This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

Digital Object Identifier IEEE ACCESS

A Comparison of CP-OFDM, PCC-OFDM

and UFMC for 5G Uplink
Communications
GAYATHRI KONGARA1 , CUIWEI HE1 , LEI YANG2 , JEAN ARMSTRONG1 (Fellow, IEEE)
1
Department of Electrical and Computer Systems Engineering, Monash University, Melbourne, Vic., Australia
2
Faculty of Information Technology, Monash University, Melbourne, Vic., Australia
Corresponding author: Jean Armstrong (e-mail: Jean.Armstrong@monash.edu).

ABSTRACT Polynomial-cancellation-coded orthogonal frequency division multiplexing (PCC-OFDM)

is a form of OFDM that has waveforms which are very well localized in both the time and frequency
domains and so it is ideally suited for use in the 5G network. This paper analyzes the performance of PCC-
OFDM in the uplink of a multiuser system using orthogonal frequency division multiple access (OFDMA)
and compares it with conventional cyclic prefix OFDM (CP-OFDM), and universal filtered multicarrier
(UFMC). PCC-OFDM is shown to be very much less sensitive to time and frequency offsets than either CP-
OFDM or UFMC. For a given constellation size, PCC-OFDM in additive white Gaussian noise (AWGN)
requires 3 dB lower signal-to-noise ratio (SNR) for a given bit-error-rate, and the SNR advantage of PCC-
OFDM increases rapidly when there are timing and/or frequency offsets. For PCC-OFDM no frequency
guard-band is required between different OFDMA users. PCC-OFDM is completely compatible with CP-
OFDM and adds negligible complexity and latency, because it uses a simple mapping of data onto pairs
of subcarriers at the transmitter, and a simple weighting-and-adding of pairs of subcarriers at the receiver.
The weighting-and-adding step, which has been omitted in some of the literature, is shown to contribute
substantially to the SNR advantage of PCC-OFDM. A disadvantage of PCC-OFDM (without overlapping) is
the potential reduction in spectral efficiency because subcarriers are modulated in pairs, but this reduction is
more than regained because no guard band is required and because, for a given channel, larger constellations
can be used.

INDEX TERMS 5G, multiuser, uplink, OFDM, PCC-OFDM, UFMC, timing offset, frequency offset

I. INTRODUCTION presence of multipath transmission, relative insensitivity to

The design of the 5G mobile network presents many new
challenges not faced by earlier generations of mobile access
technology. This is because of the wide range of diverse
services 5G will have to support [1]. There is the predicted
exponential increase in demand for very high bandwidth
connection resulting from video streaming and other high
data rate applications. At the same time, the emergence of
the internet of things (IOT) will produce a very large number
of low speed users. Reconciling these very different types of
communication is very challenging and is currently the topic
of extensive research [2]–[11].
A. CHALLENGES IN USING OFDM IN 5G
Orthogonal frequency division multiplexing (OFDM) in the
form of CP-OFDM (cyclic prefix OFDM) has been the
basis of many recent wireless communication systems. The
many advantages of CP-OFDM include robustness in the
timing offsets, compatibility with multiple-input multiple-
output (MIMO) systems and the ability to support multiple
access in the form of orthogonal frequency division multiple
access (OFDMA). The well-known disadvantages of OFDM
include high out-of-band (OOB) power, sensitivity to fre-
quency offset, and high peak-to-average power ratio (PAPR)
[12], [13].
In the current generation of mobile systems, workable
solutions have been found which overcome the disadvantages
of CP-OFDM. The OOB power can be reduced by leav-
ing some band-edge subcarriers unused and by windowing
within the cyclic prefix. Sophisticated synchronization algo-
rithms overcome the problem of frequency sensitivity. A mul-
titude of solutions to the PAPR problem have been described
in the literature [14]. In the LTE uplink a modified form
of OFDM called DFT-spread OFDM has been developed to
enhance capacity and cell-edge user throughput performance.

