27 -68.3 -68.35 -68.31 -68.31 -68.33 -68.33 -68.3
iRSS unwanted
-103.93 -117.01 -103.83 -107.73 -116.77 -103.9 -110 -116.89
(dBm)
iRSS blocking
-124.2 -132.71 -122.18 -126.08 -135.11 -121.77 -127.86 -134.76
(dBm)
ECC REPORT <No> - Page 254
iRSS unwanted
20.27 15.70 18.35 18.35 18.34 17.87 17.86 17.87
iRSS blocking (dB)
Interference
18.97 4.73 19.18 13.33 4.92 18.98 10.5 04.84
probability (%)
The results presented are conservative: the LTE BS cell radius of 750 m corresponds to a very high density
of deployment (urban environment), thus being one the most pessimistic scenario. A wider cell radius may
lead to a more favourable result for the compatibility given the lower density of LTE BSs within the LTE
network around the DTT Receiver. However, the analysis in rural environment where the coverage radius is
wider for the two systems (resulting to a lower received signal level of the DTT Rx) is not covered by the
previous analysis. Moreover, the worst case where the DTT Rx is always located inside the LTE cluster is
not covered while it could lead to more negative results.
Interference is due mainly to LTE BS unwanted emissions: indeed, the difference between iRSS unwanted
varies from 15.70 dB to about 20.3 dB. Given the ration between those two, it can be concluded that iRSS
unwanted is the parameter that contributes the most to the interference. This is confirmed by the fact that
ACIR is very close to iRSS unwanted.
The maximum LTE BS transmitted power creates too much interference: when maximum power is
considered, the level of interference received by the DTT Rx is very high so that the probability of
interference approximates 20 %, what is too much compared to the maximum acceptable value 5 %. The
reduction of the BS Tx power impacts directly the OOBE of the BS which reduced accordingly. For the three
channel bandwidths, a reduction of the BS Tx power so that the iRSS unwanted is reduced by almost 13 dB
(compared to its value for maximum BS Tx power) leads the percentage of interference to a value below the
requirement of 5 %.
Additional reduction of LTE BS unwanted emissions is required: based on the two previous points, it appears
that the LTE BS Tx unwanted emissions are the main contributors to the interference on the DTT Rx. In
order to avoid any reduction of power of the LTE BS and use it at its maximum power which is beneficial for
coverage and QoS purposes, it appear essential to reduce unwanted emissions by other means than Tx
power reduction. A solution resides in using an external filter that would provide the required reduction of the
BS Tx OOBE. Assuming that any reduction of the BS Tx power leads to a similar reduction of the OOBE, it
can be extracted that the required reduction of power is 13 dB (Max Tx power min Tx power) for the 3 MHz
and 5 MHz channel bandwidths scenarios, while the reduction is of the range of 9 dB for 1.4 MHz bandwidth.
An additional simulation with the reduced OOBE of at least 15 dB within the DTT Rx bandwidth is therefore
require in order to evaluate the accuracy of this analysis.
In order to clarify the opened issues mentioned in the previous session, supplementary analysis is required.
Therefore, an additional investigation is performed based only on the 5 MHz bandwidth and with the
following targets:
Assessment of the impact of the network density. In this case the cell radius is changed and the maximum
distance Dmax between the BS of the reference cell and the DTT Rx is set to the value of the distance
between two BS of the cluster (ISD = 3 x cell radius). This means that the DTT Rx is always located inside a
cell of one of the LTE BS composing the cluster.
Assessment of the impact of the environment. For this case, the rural environment is considered with the
typical LTE cell radius of 8 km as provided in Report ITU-R M.2292-0 for rural scenario. The evaluation is
ECC REPORT <No> - Page 255
performed at a pixel of 100 m x 100 m located at the cell edge (worst case) with the DTT Rx always located
inside the LTE cluster as in the previous bullet.
Assessment of the required additional filtering to mitigate the interference by a reduction of the level of
unwanted emissions. The considered filter attenuation in based on the previous analysis is expected to be
15 dB (2 dB margin above the calculated 13 dB). However, changing the cell radius and Dmax will probably
impact the requirements of the additional filter reduction. if one of the two initial bullets presents a higher
value of required attenuation for OOBE, the corresponding supplementary attenuation will serve as reference
for the filter attenuation, and the impact of the filter will be assessed only for the worst scenario, assuming
that it would provide much better improvement for the other cases.
DTT radius = 40.46 km, LTE Cell Radius = 750 m (ISD = 2.25 km), Dmax = 7.5km, ACLR =45
BS Tx Power (dBm) 43 43 43 43 43 46 46
iRSS unwanted (dBm) -85.29 -115.22 -103.87 -118.88 -114.63 -119.14 -134.12
iRSS blocking (dBm) -103.15 -103.09 -121.74 -121.74 -132.5 -137 -136.99
Interference probability (%) 69.37 15.06 13.02 2.67 3.55 8.91 01.99
ECC REPORT <No> - Page 257
A9.7 CONCLUSION
The analyses allow assessing the impact of the LTE BS OOBE on DTT reception. The legacy OOBE limits
specified by the 3GPP do not ensure sufficient protection of DTT reception in channel 21. Additional filtering is
required to protect the DTT receiver and the required attenuation depends mainly on the proximity between the
LTE base station and the DTT receiver.
The analysis illustrates that the wider the LTE cell radius, the lower the interference into the DTT receiver. Given
the level or attenuation required for the unwanted emissions of the base station when the probability of interference
is exceeded, it can be concluded that filters can be added to LTE BS with minimised cost when needed. This can
be managed on a case by case basis at national level.
For this scenario, the LTE BS is collocated with the DTT Receiver, so that the interfering UE is uniformly distributed
around the DTT Rx and within an area that is determined by the LTE cell radius. The Monte Carlo analysis is
performed to assess how much the DTT receiver is statistically impacted by the variations of the operation of the
UE, which implies mutually:
The UE position around the DTT Rx
The UE effective power through the power control mechanism of the system
The number of simultaneously transmitting UE
The impact of the propagation channel variations
In order for the UE to be very close to the receiving DTT Rx, a really small LTE cell radius of 250 m is considered,
which is not realistic compared to possible real deployment of LTE PMR/PAMR networks which may probably take
advantage of the good properties of this band to have wire areas covered by a single BS.
The drawback of this approach is that the UE is closed to the base station, and therefore it is more likely that it
operates at reduced transmitted power due to power control mechanism. Because of that, a second case is
considered where the UE power is limited to the range [20, 23] dB to force the transmitted power to be around its
maximum value.
DTT radius = 40.46 km, Dmax = 0, UE antenna gain = -7 dBi (body loss 4dB + -3dBi)
LTE BS center
457.5
frequency (MHz)
1 UE 5 UE 1 UE 5 UE 5UE
UE Tx Power range
-40 to 23 -40 to 23 20 to 23 20 to 23 20 to 23
(dBm)
Interference
0.02 0.18 1.98 10.06 00.54
probability (%)
A9.8.2 Conclusion
The results of the study reveal that the received interference from LTE UE unwanted emissions as well as the DTT
Rx blocking response is very low, in particular when the UE Tx power range is regular (-40 to 23 dBm). This results
in negligible probability of interference from LTE UE.
When the UE operates at maximum power, one single UE does not interfere with the DTT reception above 470
MHz. When 5 UE are considered to be transmitting at maximum power, the probability of interference is exceeded.
These results apply when the UE protection of own downlink is considered but considering a cell radius of 250 m.
It should be noted that such situation is pretty not realistic, but if the LTE network cell radius is set to 1km, the
interference is mitigate, what makes it possible to conclude that the LTE UE will not interfere the DTT reception.
However, when the value of -42 dB/8MHz is considered within channel 21 of DTT, the interference is mitigated. It
should be noted that the requirement of UE OOBE for the protection of own DL requires a lower value of unwanted
emissions (-50 dBm/MHz) what make it possible to conclude that the LTE UE will not interfere the DTT reception.
Page 259
For this study, LTE uplink band is starting at 452.5 MHz and downlink band at 462.5 MHz as illustrated by the
following Figure 145. The basis is LTE Band 31.
Studies on the protection of DTT above 470 MHz from LTE BS in the 450-470 MHz range were carried out in ECC
Report 240 section 3.5.3.1 (Minimum Coupling Loss) and 3.5.1.4 (Monte Carlo).
Especially for the Monte Carlo simulations the absolute level of the interference probability is likely to be different.
However, using the same method of interpreting the results, the same conclusion can be drawn.
LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz are given in Table 161 below.
Table 161: PPDR 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz
It should be noted that these levels will not remove interference into DTT in all cases. This is similar to LTE in 800
MHz. In such a case additional mitigation techniques will need to be applied. For a list of possible mitigation
techniques see ANNEX 2: (list of mitigation measures).
Studies carried out for compatibility between LTE800 and portable DTT reception concluded that portable DTT
reception is less susceptible to interference from base stations29. Additional studies are not required. If fixed DTT
reception is protected from base station interference, portable DTT reception is automatically protected.
The UE out-of-band emissions level necessary to protect a TV receiver using a fixed rooftop antenna from
interference from a UE located outdoors is established using a MCL analysis.
In some studies, the effect of body loss was taken into account for the LTE UE by an additional attenuation of 4 dB
(taken from Report ITU-R M.2292 [5]), in order to simulate e.g. handheld devices (mobile terminals). In other
studies, this effect was not applied in order to simulate devices not used very close to the human body, e.g. Wifi
Routers or nomadic installations.
The following assumptions have been used in the analysis of the out-of-band emissions level needed to protect
fixed DTT reception.
TV Receiver
Noise figure 6 dB
Antenna height 10 m
Note that the same directional pattern is used both in azimuth and elevation, i.e., the curves represent g,(TV)() or
g,(TV)() where and are azimuth and elevation offsets from bore sight.
20
g,(TV)()
g,(TV)()
60
UE Transmitter
UE Transmitter
In some studies, the effect of body loss was taken into account for the LTE UE by an additional attenuation of 4 dB,
in order to simulate e.g. handheld devices (mobile terminals). In other studies, this effect was not applied in order to
simulate devices not used very close to the human body, e.g. broadband wireless terminals and mobile TV
receivers.
A10.3.3 Methodology
A MCL analysis is used for evaluating the impact of adjacent-channel interference from UEs to DTT receivers. The
situation is considered where the DTT signal is received at the reference sensitivity level, the worst case separation
distance between the TV antenna and the UE is established, accounting for both the path-loss and the elevation
pattern of a typical TV antenna, and the out-of-band emissions level which would result in a 1 dB desensitization of
the TV receiver is then evaluated.
It is assumed that the TV antenna is roof mounted (at a height of 10 m) and that the UE is outdoors (at a height of
1.5 m).
ECC REPORT <No> - Page 262
The worst-case UE to TV antenna horizontal separation distance is established by considering both the path-loss
between the UE and the TV antenna and the elevation pattern of the TV antenna.
For the path-loss the free-space model is used together with the TV antenna elevation pattern from ITU-R BT.419-
3 [18], see below.
The path gain between the UE and the TV receiver is calculated as follows:
where:
As can be seen, the worst-case occurs at a horizontal separation distance of 22 m, where the total coupling gain
between the UE and the TV receiver is 44.3 dB.
Having established the total path gain for the worst-case horizontal separation between the UE and TV antenna,
the out-of-band emissions needed to meet the 1 dB desensitisation criteria is calculated.
where:
k = Boltzmanns constant;
T = Temperature (290 K);
B = Noise equivalent bandwidth of the TV receiver (7.6 MHz);
NF = DVB-T2 receiver noise figure (6 dB) .
The interference power in the TV receiver adjacent channel is calculated from a combination of the UE in-band
power (23 dBm) and the total path gain (including 4 dB body loss at the UE) at the worst-case distance as follows:
PAC PUE , IB GPG,(UE ,TV ) GBL 23.0 (44.26) (4.0) 25.26 dBm
From the above the adjacent-channel interference ratio (ACIR) can be established as follows:
ECC REPORT <No> - Page 264
without body loss (e.g. for a broadband wireless internet terminal) this would be 83.74 dB.
ACIR is related to the adjacent channel selectivity (ACS) of the victim and to the adjacent-channel interference ratio
(ACLR) of the interferer via the following expression (linear units):
However, with an assumption about reasonable improvement in TV receiver ACS by means of additional external
filtering in the antenna down lead it can be concluded that an ACS figure of 80 dB or better is achievable. Also
measurement as reported in ECC Report 240 showed that 80 dB is achievable with current receiver design.
Thus for the purposes of this calculation an ACS value of 80 dB has been used.
1 1
ACLR 10 log 10 1 79.78 / 10 80 / 10 92.84 dB
10 10
Thus for a UE transmitting at 23 dBm e.i.r.p. the out-of-band emissions will be:
This value can be rounded to -70 dBm /8 MHz. This means that an LTE user equipment BEM out-of-band
emissions limit of -70 dBm/(8 MHz) for frequencies below 790 MHz is necessary to protect fixed DTT reception.
The UE out-of-band emissions level necessary to protect portable TV reception from interference from a UE is
established using MCL analysis.
TV Receiver
UE Transmitter
General
A10.3.8 Methodology
An MCL analysis is used for evaluating the impact of adjacent-channel interference from UEs to DTT receivers.
The situation is considered where the DTT signal is received at the reference sensitivity level. The victim TV
antenna and the interfering UE are assumed to be in the same building. Some of the MCL calculations assume that
they are separated by one internal wall. It can be argued that if the victim and interferer are in the same room then
the users of both devices can negotiate a local solution in case of interference, e.g. one of them can move to
increase the distance between the victim and interferer, or, if necessary, move to another room. For various
assumed values of the UE out-of-band emissions level, the separation distance needed to meet the 1 dB
desensitisation criteria is evaluated (taking account of the wall loss). A value for the out-of-band emissions level is
then chosen which balances the need to minimise the separation distance and be achievable in a realistic terminal
design.
where:
k = Boltzmanns constant;
T = Temperature (290 K);
B = Noise equivalent bandwidth of the TV receiver (7.6 dB);
NF = TV receiver noise figure (6 dB).
The interference power at the source UE (PI,(UE)) is a combination of the UE in-band power (PIB,(UE) = 23 dBm)
the ACS of the victim TV receiver and out-of-band emission power of the UE (POOB,(UE)) within the victim receivers
channel as follows:
PI ,(UE ) 10 log 10 10
PIB ,(UE ) ACS 10
10 OOB,(UE )
P 10
For the purposes of this calculation a minimum achievable ACS value of 85 dB has been assumed. This takes into
account that an ACS is achievable with current receiver design as shown in ECC Report 240 and some rejection in
the TV receiver antenna.
Results have also been calculated for an ACS value of 100 dB to demonstrate the impact of additional rejection
filters at the portable TV receiver.
Page 267
The minimum allowed coupling gain between the interfering UE and the victim TV is therefore the difference
between the target interference power (PI) and the interference power at the source UE (PI,(UE)).
GCG PI PI ,(UE )
The total path gain between the interfering UE and the victim TV (GPG,(UE,TV)) is given by the allowed coupling
gain GCG minus the wall loss (GWL = -10.4 dB) minus the body loss at the UE (GBL = -4 dB) minus the TV antenna
gain (GA,(TV) = 2.15 dBi).
From the total path gain we can then calculate the minimum separation distance needed to meet the 1 dB
desensitisation criteria using the free-space path-loss model.
A10.3.10 Results
As indicated above, for various assumed values of the UE out-of-band emissions level, the separation distance
needed to meet the 1 dB desensitisation criteria has been evaluated. Results have been obtained for assumed TV
ACS values of both 85 dB and 100 dB (to assess the impact of rejection filters at the portable TV receiver).
TV ACS = 85 dB
The graph below illustrates the relationship between separation distance and out-of-band emissions. The lower
blue curve takes into account -10.4 dB wall loss whereas the upper pink curve does not.
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
As can be seen, the curves have essentially flattened out for a out-of-band emissions level of -75 dBm/(8 MHz) and
below i.e. for out-of-band emissions levels lower that -75 dBm/(8 MHz) there is minimal improvement in separation
distance. From this it is concluded a UE out-of-band emission level of -75 dBm/(8 MHz) is optimal.
The following table summarises the calculation of separation distance for the situation where the assumed TV
receiver ACS is 85 dB and the out-of-band emissions is set to -75 dBm/(8 MHz) for the various combinations of wall
loss and body loss.