VOLUME 4, 2016 1

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access
G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications
The advantages and challenges of using DFT-spread OFDM
in 5G are discussed in [15]. But simple adaptions to these
techniques are not adequate for 5G. This has inspired exten-
sive recent research into different waveforms [2]–[11], [15].
OFDMA [1], [16] using CP-OFDM may work well in
the downlink and meet the demand for higher bandwidth,
the requirements of massive MIMO, and the use of higher
frequency bands. However, it is not suitable for the uplink
of the massive machine-type communication (mMTC) which
will result from the IOT. Wunder et al. [17] conclude that the
strict synchronization required in OFDMA using CP-OFDM
will not be appropriate for the sporadic traffic generated
by the IOT for two reasons. Firstly many of these devices
will be battery powered and will spend most of the time
in a dormant state, awakening only periodically to transmit
data, and secondly because of the hardware, time, and en-
ergy, that is required for strict synchronization. Hence the
search for waveforms that can be used by these devices in
the uplink, and which are tolerant to time and frequency
offsets. The waveforms which have been considered so far,
are well localized in the time and frequency domain but
are not necessarily orthogonal. Techniques which have been
proposed include filterbank multicarrier (FBMC) [7], [9]–
[11] universal filtered multicarrier (UFMC) [18], and gen-
eralized frequency division multiplexing (GFDM) [7]. These
typically involve some form of frequency domain filtering,
or time domain windowing, or some combination of both
filtering and windowing.
B. BACKGROUND ON PCC-OFDM
Polynomial cancellation coded OFDM (PCC-OFDM) is one
form of OFDM which is very well localized in both the time
and frequency domain, and which is compatible with CP-
OFDM and also with MIMO [19], [20]. The first papers on
PCC-OFDM were published around two decades ago [21]–
[24] and showed that PCC-OFDM is robust to frequency
offset [22]. A number of papers quickly followed showing
that PCC-OFDM is also robust to time offset [25], phase
noise [26], and time-varying channels including Rayleigh
fading channels [21], [27], [28], and has a very rapid spectral
roll-off [29]. There has also been some recent work on PCC-
OFDM extending the earlier work on PCC-OFDM in single
user systems [30], [31]. The simplest form of PCC-OFDM
involves mapping data to adjacent pairs of subcarriers. This
results in substantial cancellation of the frequency domain
sidelobes, and a form of windowing in the time domain1 .
Its main disadvantage is that it is not spectrally efficient.
To overcome this, a form of PCC-OFDM using overlapping
symbol periods was developed [32]. This uses a concept
similar to the weighted-overlap-and-add (WOLA) technique
that has recently been proposed for 5G [33]. Interest in PCC-
1 One misconception that has occurred in the past is that PCC-OFDM
is simply a form of repetition coding. While repetition coding would give
a 3 dB gain improvement in an AWGN channel it would not provide the
improved performance in the presence of time and frequency offsets that
PCC-OFDM demonstrates.
2 VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
OFDM reduced as research in MIMO systems exploded in
response to the landmark papers on multiple antenna systems
[34], [35]. However, as we will show the basic form of PCC-
OFDM is ideally suited to the uplink in mMTC and has the
potential to be a key technology for 5G.
A few papers have recently been published on the appli-
cation of PCC-OFDM to 5G. Some theoretical analysis and
simulation results for PCC-OFDM were presented in [36]
and [37], but these did not include the receiver weighting and
adding operation that is required to optimize performance.
We will show that the receiver processing gives substantial
extra benefit. Experimental work using a software defined
radio testbed has confirmed the potential of PCC-OFDM
[38]–[40].
In PCC-OFDM data is mapped onto adjacent subcarriers
using opposite polarities. Since the first publications on PCC-
OFDM [21]–[24] a number of other techniques have been
described which also map data onto pairs of OFDM sub-
carriers. In symmetric cancellation coding, data is mapped
onto two subcarriers symmetrically spaced in the data vector
[41]. Other schemes use a complex conjugate weighting to
map the data to pairs of adjacent or symmetrically spaced
subcarriers [42]. A number of other related schemes have
also been described (see [30], [31] and references therein).
The various techniques have slightly different properties.
Most are less sensitivity to frequency offset than conventional
CP-OFDM and some reduce the sensitivity to phase offsets
or I/Q imbalance. In general these techniques, unlike PCC-
OFDM, do not result in a windowing effect on the transmitted
OFDM symbol. As a result, unlike PCC-OFDM, they are
very sensitive to time offsets and delay spread, unless a cyclic
prefix is used. The advantage of these other techniques is that
in general their PAPR statistics are similar to CP-OFDM and
so they have a lower PAPR than PCC-OFDM, but this is at
the cost of significantly increased sensitivity to time offsets if
a cyclic prefix is not used, or higher energy-per-bit and lower
spectral efficiency if a cyclic prefix is used.
C. CONTRIBUTIONS, NOVELTY AND SIGNIFICANCE
In this paper, we present a study of the performance of PCC-
OFDM in the context of multiuser uplink communications
for 5G. The work is highly significant as it presents a new
technique which clearly outperforms other techniques that
have been proposed for this important emerging 5G appli-
cation. The contributions include:
1) The first comprehensive study of multiuser PCC-
OFDM. This is the first paper which takes into account
both the transmitter and receiver processing. The re-
ceiver weighting-and-adding operation is crucial, and
the only previous work on multiuser PCC-OFDM [36],
[37] does not include this receiver operation. As a re-
sult the earlier work very substantially underestimates
the performance of PCC-OFDM. The contribution of
this new work is significant because it shows that,
when correctly configured, PCC-OFDM substantially
outperforms UFMC and CP-OFDM, which are two
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