ECC REPORT <No> - Page 268
Target performance
Receiver NF dB 6.00 6.00 6.00 6.00 NF
Thermal Noise floor (9
dBm -99.17 -99.17 -99.17 -99.17 Pn= 10 log(kTB) + NF + 30
MHz)
INR dB -6.00 -6.00 -6.00 -6.00 INR
Target interference
dBm -105.17 -105.17 -105.17 -105.17 PItarget= Pn + INR
power
Victim's performance
Receiver selectivity
dB 85.00 85.00 85.00 85.00
(ACS)
BEM limits
In-block transmit dBm/
23.00 23.00 23.00 23.00 Pib,tr
power 10MHz
Interferer antenna gain dBi 0.00 0.00 0.00 0.00 Ga,i
dBm/
e.i.r.p. 23.00 23.00 23.00 23.00 Pib
10MHz
dBm/
Out-of-block -75.00 -75.00 -75.00 -75.00 Poob
8MHz
"Total" interference at Linear: Px = Pib/ACS + Poob,
dBm -61.79 -61.79 -61.79 -61.79
"source" where PItarget = G Px
ACIR calculation
ACLR dB 98.00 98.00 98.00 98.00 Pib - Poob
Linear = 1/((1/ACLR) +
ACIR dB 98.21 98.21 98.21 98.21
(1/ACS))
Coupling calculation
Coupling gain dB -43.48 -43.48 -43.48 -43.48 Linear: G = PItarget -Px
Link budget
Interferer body gain dB -4.00 -4.00 0.00 0.00 Gb,I
Wall gain dB -10.40 0.00 -10.40 0.00 GWl
Victim body gain dB 0.00 0.00 0.00 0.00 Gb,v
Victim ant. Elevation
dB 0.00 0.00 0.00 0.00 gb,v (assumed zero)
pattern
Victim antenna gain dB 2.15 2.15 2.15 2.15 Ga,v
GpI = G -Gb,I -GWl -gb,v -Ga,v
Path gain dB -31.13 -41.53 -35.13 -45.53
-Gb,v
Geometry
d, where GpI = 147.56 -
Protection distance m 1.89 6.26 3.00 9.92
20log10(f) - 20log10(d) dB
Page 269
TV ACS = 100 dB
In order to assess the impact of a rejection filter fitted to the portable TV receiver a further set of results are
calculated but with an ACS value of 100 dB (rather than 85 dB assumed above).
The graph below provides results where the UE body loss is set to 4 dB.
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
The graph below provides results where the UE body loss is set to zero.
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-90.00 -80.00 -70.00 -60.00 -50.00 -40.00 -30.00
OOBE level (dBm/8 MHz)
No wall One wall
The following table summarises the calculation of separation distance for the situation where the assumed TV
receiver ACS is 100 dB and the out-of-band emissions is set to -75 dBm/(8 MHz) for the various combinations of
wall loss and body loss.
Target performance
Receiver NF dB 6.00 6.00 6.00 6.00 NF
Thermal Noise floor (9
dBm -99.17 -99.17 -99.17 -99.17 Pn= 10 log(kTB) + NF + 30
MHz)
INR dB -6.00 -6.00 -6.00 -6.00 INR
Target interference
dBm -105.17 -105.17 -105.17 -105.17 PItarget= Pn + INR
power
Victim's performance
Receiver selectivity
dB 100.00 100.00 100.00 100.00
(ACS)
BEM limits
In-block transmit dBm/
23.00 23.00 23.00 23.00 Pib,tr
power 10MHz
Interferer antenna gain dBi 0.00 0.00 0.00 0.00 Ga,i
dBm/
e.i.r.p. 23.00 23.00 23.00 23.00 Pib
10MHz
dBm/
Out-of-block -75.00 -75.00 -75.00 -75.00 Poob
8MHz
"Total" interference at Linear: Px = Pib/ACS + Poob,
dBm -72.88 -72.88 -72.88 -72.88
"source" where PItarget = G Px
ACIR calculation
ACLR dB 98.00 98.00 98.00 98.00 Pib - Poob
Linear = 1/((1/ACLR) +
ACIR dB 102.12 102.12 102.12 102.12
(1/ACS))
Coupling calculation
Coupling gain dB -32.29 -32.29 -32.29 -32.29 Linear: G = PItarget -Px
Link budget
Interferer body gain dB -4.00 -4.00 0.00 0.00 Gb,I
Wall gain dB -10.40 0.00 -10.40 0.00 GWl
Victim body gain dB 0.00 0.00 0.00 0.00 Gb,v
Victim ant. Elevation
dB 0.00 0.00 0.00 0.00 gb,v (assumed zero)
pattern
Victim antenna gain dB 2.15 2.15 2.15 2.15 Ga,v
GpI = G -Gb,I -GWl -gb,v -Ga,v
Path gain dB -20.04 -30.44 -24.04 -34.44
-Gb,v
Page 271
Geometry
d, where GpI = 147.56 -
Protection distance m 0.53 1.75 0.84 2.77
20log10(f) - 20log10(d) dB
A10.4 CONCLUSION
The studies carried out for PPDR base stations in 400 MHz in ECC Report 240 concluded on a set of out-of-band
emissions from base stations. The scenario for this report is similar and therefore the same limits should apply.
However, additional mitigation measures may be required to solve possible residual interference from LTEin400
BSs on a case by case basis in a manner similar to the situation between LTE800 and DTT (see also ANNEX 2:
(list of mitigation measures)).
LTE 400 Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz are given in Table 171 below.
Table 171: LTE Base Station OOBE e.i.r.p. levels for protection of DTT above 470 MHz
The MCL studies have shown that the unwanted emissions above 470 MHz need to be limited in order to minimize
interference.
In order to protect DTT in channel 21 and above the LTE unwanted emission level should not exceed -70
dBm/8 MHz for fixed reception and -75 dBm/8 MHz for portable reception.
ECC REPORT <No> - Page 272
A11.1 INTRODUCTION
ECC Report 240 concluded that a LTE UE unwanted emission level of -42 dBm/8MHz. This level was initially
derived for the protection of DTT below 694 MHz for UE emissions above 703 MHz. The assumed level of DTT
degradation for DTT channels below 694 MHz was 5.6 dB. ECC Report 240 concluded that Monte-Carlo
simulations demonstrated limited interferences to DTT for high power UE (37 dBm) with improved ACLR (79 dB, Commented [BP34]: Consistency with or without
i.e. OOBE of -42 dBm / 8 MHz) in Channel 21. hyphnens?
A11.2 UE OOB EMISSIONS LEVEL TO PROTECT FIXED DTT RECEPTION BASED ON SEAMCAT
SIMULATIONS
ECC Report 240 concluded that Monte-Carlo simulations demonstrated limited interferences to DTT for high power
UE (37 dBm) with improved ACLR (79 dB, i.e. OOBE of -42 dBm / 8 MHz) in Channel 21. Below, SEAMCAT
simulations between IoT UE and DTT Channel 21 have been performed using the parameters described in Annex
A1.1, A1.8 and assuming a UE IoT unwanted emission level above 470MHz of -42 dBm/8MHz. In ECC Report 240
BB PPDR devices were used which in comparison to IoT devices can use many RBs simultaneously, have higher
transmit power and are mainly outdoor.
Two scenarios have been simulated. Scenario 1 describes the worst case, on which fixed DTT receivers are
located at the DTT cell edge while in scenario 2, the DTT receivers are randomly allocated within the DTT cell area,
representing a more realistic scenario. The IoT network is intentionally placed around the DTT receiver to ensure
proximity between the IoT UE and the DTT receiver. In both scenarios the IoT UEs are placed within 50 meters of
the DTT receivers. 3 MHz bandwidth is used for the LTE base station where 15 UEs are actively transmitting in
both of the scenarios with one RB each of 180 kHz. Transmit power for the IoT UEs are power controlled between -
40 to 23 dBm.
Table 172:XX
Scenario 1 4.59 %
Scenario 2 0.00 %
The MCL studies have shown that for LTE UE transmitting at maximum output power and at the worst case
separation from DTT, the unwanted emissions above 470 MHz need to be limited in order to to protect DTT in
channel 21 to-70 dBm/8 MHz for fixed reception and -75 dBm/8 MHz for portable reception. This is under the
condition of DTT ACS values of 80 dB and 85 dB, respectively. It should be noted that LTE includes power control
for the UEs and that the UEs are moving devices. The level of DTT sensitivity degradation considered in this
analysis is 1dB, while in ECC Report 240, a level of 5.6 dB was considered.
SEAMCAT simulations show limited probability of interference assuming a DTT protection level of -42 dBm/8MHz.
This is in line with the conclusion of BB PPDR UE emissions towards DTT in ECC Report 240,
The study presented in this contribution is based on the SEAMCAT files kindly provided by ANFR. The files were
used in the study presented in contribution SE7(17)035. The scenario use 3 MHz base station bandwidth. The
following parameters have been modified for the new study presented in this contribution:
In total the new modified workspaces only contain a few changes from the original study file as presented above.
This makes it easier to compare the studies.
A11.5 RESULTS
Scenario Pinterference
LTE cell radius 2.934 km 2.78%
LTE cell radius 8.5 km 1.44%
ECC REPORT <No> - Page 274
ANNEX 12:ANNEX 9: GUIDANCE ON MEANS TO SOLVE INTERFERENCE CASES BETWEEN LTE AND DTT
This annex provides a list of potential mitigation techniques which may be considered by national administrations to
solve or minimise the interference cases between LTE450 and terrestrial broadcasting on a local / regional /
national basis. They would need to be implemented in addition to the techniques (BEM and guard band) addressed
in this Report. It should be noted that this list is not exhaustive and that, for example, additional spectrum
engineering techniques may be considered, such as additional frequency offset or restricted BEM.
LTE cell coverage area is reduced due to insertion loss of the filter.
An improved filter would limit the OOB emissions but not the in-band
Improved filters in LTE emissions. This improves adjacent channel compatibility but not blocking
BS transmitters and overloading.
(at 470 MHz) All countries that introduced LTE800 ensured that the OOB emissions
were less than the most stringent limit in CEPT Report 30 [11][11]. This
makes it a generally applicable mitigation solution.
A12.3A9.3 DISCUSSION
There may be areas/regions where interference to the fixed and/or portable indoor DTT reception is likely to occur.
From this first assessment, it can be assumed that a single mitigation technique may not be sufficient to protect
broadcasting services from interference by LTE. A combination of two or more mitigation techniques may lead to a
sufficient protection of broadcasting services.
The mitigation measures to avoid interference caused by adjacent OOB emissions differ from those for blocking or
overloading by in-band emissions. Blocking and overloading are likely to occur by LTE transmission in close vicinity
to the DTT reception; in the case of portable reception the interference will be dominated by the LTE terminal. The
adjacent OOB interference is likely to be caused by the LTE BS operating just below 470 MHz.
For most of the techniques mentioned above e.g. appropriate filters and for choosing a proper combination of
mitigation techniques, national experience has been collected for the similar case of LTE800. It provides a basis to
apply them in a similar manner for LTE450.
Page 277
Transmission loss, calculating with free space loss and ITU-R P.526-13
ECC REPORT <No> - Page 278
The results of the two models ITU-R P. 526 and EPM73 are coherent. Beyond a specific distance, the two curves
are differing. This is because EPM73 model is considering that tropospheric diffusion is the main mode of
propagation. This mode of propagation (tropospheric diffusion) is not taken into account by ITU-R P. 526 model.
Page 279
The additional transmission loss due to diffraction over a spherical Earth can be computed by the classical residue
series formula. The ITU-R P.526-13 contains the applicable calculation method. A computer program GRWAVE,
available from the ITU, provides the complete method. GRWAVE was used to obtain the pathloss.
Figure xx: Transmission loss, calculating with free space loss and ITU-R P.526-13
ECC REPORT <No> - Page 280
Propagation model
LOS
LOS/NLOS reflection
Propagation LOS LOS
reflection diffraction
mechanism diffraction diffraction
diffraction
tropospheric
scatter
The three distance regions can be specified in the VHF band. In the first region the most decisive propagation
mechanism is LOS propagation and reflection. The limit of the first region is d1 distance. This distance is close to
the radio horizon. Propagation by diffraction is determinative in the second region, called diffraction region. In the
last region the most decisive propagation mechanism is tropospheric scatter. Boundary of the second and third
regions is represented by the d2. It is shown by the Figure 50.
Page 281
The accuracy of the models that predict the pathloss depends on the input parameters and the used propagation
mechanism. For short distances (< 60 km) General 450 and MYRIAD model provide more realistic and accurate
result, than EPM73 and ITU-R P.526-13 [31][31], because these models use digital maps tuned by measurements.
Over these distances the EPM73 and ITU-R P.526-13 ensure very similar result. It is because both of the models
based on propagation by diffraction in the diffraction region. In the tropospheric scatter region only EPM73 provides
reliable result, because only this model uses the tropospheric scatter propagation mechanism.
ECC REPORT <No> - Page 282
Dec Dec
LTE-BS Radar LTE-BS Ground Radar MCL (dB) Separation distance (km)
(Polar) (Ant)
dBm/MHz Be/Br=0
Victim main
Adjacent- 6 15 2 38.5 0 -115.9 0 3 170.4 / / 72.5
beam
channel
Victim main
Adjacent- -1 15 2 38.5 0 -115.9 0 3 163.4 / / 62.9 58
beam
channel
Dec Dec
LTE-BS Radar LTE-BS Ground Radar MCL (dB) Separation distance (km)
(Polar) (Ant)
Victim main
-26 15 2 38.5 0 -115.9 0 3 138.4 / / 31.3 35
Spurious beam
Note 1: He = 30 m, Hr = 8 m
Note 2: In adjacent band Pe corresponds to OOB level, for a 1.4 MHz channel. For this reason MCL columns (3 MHz and 5 MHz) are colored in grey
ECC REPORT <No> - Page 285
*: The additional transmission loss due to diffraction over a spherical Earth can be computed by the GRWAVE
computer program (available from the ITU) to obtain the pathloss according to the classical ITU-R P.526-13
formula. The separtation distance is 111 km for He=30 m, 116 km for He=40 m and 119 km for He=50 m.
Dec Dec
Separation
LTE-BS Radar LTE-BS Ground Radar MCL (dB)
(Polar (Ant distance (km)
) )
Interfere Pe
UIT-
r main IC
G Pf PF R
beam dBm Gr dBm EPM7
Scenario e e r
dB (1MHz 3 (1)
dB dB dB P.52
directed dBm/MH )
6 (2)
towards: z
Be/Br=0
Victim
38. 177.
Adjacent main 6 15 2 0 -115.9 0 10 / / 82.4 82
5 4
-channel beam
F_1 Victim
(20) 143.
side 6 15 2 5 0 -115.9 0 10 / / 37.8 40
9
lobes
Victim
38. 156.
Adjacent main -1 15 2 0 -115.9 0 10 / / 53.6 54
5 4
-channel beam
F_2 Victim
(13) 122.
side -1 15 2 5 0 -115.9 0 10 / / 15.6 18
9
lobes
Victim
38. 151.
Adjacent main -6 15 2 0 -115.9 0 10 / / 47.1 45
5 4
-channel beam
F_3 Victim
(8) 117.
side -6 15 2 5 0 -115.9 0 10 / / 11.6 12
9
lobes
Note 1: He = 30 m, Hr = 8 m
Note 2: In adjacent band Pe corresponds to OOB level, for a 1.4MHz channel. For this reason MCL columns (3
MHz and 5 MHz) are colored in grey
There were two different tuned propagation models used for the adjacent channel scenario.
The General 450 model, applicable in the frequency range extending from 150 MHz to 2 GHz, is based upon the
Okumura-Hata (COST 231) empirical model, with a number of additional features to enhance its flexibility.
The MYRIAD model, applicable in the frequency range extending from 200 MHz to 5 GHz, is a sophisticated and
versatile propagation model that relies on highly realistic modelling of the propagation channel, and therefore
achieves synergy of the following three vital physical elements:
In addition, this model can produce LOS/NLOS information for each predicted location.
Based on the simulations the minimum separation distances are the followings (for 1.4 MHz channel bandwidth):
Table 164178: Propagation losses are corresponded to minimum distances (Channel width: 1.4MHz)
PunwantedemPPDR
5.43 -1 -6 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
18.43 12 7 -13
(dBm/MHz)
35 27 22 6
d (EPM73) 36 28 23 6
37 29 23 7
Page 287
33 26 21 8
40 32 26 10
Figure 144157: Separation distance as a function of frequency shift (Channel width: 1.4 MHz)
The distance between LTE and ground radar can be further reduced with appropriate filtering.
Table 165179: Propagation losses are corresponded to minimum distances (Channel width: 1.4 MHz)
PunwantedemPPDR
5.43 -1 -6 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
18.43 12 7 -13
(dBm/MHz)
Based on the simulations the minimum separation distances are the followings (for 3 MHz channel bandwidth):
Table 166180: Propagation losses are corresponded to minimum distances (Channel width: 3 MHz)
PunwantedemPPDR
4.44 -5 -6 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
17.44 8 7 -13
(dBm/MHz)
34 24 22 6
d (EPM73) 35 28 23 6
36 28 24 7
32 22 21 8
d (ITU-R P.526-13) (km)
37 26 25 9
Page 289
38 27 26 10
Figure 145158: Separation distance as a function of frequency shift (Channel width: 3 MHz)
The distance between LTE and ground radar can be further reduced with appropriate filtering.