xNi  NCP
X 2i k 

..
x0i x0i Filter
X 0i Add Dki 
D/A
X 1i x1i Cyclic x1i

N-IFFT
P/S
Prefix
upconvert  X 2i k 1
…

…
X Ni 1 xNi 1 xNi 1 1

(a) (b)

FIGURE 1. OFDM and PCC OFDM transmitter (a) OFDM transmitter and (b) PCC mapping at transmitter.

Single Tap
Equalizer
Downconvert q Y2ql 2
y0 Y0q
Filter
 

A/D y1q Y1q Z lq

N-FFT

S/P
…

…
Remove CP q Y2ql1 2 
q
y N 1 Y
 N 1

(a) (b)

FIGURE 2. OFDM and PCC-OFDM receiver (a) OFDM receiver and (b) PCC OFDM weighting and adding at receiver.

waveforms which have been considered for the mul-
tiuser uplink in 5G.
2) New analytical and simulation results which
demonstrate the separate contributions of the
transmitter and receiver processing. These results
show the important contribution of receiver processing
to the robustness of PCC-OFDM in the presence of
timing and frequency offsets. This is essential to under-
standing why this work reaches different conclusions
from earlier research [36], [37].
3) The first study of a multiuser OFDMA system using
PCC-OFDM that considers frequency offsets, and
the combination of timing and frequency offsets.
This is an important factor in an mMTC environment,
where strict synchronization is not practicable. (The
earlier papers [36], [37] considered only timing offset,
not frequency offset, or the combination of timing and
frequency offset).
4) Detailed simulation results for the performance of
PCC-OFDM in a multiuser context. These include
results for much larger time and frequency offsets than
have been considered in single-user OFDM. They also
include the effect of interference from other users. The
body of work on single-user OFDM typically considers
only the small offsets that may remain after the receiver
has been synchronized to the received waveform. They
also usually assume that there are no unsynchronized
interfering signals.
VOLUME 4, 2016
5) A clear description of the aspects of earlier PCC-
OFDM research that are relevant to this new work.
D. ORGANIZATION OF PAPER
The remainder of this paper is organized as follows. Section
II describes the multiuser systems and the CP-OFDM, PCC-
OFDM and UFMC transmitters and receivers that are studied
in this paper. In Section III, the time and frequency domain
properties of the three waveforms are described, and the ICI
caused by time and frequency offsets is analyzed. Detailed
simulations of the three waveforms in single user systems
in the presence of time and frequency offsets are presented
in Section IV. In Section V the simulations are extended
to consider the multiuser case. The many advantages and
the possible disadvantages of PCC-OFDM are discussed in
Section VI, and finally conclusions are drawn in Section VI.
II. MULTIUSER OFDM SYSTEMS
In this section we describe the transmitter and receiver struc-
tures for the three forms of OFDM that we analyze and their
application in a multiuser system.
Fig. 1 shows the key elements of an OFDM transmitter,
and the mapping function required for PCC-OFDM. In the i-
th symbol period the data to be transmitted is mapped onto
a complex vector Xi = [X0i , X1i · · · Xki · · · XN i
−1 ] which
is input to an N -point-IFFT. The corresponding vector at
the IFFT is xi = [xi0 , xi1 · · · xim · · · xiN −1 ] where from the
3