Table 167181: Propagation losses are corresponded to minimum distances (Channel width: 3 MHz)
PunwantedemPPDR
4.44 -5 -6 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
17.44 8 7 -13
(dBm/MHz)
Based on the simulations the minimum separation distances are the followings (for 5 MHz channel bandwidth):
Table 168182: Propagation losses are corresponded to minimum distances (Channel width: 5 MHz)
PunwantedemPPDR
2.3 -4 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
15.3 9 -13
(dBm/MHz)
32 25 6
d (EPM73) 33 25 6
33 26 7
30 23 8
37 28 10
Figure 146159: Separation distance as a function of frequency shift (Channel width: 5 MHz)
The distance between LTE and ground radar can be further reduced with appropriate filtering.
Table 169183: Separation distance between LTE400 BS and ground radar (Channel width: 5 MHz)
PunwantedemPPDR
2.3 -4 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
15.3 9 -13
(dBm/MHz)
Dec Dec
Separatio
LTE-BS Radar LTE-BS Airborne Radar MCL (dB) n distance
(Pola (Ant
(km)
r) )
Interfer
Be/Br Be/Br
er main Be/Br=
= =
beam G G IC ITU-
Pe Pf PF
Scenari e r dBm 1/3 R
e r 1/1.4 1/5 EL
o directe d d (1MH P.52
dBm dB dB
d B B z) (4.8dB 8
(1.5d (6.9d
toward )
B) B)
s
Victim
37.5 41 2 - 400
main 15 2 0 1.5 3 182.4 182.6 182.5
43 2 115.9 (1)
Co- beam
channel
V side - 400
15 2 0 0 1.5 3 160 160.6 160.5
lobes 115.9 (1)
dBm/M Be/Br=
Hz 0
Adjacen Victim
2 - 400
t- main 6 15 2 0 0 3 153.9 / /
2 115.9 (1)
channel beam
F_1 V side -
(20) 6 15 2 0 0 0 3 131.9 / / 223
lobes 115.9
Adjacen Victim
2 - 400
t- main -1 15 2 0 0 3 146.9 / /
2 115.9 (1)
channel beam
F_2 V side -
(13) -1 15 2 0 0 0 3 124.9 / /
lobes 115.9
Adjacen Victim
2 - 400
t- main -6 15 2 0 0 3 141.9 / /
2 115.9 (1)
channel beam
F_3
(8) V side -6 15 2 0 0 - 0 3 119.9 / / 99.
Page 293
Dec Dec
Separatio
LTE-BS Radar LTE-BS Airborne Radar MCL (dB) n distance
(Pola (Ant
(km)
r) )
lobes 115.9 9
Victim
2 - 70.
Spuriou main -26 15 2 0 0 3 121.9 / /
2 115.9 7
s beam
(-13) V side -
-26 15 2 0 0 0 3 99.9 / / 5.6
lobes 115.9
Note 2: airborne radars do not operate in the 420-430 MHz band according to the ITU database
Table 171185: Propagation losses are corresponded to minimum distances (Channel width: 1.4 MHz)
PunwantedemPPDR
5.43 -1 -6 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
18.43 12 7 -13
(dBm/MHz)
Table 172186: Propagation losses are corresponded to minimum distances (Channel width: 3 MHz)
PunwantedemPPDR
4.44 -5 -6 -26
(dBm/MHz) 3GPP
ECC REPORT <No> - Page 294
e.i.r.p.unwantedemPPDR
17.44 8 7 -13
(dBm/MHz)
Table 173187: Propagation losses are corresponded to minimum distances (Channel width: 5 MHz)
PunwantedemPPDR
2.3 -4 -26
(dBm/MHz) 3GPP
e.i.r.p.unwantedemPPDR
15.3 9 -13
(dBm/MHz)
The distance between LTE and airborne radar can be reduced with appropriate filtering.
Table 174188: Separation distance between LTE400 BS and airborne radar (Channel width: 1.4 MHz)
Amount of filtering
10 20 30 40 50 60
[dB]
Page 295
Amount of filtering
10 20 30 40 50 60
[dB]
Table 175189: Separation distance between LTE400 BS and airborne radar (Channel width: 3 MHz)
Amount of filtering
10 20 30 40 50 60
[dB]
Table 176190: Separation distance between LTE400 BS and airborne radar (Channel width: 5 MHz)
Amount of filtering
10 20 30 40 50 60
[dB]
Dec Dec
Separation
LTE-BS Radar LTE-BS Airborne Radar MCL (dB)
distance (km)
(Polar) (Ant)
Interferer
Pe
main
IC
beam Ge Pfe Gr PFr UIT-
Scenario dBm dBm EL
dB dB dB dB R528
(1MHz)
directed
dBm/MHz
towards
Be/Br=0
Adjacent-
channel Victim
400km
main 6 15 2 22 0 -115.9 0 10 146.9 / /
(1)
F_1 beam
(20)
Adjacent-
channel Victim
400km
main -1 15 2 22 0 -115.9 0 10 139.9 / /
(1)
F_2 beam
(13)
Adjacent-
Victim
channel
main -6 15 2 22 0 -115.9 0 10 134.9 / / 316
beam
F_3 (8)
Spurious Victim
main -26 15 2 22 0 -115.9 0 10 114.9 / / 31.6
(-13) beam
For airborne radars the decoupling antenna factor depends on the radar positions. The decoupling antenna factor
also depends on the LTE BS antenna mask, and the airborne radar pattern. Assuming the LTE BS antenna is not
tilted, and the airborne radar antenna main lobe can scan 60 elevation and 360azimuth.
The attenuation of LTE BS antenna (Kathrein, K742 242) has been calculated for several airborne positions and
angles which could be considered as DECant factor.
Table 178192: The attenuation of LTE BS antenna (Kathrein, K742 242) as a function of airborne position
Table 179193: Separation distance between LTE400 BS and airborne radar (Channel width: 1.4 MHz)
ECC REPORT <No> - Page 298
Scenarios 1 2 3 4 5 6
Airborne radar
9000 9000 9000 9000 9000 9000
height [m]
Distance 391
between LTE
BS and 10 20 50 100 200 (radio
airborne radar horizon on
[km] 9000 m)
Angle above
65 26.7 10.4 5 2.5 1.3
horizontal*
K742 242
27.3 14.6 11.8 4.6 2.5 1.7
attenuation(*)
[dB]
Scenarios 7 8 9 10 11 12
Airborne radar
5000 5000 5000 5000 5000 5000
height [m]
Distance 291
between LTE
BS and 10 20 50 100 200 (radio
airborne radar horizon on
[km] 5000 m)
Angle above
30 14.5 5.7 2.9 1.4 1
horizontal*
K742 242
18.3 25 5.1 3 1.7 1.7
attenuation(*)
[dB]
Page 299
Illustration of PPDR LTE mask on spectrum with a 1 MHz guard band related to 430 MHz
60
PPDR
1,4
50 PPDR
3
40 PPDR
5
30
20
10
MHz
0
405 410 415 420 425 430 435 440 445
-10
-20
-30
RADIOLOCATION
-40
The figure below illustrates what are the possible center frequency offsets () of the LTE BS channel and
corresponding guard band with respect to the radiolocation primary band. Adjacent channel compatibility is feasible
if the OOB and spurious emissions of the LTE falling within the radiolocation primary band do not exceed the
protection criteria
60 F3
40
20
0
420 425 430
-20
-40
RADIOLOCATION
F1 = 3.5 MHz, and improving F2 = 4.5MHz and improving OOB: F3 = 7.5 MHz and Improving
OOB: 31 dB 32 dB OOB : 33 dB
Considering the proposed value of GB = 2.5 MHz, the corresponding minimum offsets are:
Channel Bandwidth
(MHz) 3.2 4 5
Page 301
Conditions: Propagation loss 135.8 dB, BS Height 30 m, radar height 8m, calculation frequency 420 MHz, on a site
in FRANCE
First calculation: with a statistic propagation model Second calculation: with a deterministic propagation
which takes into account antenna heights model named CARDIF (calculation including clutter), on a
site in France. Result below
EPM73 propagation model (it has been shown that
distances calculated with EPM73 and UIT-R P.526 are
very close, for distances up to about 150km)
To confirm the previous results, another case is computed in the same condition, but with a different simulation
tool and on a different site (Sweden)
Page 303
Radius of 28km With Cardif the area is wider than EPM73 area in some
directions
A13.9A10.9 LTE400 BS OOB AND SPURIOUS EMISSION LIMIT (DBM) AND UNWANTED EMISSIONS
TARGET
The table below illustrates how much additional attenuation is required for LTE OOBE and spurious emissions so
as to meet acceptable power falling within the radiolocation primary allocated frequencies (430-440 MHz). Less
stringent additional filter attenuation is required if 10 dB antenna decoupling is considered. It should be noted that
the proposed values correspond to a free space propagation model between the airborne radar at 9000 m above
the ground and the LTE BS.
The required 9 km separation distance for free space model could be reduced for ground radar due to additional
attenuation from the environment. As an example, with the EPM73 propagation model, the same targeted OOBE
would correspond to a separation distance of 4 km
ECC REPORT <No> - Page 305
1.4 MHz 0.7 to 2.1 -4dBm 100 kHz 49.9 -53.9 dBm 42.9 -46.9 dBm
2.1 to 3.5 -11 dBm 100 kHz 42.9 -53.9 dBm 35.9 -46.9 dBm
3.5 to 9.95 -16 dBm 100 kHz 37.9 -53.9 dBm 30.9 -46.9 dBm
3 MHz 1.5 to 4.5 -5 100 kHz 48.9 -53.9 dBm 41.9 -46.9 dBm
4.5 to 7.5 -15 100 kHz 38.9 -53.9 dBm 31.9 -46.9 dBm
7.5 to 9.995 -16 100 kHz 37.9 -53.9 dBm 30.9 -46.9 dBm
5 MHz 2.5 to 7.5 -8 100 kHz 45.9 -53.9 dBm 38.9 -46.9 dBm
7.5 to 9.95 -14 100 kHz 39.9 -53.9 dBm 32.9 -46.9 dBm
30 MHz f < - 36 dBm 100 kHz 17.9 -53.9 dBm 10.9 -46.9 dBm
1000 MHz
Draft ECC REPORT [LTE 400] - Page 306
Draft ECC REPORT [LTE 400] - Page 307
The studies in ECC Report 240 considered a frequency offset of more than 10 MHz between the PPDR MS
and the PMSE MS. It implies that the Tx power of the system is going to be -36 dBm in 100 kHz. The
following table provides the results of calculations to assess the separation distances considering a
frequency offset of 10 MHz. It should be noted that considering the Extended Hata (Urban) model, the
distance is about 66 meters. Additional propagation models are also considered for the purpose of
comparison.
e.i.r.p. (dBm) -36 dBm in 100 kHz -36 dBm in 100 kHz -36 dBm in 100 kHz -36 dBm in 100 kHz
Tx power
-36 dBm in 100 kHz -36 dBm in 100 kHz -36 dBm in 100 kHz -36 dBm in 100 kHz
(dBm)
Gain -3 dB -3 dB -3 dB -3 dB
Body loss 4 dB 4 dB 4 dB 4 dB
e.i.r.p. in the -43 dBm in 100 kHz -43 dBm in 100 kHz -43 dBm in 100 kHz
direction of
-43 dBm in 100 kHz
the PMSE
equipment
e.i.r.p. in the -51.24 dBm in 15 kHz -51.24 dBm in 15 kHz -51.24 dBm in 15
direction of -51.24 dBm in 15 kHz
the PMSE kHz
equipment
PMSE
-117 dBm -117 dBm -117 dBm -117 dBm
sensitivity
PMSE
sensitivity + 3 -114 dBm -114 dBm -114 dBm -114 dBm
dB
C/I 17 dB 17 dB 17 dB 17 dB
I -131 dBm -131 dBm -131 dBm -131 dBm
Gain 0 dBi 0 dBi 0 dBi 0 dBi
Attenuation 79.76 dB 79.76 dB 79.76 dB
to meet the 79.76 dB
criterion
Distances 55 m 55 m 501 m 69 m
Assuming 4 dB body loss, it can be seen that considering a frequency offset of 10 MHz, the separation
distances between the PPDR MS and the PMSE MS would be at least 55 m.
Draft ECC REPORT [LTE 400] - Page 308
For this case, ECC Report 240 considered the two systems operating in adjacent blocks of spectrum. This is
further considered in the following tables where an offset of 1 MHz is considered between the frequency
used by the PMSE BS and the edge of the PPDR UE band.
Table 183197: LTEin400 MS impact on 25 kHz analogue FM BS offset 500 kHz TS.36.101 and 3/5 RB
Emission limits
-10 dBm in 30 kHz -13 dBm in 30 kHz -15 dBm in 30 kHz -10 dBm in 1 MHz
(dBm)
Gain -3 dB -3 dB -3 dB -3 dB
Body loss 4 dB 4 dB 4 dB 4 dB
e.i.r.p. in the
direction of the -35.24 dBm in 15
-20 dBm in 15 kHz -23 dBm in 15 kHz -25 dBm in 15 kHz
PMSE kHz
equipment
PMSE
-120 dBm -120 dBm -120 dBm -120 dBm
sensitivity
PMSE
sensitivity + 3 -117 dBm -117 dBm -117 dBm -117 dBm
dB
C/I 17 dB 17 dB 17 dB 17 dB
I -134 dBm -134 dBm -134 dBm -134 dBm
Gain 3 dBi 3 dBi 3 dBi 3 dBi
Attenuation to
meet the 117 dB 114 dB 112 dB 101.76 dB
criterion
Distances 876 m 720 m 631 m 323 m
The separation distances are ranging from 323 m to more than 800 m.
At a frequency offset of 1 MHz, the Tx power will be -10 dBm in 1 MHz, therefore a distance of 323 m
considering 4 dB body loss.
In 4his case, the systems are assumed to operate in adjacent block. The following table provides results for
an offset of 500 kHz.
Table 184198: LTEin400 BS impact on 25 kHz analogue FM MS 500 kHz frequency offset
Tx power (dBm) -4.57 dBm in 100 kHz -6.67 dBm in 100 kHz -7.7 dBm in 100 kHz
e.i.r.p. in 15 kHz 0.19 dBm in 15 kHz -1.91 dBm in 15 kHz -2.94 dBm in 15 kHz
PMSE sensitivity +
-114 dBm -114 dBm -114 dBm
3 dB
C/I 17 dB 17 dB 17 dB
Attenuation to
131.19 dB 129.09 dB 128.06 dB
meet the criterion
Offset of 1 MHz
Parameter 1.4 MHz 3 MHz 5 MHz
(ECC DEC(16)02
Tx power -8,14 dBm in 100 kHz -8.3 dBm in 100 kHz -8.4 dBm in 100 kHz
-51.24 dBm in 15
e.i.r.p. in 15 kHz -3.38 dBm in 15 kHz -3.57 dBm in 15 kHz -3.64 dBm in 15 kHz
kHz
PMSE sensitivity -117 dBm -117 dBm -117 dBm -117 dBm
PMSE sensitivity
-114 dBm -114 dBm -114 dBm -114 dBm
+ 3 dB
C/I 17 dB 17 dB 17 dB 17 dB
I -131 dBm -131 dBm -131 dBm -131 dBm
Gain MS PMSE 0 dBi 0 dBi 0 dBi 0 dBi
Attenuation to
79.76 dB 127.62 dB 127.43 dB 127.36 dB
meet the criterion
Distances 85 m 1.753 km 1.731 km 1.723 km
At 1 MHz, the separation distances are more than 1 km except if the limit given in ECC DEC(16)04 is
implemented.
Draft ECC REPORT [LTE 400] - Page 310
The following table provides the separation distances assuming an offset of at least 5 MHz as in ECC Report
240.
Table 186200: LTEin400 BS impact on 25 kHz analogue FM BS-5 MHz frequency offset
Tx power -16 dBm in 100 kHz -15 dBm in 100 kHz -14 dBm in 100 kHz
-51.24 dBm in 15
e.i.r.p. in 15 kHz -11.24 dBm in 15 kHz -10.24 dBm in 15 kHz -9.24 dBm in 15 kHz
kHz
PMSE sensitivity -120 dBm -120 dBm -120 dBm -120 dBm
PMSE sensitivity
-117 dBm -117 dBm -117 dBm -117 dBm
+ 3 dB
C/I 17 dB 17 dB 17 dB 17 dB
I -134 dBm -134 dBm -134 dBm -134 dBm
Gain MS PMSE 3 dBi 3 dBi 3 dBi 3 dBi
Attenuation to
85.76 dB 122.76 dB 123.76 dB 124.76 dB
meet the criterion
Distances 0.733 km 8.23 km 8.8 km 9.37 km
The separation distances are more than 8 km except if the limit given in ECC DEC(16)04 is implemented.