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

Filter
IFFT P/S

...

...
Length L
Subband
Other subbands  D/A

...
Mapper Upconvert
Filter
IFFT P/S

...

...
Length L

(a)

Downconvert 

…
Filter
A/D, S/P 2N  Subband
point
Demapper

…
FFT
Zero
…

padding 
Single Tap
Equalizer
(b)

FIGURE 3. UFMC Transmitter and Receiver.

definition of the IFFT Because the noise in the two subcarriers is independent,
N −1   weighing and adding the two subcarriers in the receiver im-
1 X i j2πkm proves the signal-to-noise ratio (SNR) by 3 dB (see Appendix
xim = √ Xk exp . (1)
N k=0 N A). We will show that it also further reduces the overall
sensitivity of PCC-OFDM to frequency and time offset. In
Usually a cyclic prefix is appended to xi before parallel-to- the simulations we will show that no cyclic prefix is required
serial conversion, digital-to-analog conversion, filtering, and in PCC-OFDM in this application, but one can be used if
upconversion to a radio frequency carrier. The corresponding necessary, for example if the same equipment is used for both
OFDM receiver and PCC weighting-and-adding block are CP-OFDM and PCC-OFDM.
shown in Fig. 2, where the q-th input  to the receiver FFT In this paper, we consider the uplink of a multi-user OFDM
is vector yq = y0q , y1q · · · ynq · · · yN
q
 −1 and the output vector based system, where multi-user access is achieved by using
is Yq = Y0q , Y1q · · · Ylq · · · YNq −1 where frequency-division-multiplexing with different users being
N −1   allocated different subbands, and where each subband is a
1 X q −j2πln
Ylq = √ yn exp . (2) group of adjacent OFDM subcarriers. Groups of 12 subcarri-
N n=0 N ers are often considered in the literature, and this is what is
In many applications some subcarriers, for example the used in our simulations. Each user is allocated one or more
band-edge subcarriers, are unused and the corresponding subbands. To limit interuser interference, guard bands of
inputs are set to zero. The only difference between a PCC- unused subcarriers may be inserted between the subcarriers
OFDM transmitter and a simple CP-OFDM transmitter allocated to each user. In this paper we consider two cases:
shown is the mapping of data onto adjacent pairs of sub- systems with guard bands of 12 subcarriers, and systems with
carriers2 . Each input Dki 0 is mapped onto adjacent pairs of no guard bands.
i
subcarriers , so that X2k i i i
0 = Dk 0 and X2k 0 +1 = −Dk 0 . In a
The system description for UFMC is more complicated
PCC-OFDM receiver, after equalization, the two subcarriers than for CP-OFDM or PCC-OFDM as a separate IFFT and
in each pair are combined as shown in Fig. 2(b) to give a separate time-domain filter are required for each subband.
Fig. 3 shows a UFMC transmitter and receiver. Each trans-
mitter will in general require multiple N -point IFFTs: one
(Y2li 0 − Y2li 0 +1 ) IFFT for each subband. To reduce the spectral leakage into
Zli0 = . (3)
2 other subbands, each subband is separately filtered before the
2 To simplify the discussions in the paper, we consider the case of mapping
filtered signals representing each subband are combined and
onto pairs of subcarriers where the lower index is even. This is not necessary. transmitted. At the receiver the signal is zero-padded before
Any adjacent pairs of subcarriers will provide the same performance gains. input to a 2N -point FFT. It can be shown that the wanted
4 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

User 1
CP/PCC/
Subband UFMC
Channel

...
Mapper OFDM
Transmitter

CP/PCC/
Subband


...
Other users UFMC
De-

...
OFDM
mapper
Receiver
User d
AWGN
CP/PCC/
Subband UFMC
Channel
...

Mapper OFDM
Transmitter

FIGURE 4. Configuration of multi-user uplink.

(a) (b) (c)

FIGURE 5. Subband spectra for two users for (a) CP-OFDM, (b) PCC-OFDM and (c) UFMC for N = 256, NCP = 32 and L = 33.

signals appear on alternate outputs of the IFFT [18]. These
outputs are then equalized before sub-band demapping. A
major disadvantage of UFMC is the requirement for multiple
FFTs and filters. These increase both the complexity and
latency of UFMC systems.
Fig. 4 shows the block diagram that describes the multi-
user configuration for the three waveforms that we analyze.
A number of different transmitters transmit on different sub-
bands. In this paper we consider the case where all users
within a system use the same waveform (i.e. CP-OFDM,
PCC-OFDM or UFMC). At each transmitter a subband
mapper maps the data onto the subbands allocated to that
user. The received signal is the sum of the received signals
from each of the users plus additive white Gaussian noise
(AWGN). The receiver demodulates the signal and then,
based on their allocated subbands, the subband demapper
separates the data transmitted by each user.
VOLUME 4, 2016
III. TIME AND FREQUENCY DOMAIN PROPERTIES OF
CP-OFDM, PCC-OFDM AND UFMC
The relative performances of CP-OFDM, PCC-OFDM and
UCMC in a multiuser system depend very much on the time
and frequency domain properties of the three waveforms, and
the level of ICI that any time or frequency offset causes.
In this section we describe these properties and the ICI that
results from time and frequency domain offsets.
A. SPECTRA OF CP-OFDM, PCC-OFDM AND UFMC
Fig. 5 shows the spectra of two 12-subcarrier subbands
separated by a guard band of 12 subcarriers for systems with
IFFT size, N = 256, for the three waveforms we consider.
Fig. 6 shows the band-edge of the spectra in more detail.
Fig. 5(a) shows the case of conventional CP-OFDM with
CP length, NCP = 32. The ripple in the pass band of CP-
OFDM and the absence of deep nulls in the OOB are due
to the CP [12], [43]. The corresponding spectrum for PCC-
OFDM (with no CP) is shown in Fig. 5(b). The spectral
roll-off is much more rapid for PCC-OFDM. This rapid roll-
5