Draft ECC REPORT [LTE 400] - Page 311
A14.3A11.3 TDD
In case TDD links are considered, then, both the MS and the BS will be operating on the same frequency,
resulting in frequency offsets smaller than those considered in ECC Report 240. This section considered the
case of PPDR MS on PMSE MS and PPDR BS on PMSE BS.
The following table provides the results for an offset of 500 kHz.
Table 187201: LTEin400 MS impact on 25 kHz analogue FM MS-TS.36.101 and 3/5 RB500 kHz offset
Emission
-10 dBm in 30 kHz -13 dBm in 30 kHz -15 dBm in 30 kHz -10 dBm in 1 MHz
limits (dBm)
Gain -3 dB -3 dB -3 dB -3 dB
Body loss 4 dB 4 dB 4 dB 4 dB
e.i.r.p. in the
direction of -35.24 dBm in 15
-20 dBm in 15 kHz -23 dBm in 15 kHz -25 dBm in 15 kHz
the PMSE kHz
equipment
PMSE
-117 dBm -117 dBm -117 dBm -117 dBm
sensitivity
PMSE
sensitivity + 3 -114 dBm -114 dBm -114 dBm -114 dBm
dB
C/I 17 dB 17 dB 17 dB 17 dB
I -131 dBm -131 dBm -131 dBm -131 dBm
Gain 0 dBi 0 dBi 0 dBi 0 dBi
Attenuation to
meet the 111 dB 108 dB 106 dB 95.76 dB
criterion
Distances 108 m 97 m 93 m 77 m
The following table provides the results for an offset of 500 kHz.
Table 188202: LTEin400 BS impact on 25 kHz analogue FM BS 500 kHz frequency offset
Tx power -4.57 dBm in 100 kHz -6.67 dBm in 100 kHz -7.7 dBm in 100 kHz
e.i.r.p. in 15 kHz 0.19 dBm in 15 kHz -1.91 dBm in 15 kHz -2.94 dBm in 15 kHz
PMSE sensitivity -120 dBm -120 dBm -120 dBm
Draft ECC REPORT [LTE 400] - Page 312
PMSE sensitivity +
-117 dBm -117 dBm -117 dBm
3 dB
C/I 17 dB 17 dB 17 dB
I -134 dBm -134 dBm -134 dBm
Gain MS PMSE 3 dBi 3 dBi 3 dBi
Attenuation to
137.19 dB 135.09 dB 134.06 dB
meet the criterion
Distances 20.9 km 18.3 km 17.22 km
Draft ECC REPORT [LTE 400] - Page 313
According to Error! Reference source not found.Error! Reference source not found., the formula is the
amplitude distribution (note: not the power distribution)
U2
U
f AU e 2 2
2
Whereby: U: Amplitude of the receiving signal (field strength, etc.)
: Standard distribution
2 2: Quadratic mean value of the receiving signal amplitude
Transferal of the above relationship to power values. Whereby the following applies:
U2
P U PR
R
Whereby 00:
PR
PR
f A P e 2 2
2
2 2 U mittel Pmean R
2
and
P
PR
f A P e Pmean
1
Pmean R
2
Draft ECC REPORT [LTE 400] - Page 314
P
2 P
f A P e Pmean
R Pmean
The following relationship accurately reflects the reality of the received level distribution as it is quantified in
the interior of actual vehicles.
Probability density distribution for level below a power "P" (linear statement of power) if a Rayleigh
distribution can be assumed:
P
p A P 1 e Pmean
P
p A P 1 e Pmean
P
p A P 1 e Pmean
P dBm
10 10
Pmean dBm
P dBm Pmean dBm
10 10
10
1 e 10 10
1 e
1
P dBm Pmean dBm
1 e 10
10
p A P
The relationship found very accurately reflects the reality in the vehicle. The only drawback is that Pmean is
not the mean value defined across all measuring points as per the measuring instructions
1 mn
Pmean _ valuedBm m _ measured _ valuedBm
n m 1 30. Rather, if P = Pmean, pA (P) = 63.21 % and not the
required 50%. To counteract this deficit, the above relationship is provided with a correction value "kP":
1
P dBm Pmean_ va lue dBm k P
1 e 10
10
p A P
Calculation of "kP":
kP
1
0,5 1 e 10
10
k P
1
k P 1
ln 0,5 1010 log ln 0,5 k P
1
0,5 e 10
10
10
10 log ln 0,5 k P
k P 1,59dB
1
P dBm Pmean _ value dBm 1, 59dB
1 e 10
10
p A P
Whereby:
P[dBm]: e.g. Sensitivity of a receiver in dBm
1 mn
m _ measured _ valuedBm
PMean value[dBm]: dBm mean value ( n m1 )
across all levels at the receiver
pA (P): Describes the share of received levels below P[dBm].
1 mn
m _ measured _ valuedBm
Example: Averaged dBm level: -60 dBm ( n m1 )
Sought: Probability of received levels below -100 dBm:
1 e 10 0,00693%
10
p A P
Figure 190204: Calculated probabiltiy level density for Pmean_value = -60 dBm
100.0 %
10.0 %
1.0 %
0.1 %
0.01 %
0.001 %
-120dBm
-110dBm
-100dBm
-40dBm
-90dBm
-80dBm
-70dBm
-60dBm
-50dBm
-30dBm
Draft ECC REPORT [LTE 400] - Page 317
Now the density distribution should be considered. To this end, the probability distribution is derived:
Linear:
pAP 1 e Pmean
d
dP Pmean
Logarithmic:
ln 10 1 1010
d
10
dP
1
P dBm Pmean_ value dBm 1, 59dB
pAP ln 1010
d 10 1
P dBm Pmean_ value dBm1, 59dB
e 10
10
dP 10
1
P dBm Pmean_ value dBm 1, 59dB
ln 10 1010
1
P dBm Pmean _ value dBm1, 59dB
f A P e 10
10
10
Whereby:
M: Number of all measuring points
P[dBm]: Power in dBm for which the probability density applies
(e.g. sensitivity of a receiver)
1 mn
m _ measured _ valuedBm
PMean value[dBm]: dBm mean value (
n m1 )
across all levels at the receiver
fA (P):Describes the number of measuring points for an interval of 1 dB.
In the case of real measurements with "M" measuring points, the density distribution "fA (P)" is as follows:
1
P dBm Pmean_ value dBm 1, 59dB
ln 101010
1
P dBm Pmean _ value dBm1, 59dB
f A P M e 10
10
10
Whereby:
M: Number of all measuring points
P[dBm]: Power in dBm for which the probability density applies
Draft ECC REPORT [LTE 400] - Page 318
The following measurement relates to a real measurement taken in the vehicle. The original measurements
can be found in Error! Reference source not found.Error! Reference source not found. and Error!
Reference source not found.Error! Reference source not found..
m 20000
1
61,8dBm m _ measured _ valuedBm
20000 m 1
Messwerte values
real measurement
-40 dBm
-50 dBm
-60 dBm
-70 dBm
-80 dBm
-90 dBm
-100 dBm
0,0s 2,0s 4,0s 6,0s 8,0s 10,0s 12,0s 14,0s 16,0s 18,0s 20,0s
Density: Number
Dichte of
dermeasurement values
Messpunkte pro 1dBper dB Anzahl
Number of der Messpunkte
measurement mit geringerem
values with levelsPegel
below than
2500 100000
2000 10000
1000
1500
100
1000
10
500
1
0 -120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20
-120 dBm -110 dBm -100 dBm -90 dBm -80 dBm -70 dBm -60 dBm -50 dBm -40 dBm dBm dBm dBm dBm dBm dBm dBm dBm dBm dBm dBm
Draft ECC REPORT [LTE 400] - Page 320
1
P dBm Pmean_ value dBm1,59dB
ln 10 1010
1
1
P dBm Pmean_ value dBm1, 59dB P dBm Pmean_ value dBm 1, 59dB
1 e10
10
f A P M e 10
10
p A P
10
If an interference signal is present in the same space, the following correlations apply:
A15.2.4.1A12.2.4.1 Example
The cellular phone and key are located in the same place in the vehicle. E.g. the key and cellular phone are
carried by the vehicle user, whereby both components do not necessarily need to be in the same pocket. It is
sufficient if both devices are located in the direct vicinity of the individual (e.g. key in right pants pocket and
cellular phone in left pants pocket)
It is also assumed that the spurious emissions can be compared to white noise, which leads to a necessary
noise level signal for the useful signal receiver (vehicle) of
S/N =+13 dB
Figure 193207: Example of Probabiltiy density for key (-20 dBm) and LTE mobile (-36 dBm@100 kHz
spurious emission), while MS and key are located in the same area
Reception not possible
S/N = 13 dB
Successful
Reception
As the key and cellular phone do not have a defined position with respect to each other and are accepted at
random, the key and cellular phone levels received by the vehicle antennas are also distributed at random.