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

off is an important advantage in subband multiplexing, as it

means that there is much less interference between users if
the frequencies are not precisely aligned. Fig. 5(c) shows the
spectra for UFMC for a Dolph-Chebyshev filter of length,
L = 33. Comparing PCC-OFDM and UFMC it can be seen
that the OOB power of PCC-OFDM initially falls off more
quickly than that of UFMC, but is higher for frequencies
further from the subband. It can be seen that for CP-OFDM, if
timing or frequency offsets disrupt the strict orthogonality be-
tween subcarriers, there is potentially significant interference
between subbands. While there is more overlap for PCC-
OFDM than UFMC, the overall performance also depends on
the receiver processing. The combining of adjacent subcarri-
ers in PCC-OFDM results in further interference reduction
FIGURE 6. Band-edge detail of spectra for CP-OFDM, PCC-OFDM and
so that, as will be shown in Section IV, in the overall system UFMC, for N = 256, NCP = 32 and L = 33.
PCC-OFDM outperforms UFMC. The spectrum of UFMC
depends on the length of the filter used. A longer filter will
result in more rapid spectral roll-off, at the cost of some loss
in overall spectral efficiency and increase in signal processing
complexity.
The rapid roll-off of PCC-OFDM can be understood by
comparing the power spectral density (PSD) of an individual
OFDM subcarrier with that of a PCC-OFDM subcarrier pair.
It is well known that the Fourier transform on an OFDM
subcarrier has the form of a sinc function. In [29] it is shown n
both mathematically and graphically (see [29] Fig. 2) that the (a)
sidelobes of adjacent subcarriers are very similar and so the
[+1, -1] PCC mapping results in substantial cancellation of
the sidelobes. This was then used to show that, subject to a
number of assumptions and conditions, the PSD of an OFDM
subcarrier falls off with frequency, f , as 1/(f 2 N 2 ), while for
N  1, the PSD of a subcarrier pair in PCC-OFDM falls off
as 1/(f 4 N 4 ). For OFDM, it was assumed that the complex
data mapped onto the OFDM subcarriers comprised indepen-
dent zero-mean random variables [29], and that there was no n
CP. Rectangular pulse shaping and perfect interpolation were (b)
also implicitly assumed. For PCC-OFDM the complex data
mapped to the PCC-OFDM subcarrier pairs was assumed to
comprise independent zero-mean random variables.
The overall spectrum of an OFDM signal has been shown
to depend on a number of aspects of the transmitter de-
sign, including N , NCP , the number of unused band-edge
subcarriers, the pulse shaping, and the interpolation filtering
[43]. (Note: the comments about OOB power in [29] are n
inaccurate). (c)

FIGURE 7. Time domain envelopes for (a) CP-OFDM, (b) PCC-OFDM and (c)
UFMC for N = 256, NCP = 32 and L = 33.
B. TIME-DOMAIN ENVELOPES OF CP-OFDM,
PCC-OFDM AND UFMC

Fig. 7 shows the time domain envelopes for the three wave- is given by
forms we consider. CP-OFDM has the familiar square win-
dow and has a symbol length of N + NCP = 288. The N −1  
formula for the PCC-OFDM envelope shown in Fig. 7(b) can 1 X i j2πkm
xim = √ Xk exp (4)
be calculated as follows. From the definition of the FFT, xm N k=0 N
6 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

where the component of xim due to the input Xki is yq = xi and Yq = Xi . Now consider the effect of a time
  offset of p samples where p ≥ 1, and for NCP = 0 , so
i 1 i j2πkm
xm,k = √ Xk exp yi = y0i , y1i , y2i · · · yN
i
 
. (5) −1
N N (8)
= xip · · · xiN −1 , xi+1 · · · xi+1
 
0 p−1 .
So the component of xim due to the pair of PCC coded inputs,
It can be seen that yi is a function of both Xi and Xi+1 .
Xki and Xk+1
i i
, where Xk+1 = −Xki is
Both Xi and Xi+1 cause ICI, but Xi+1 also causes ISI.
Consider first the components of yi and Yi due only to
xim,k + xim,k+1
     Xi , which i,i i,i
 i we i denote by y , and Y respectively, so
1 i j2πkm i j2π(k + 1)m i,i
y = xp · · · xN −1 , 0 · · · 0 . In Appendix B, it is shown that
=√ Xk exp − Xk exp
N N N  i,i  N −1−p
    Y   
1 j2πkm j2πm  l,k  1  X j2πn (k − l) 
=√ Xki exp 1 − exp  i = exp (9)
N N  Xk  N  N
 