What can be derived from the above graphic: Let us assume that the key generates a level of -80 dBm at the
vehicle antenna. The vehicle receiver can only successfully receive this signal if
The minimum required distance between the useful signal and interference signal (here: 13 dB) is fulfilled.
(see green arrow at top)
If one now carries out a projection with respect to the probability density for the cellular phone (green line
from tip of arrow to red curve), there is a probability of approx. 1.3 %. I.e. 1.3 % of all base noises from the
cellular phone are sufficiently low and do not cause any interference. HOWEVER, all remaining levels cause
a failure consistent with
A15.2.4.2A12.2.4.2 Invoice
In order to acquire the sought intersecting point along the probability density curve for the interferer (red line),
one can assume the general probability density function
1
P dBm Pmean _ value dBm1, 59dB
1 e 10
10
p A P
1 mn
Pmean _ valuedBm measured _ valuedBm _ m
whereby n m 1
is introduced.
For the borderline case from which the levels of an interference signal (PInterf.[dBm]) prevent the useful
signal from being interpreted with PUseful[dBm] by the receiver, the following applies:
Applied to Error! Reference source not found.Error! Reference source not found., one finds
1
PIn terf
10 d BmS / N d B Pmea n _ va lu e_ S t r d Bm1, 5 9d B
p APUsefu ll 1 e 10
d Bm
Equation 00 describes the probability of base noises that lie above the limit value. These do not cause
interference with the receiver, however. The number of base noises that lie above the limit value is:
Draft ECC REPORT [LTE 400] - Page 323
p AInterf PUsefull
dBm
1 p AP Usefull dBm
1
dBm S / N dB Pmean_ value_ Interf dBm1, 59dB
PUsefull
1010
p AInterf PUsefull
dBm
e
Sample calculation:
S/N[dB] = 13 dB
e10 98,63%
10
p AInterf 80dBm
Now the frequency (statistically) of the useful signal must also be considered to obtain a quantifiable
statement in the end.
1
P dBm Pmean _ value dBm1, 59dB
ln 10 1010
1
P dBm Pmean _ value dBm1, 59dB
f A P e 10
10
10
and with the conventions from 00, the following formula can be derived:
1
PUseful dBm Pmean _ value Useful dBm1, 59dB
ln 10 1010
1
PUseful
10 dBmPmean _ valueUseful dBm1, 59dB
f APUseful e 10
dBm 10
f APNutzdBm
describes the frequency (statistically) of received levels in a 1 dB window.
For the failure probability in the presence of an interferer, all relevant received levels starting with the
lowest processable input power by the receiver (= sensitivity) must be added and weighted with
f A PUsefu l
d Bm
:
Draft ECC REPORT [LTE 400] - Page 324
To be added to the above failure probability are all failure probabilities that due to the limited reception
sensitivity of the receiver cannot be processed, which is also the case in the undisturbed state.
1
P dBm Pmean _ value dBm1, 59dB
1 e 10
10
p A P
Whereby P[dBm] becomes Psensitivity[dBm] and Pmean value[dBm] becomes Pmean value_usef.[dBm].
1 e 10
n 0 dB 10
1
PSensitivity n dB Pmean _ value_ Usefu l 1, 59dB
e 1010 dBm dBm
Whereby:
Psensitivity[dBm]: e.g. sensitivity of a receiver)
Draft ECC REPORT [LTE 400] - Page 325
1 mn
m _ measured _ valuedBm
Pmean_value_usef.[dBm]: dBm mean value ( n m1 )
across all
levels at receiver for the useful signal
1 mn
m _ measured _ valuedBm
Pmean_value_interf.[dBm]: dBm mean value ( n m1 )
across all levels at receiver for the interference signal
S/N[dB]: Minimum required useful signal for interference signal (signal to noise) in dB
so that the receiver can reliably process the useful signals.
Key -22 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Table 194208: Caluclation with key -22 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Useful
Key Mean reception performance -62.00 dBm
signal
Interference
LTEin400 Mean reception performance -76.00 dBm
signal
44.269%
Figure 195209: Caluclation with key -22 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Key -20 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Table 196210: Caluclation with key -20 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
bandwidth. As such, the value for 100 kHz represents a good estimate.
33.386%
Figure 197211: Caluclation with key -20 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Draft ECC REPORT [LTE 400] - Page 328
Key -18 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Table 198212: Caluclation with key 18 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
bandwidth. As such, the value for 100 kHz represents a good estimate.
24,025%
Figure 199213: Caluclation with key -18 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
Draft ECC REPORT [LTE 400] - Page 330
Key -18 dBm; LTE 400 = -46 dBm, same location (e.g. driver's seat)
Table 200214: Caluclation with key -18 dBm; LTE 400 = -36 dBm, same location (e.g. driver's seat)
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry. The system-determining bandwidth,
however, is the demodulation bandwidth, which is lower than the IF bandwidth. As
such, the value for 100 kHz represents a good estimate.
3.065%
Figure 201215: Caluclation with key -18 dBm; LTE 400 = -46 dBm, same location (e.g. driver's seat)
Draft ECC REPORT [LTE 400] - Page 331
Key -18 dBm (driver's seat); LTE 400 = -36 dBm (rear seat bench, 10 dB lower path attenuation)
Table 202216: Caluclation with key -18dBm; LTE 400 = -36 dBm, (rear seat bench, 10 dB lower path
attenuation)
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
bandwidth. As such, the value for 100 kHz represents a good estimate.
Draft ECC REPORT [LTE 400] - Page 332
75.975%
Figure 203217: Caluclation with key -18 dBm; LTE 400 = -36 dBm, (rear seat bench, 10 dB lower path
attenuation)
Draft ECC REPORT [LTE 400] - Page 333
Draft ECC REPORT [LTE 400] - Page 334
Key -18 dBm (driver's seat); LTE 400 = -96 dBm (rear seat bench, 10 dB lower path attenuation)
Table 204218: Caluclation with key -18dBm; LTE 400 = -96 dBm, (rear seat bench, 10 dB lower path
attenuation)
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
bandwidth. As such, the value for 100 kHz represents a good estimate.
0.001%
Figure 205219: Caluclation with key -18 dBm; LTE 400 = -96 dBm, (rear seat bench, 10 dB lower path
attenuation)
Draft ECC REPORT [LTE 400] - Page 335
Key -24 dBm (worst case); undisturbed (no LTE 400), receiver sensitivity -98.5 dBm (worst case)
Table 206220: Caluclation with key -24 Bm; worse receiver sensitivity (-98,5 dBm); no LTE 400
*) The receiver IF bandwidth is 165 kHz, which leads to a 10xlog (165 kHz/100
kHz) = 2.17 dB higher interference entry (-33.83 dBm). The system-determining
bandwidth, however, is the demodulation bandwidth, which is lower than the IF
Draft ECC REPORT [LTE 400] - Page 336
bandwidth. As such, the value for 100 kHz represents a good estimate.
0.869%
Figure 207221: Caluclation with key -24 Bm; worse receiver sensitivity (-98,5 dBm); no LTE 400
Draft ECC REPORT [LTE 400] - Page 337
Draft ECC REPORT [LTE 400] - Page 338
[29] 3GPP specification TS 36.101 v12.5.0 (2014-09); 3rd Generation Partnership Project; Technical
Specification Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA); User
Equipment (UE) radio transmission and reception (Release 12)
[30] 3GPP specification TS 36.104 v12.5.0 (2014-09) ; 3rd Generation Partnership Project; Technical
Specification Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA);
Base Station (BS) radio transmission and reception (Release 12)
[31] ITU-R P.526-13 (11/2013): Propagation by diffraction
[32] ITU-R P.525-3 (11/2016): Calculation of free-space attenuation
[33] SE7(17)013A1 draft ECC Report 338323
[34] ECC Report 252: SEAMCAT Handbook