N n=0

(6)
i,i
where Yl,k is the component of Yli,i due to Xki . Equation (9)
which is a complex exponential of frequency, k, multiplied
 
 i,i i 
by a complex window (1 − exp (j2πm/N )). It is important
to note that the window function is independent of k, so is
the envelope of every PCC pair and hence of the entire PCC-
OFDM symbol. The magnitude of the window function is
shows that the magnitude of a component of ICI, Yl,k /Xk ,
depends on the time offset, p, and (k − l), the difference
between the subcarrier indices. The magnitude of the ICI
does not depend
 on the separate indices, k and l directly. Fig.
 i,i i 
given by
p offset, p/N , and (k − l). For p = 0 and k = l, there is
|(1 − exp (j2πm/N ))| = 2 (1 − cos (2πm/N )). (7)
For PCC, as no CP is used, the symbol length is equal to
the FFT size, which in this case is N = 256. For UFMC
the envelope depends on the filter length. Fig. 7(c) shows the
result for N = 256 and L = 33, which results in an overall
symbol length of N + L − 1 = 288. The filtering in UFMC
results in an envelope which tapers at the start and end of
the symbol, and this reduces the sensitivity of UFMC to time
offsets.
8(a) shows Yl,k /Xk  for N = 256 as a function of time
no ICI: all of the energy transmitted on a given subcarrier
is received on the same subcarrier. When the time offset
increases, the ICI increases, and there is substantial ICI even
in quite distant subcarriers.
2) ICI in PCC-OFDM with transmitter weighting only
When the data at the IFFT input is PCC encoded, then
Xk+1 = −Xk , and the l-th output of the receiver FFT is
given by the sum of the components due to Xk and Xk+1 . In
Appendix C, it is shown that
 i,i  N −1−p
 Y + Y i,i 
  
C. ANALYSIS OF INTERCARRIER INTERFERENCE  l,k l,k+1  1  X j2πn (k − l)
= exp
Xki  N N
  
CAUSED BY TIME OFFSET 
n=0
In this section the effect of timing offset on OFDM and PCC-     (10)
j2π (n + p)
× 1 − exp .

OFDM is analyzed in detail, and the importance of receiver N 
processing in PCC-OFDM is explained. This extends the
analysis in [29] to include the derivation of expressions for Like the result for conventional OFDM given in (9), the mag-
the PCC-OFDM signals before and after receiver weighting nitude of a component of ICI depends only on the time offset,
and adding, and for intercarrier as well as intersymbol inter- p, and (k −l),
 and does notdepend  on k or l separately. Fig.
ference.  i,i i,i i 
8(b) shows  Yl,k + Yl,k+1 /Xk  for N = 256. Again, like
Fig. 8(a), for p = 0 and k = l there is no ICI: all of the
1) ICI in OFDM energy transmitted on a given subcarrier is received on the
For CP-OFDM, as long as the there is no frequency same subcarrier, however because the data is transmitted on
offset, and any timing offset is less than the length of both the k-th and the (k + 1)-th subcarriers, there are two
the CP, the received subcarriers are orthogonal, and can non-zero values for p = 0. When the time offset increases,
be recovered without any ICI or intersymbol interfer- the ICI increases, but for all offsets, p, the ICI is less than in
ence (ISI). However when this condition is not satisfied Fig. 8(a).
the performance degrades rapidly. To understand the ef-
fect of timing offset consider  i i ai sequence  of twoi+1IFFT 3) ICI in PCC-OFDM with both transmitter and receiver
i i
output vectors
 i+1 i+1 i+1 x = x , x
0 1 , x 2 · · · x N −1 and x = processing
i+1 q
x ,
 q0 q 1qx , x 2 · · · x
 N −1 and one FFT input vector y = We now analyze the effect of time offset on the overall PCC-
q
y0 , y1 , y2 · · · yN −1 . If the receiver is perfectly synchro- OFDM system including both the transmitter mapping, and
nized, the channel is flat, there is no noise, and q = i, then the receiver weighting-and-adding operation. In this case the
VOLUME 4, 2016 7

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

(a) (b) (c)

FIGURE 8. ICI and ISCHI as a function of time offset for N = 256 (a) CP-OFDM, (b) PCC-OFDM no receiver weighting and adding and (c) PCC-OFDM with
receiver weighting and adding.

(a) (b) (c)

FIGURE 9. ICI and ISCHI due to ISI as a function of time offset for N = 256 (a) CP-OFDM, (b) PCC-OFDM no receiver weighting and adding and (c) PCC-OFDM
with receiver weighting and adding.

(a) (b) (c)

FIGURE 10. ICI and ISCHI as a function of frequency offset for N = 256 (a) CP-OFDM, (b) PCC-OFDM no receiver weighting and adding and (c) PCC-OFDM
with receiver weighting and adding.

8 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

performance depends on how Dki 0 varies as a function of Zli0 ,
and as the time offset is varied. Denoting the components of
Zi due to Di by Zi,i , and the component of Zli,i 0 due to Dk 0
i
i,i
as Zl0 ,k0 , and using Zl0 = (Y2l0 − Y2l0 +1 )/2 and Dk0 =
X2k0 = −X2k0 −1 , it is shown in Appendix D that
From (16) it can be seen that the magnitude of ICI due
to frequency offset depends only on (k − l), the difference
between the subcarrier indices and ∆f T , where ∆f T is the
frequency offset normalized to the subcarrier spacing. For
PCC-OFDM considering only the transmitter weighting, it
 i,i   N −1−p  is shown in Appendix F that
Z 0 0  1  X

j2πn

 l ,k   i,i
 Y + Y i,i 
  N −1
 i = 1 − exp 1  X

j2πn (k − l + ∆f T )

 Dk0  2N  N
  l,k l,k+1 
n=0 =  exp
Xki  N N

  
n=0
j4πn (k 0 − l0 )
   
j2π (n + p)  
× exp 1 − exp .
  
N N j2πn
× 1 − exp .
 
(11) N 
(17)
It can be seen from (11) that the interference for PCC after
both receiver and transmitter processing, depends only on p For PCC-OFDM when the receiver weighting-and-adding
0 0 0 0
 (k − l ) and not on k or l separately. Fig. 8(c) shows
and step is included
 i,i i
Zl0 ,k0 /D2k 0  for N = 256 as a function of time offset
  i,i   N −1 2
Z 0 0  1  X

j2πn
 l ,k 
p/N and (k − l0 ). Comparing Fig. 8(c) with Fig. 8(b) it can
0  i = 1 − exp
 X2k0  2N  N

be seen that receiver weighting-and-adding step reduces the n=0

j2πn (2k 0 − 2l0 + ∆f T ) 

interference considerably.
× exp .
N 
4) ICI due to intersymbol interference (18)
Expressions for the components of interference caused by
intersymbol interference from the i + 1 th symbol can be Fig. 10 shows the ICI as a function of frequency offset for
derived in a similar way. In this case CP-OFDM and PCC-OFDM with and without weighting
assuming θ0 = 0 . The normalized frequency offset is varied
yi,i+1 = 0 · · · 0, xi+1 · · · xi+1 for 0 ≤ ∆f T ≤ 2, which corresponds to two subcarrier
 
0 p−1 (12)
spacings. For CP-OFDM it can be seen in Fig 10(a) that, as
and it can be shown in Appendix E that for OFDM ∆f T varies from 0 to 1, the ICI gradually shifts from one
 i,i+1  
 N −1
 subcarrier to the next, and that there is significant interference
Y  1  X
 
j2πn (k − l)  across a number of subcarriers. Fig. 10(b) shows the ICI as
 l,k 
 i+1  = exp . (13) a function of frequency offset for PCC-OFDM without the
 Xk  N  N
n=N −p 
Similarly for PCC-OFDM encoding at the transmitter only,
the ICI is given by (14). Performing a similar analysis for
PCC-OFDM with both transmitter mapping and receiver
weighting-and-adding, and denoting the component of Zli,i
due to Dki+1
0 as Zli,i+1
0 ,k 0 , the interference is given by (15).
receiver weighting-and-adding step, while Fig 10(c) shows
the ICI in PCC-OFDM with receiver weighting-and-adding.
It can be seen that, as for time offset, both the transmitter
mapping and the receiver weighting-and-adding contribute to
0
the reduction in interference in PCC-OFDM.

These results are shown in Fig. 9. It can be seen that as the
time offset increases, the ICI due to ISI also increases.
D. ANALYSIS OF INTERCARRIER INTERFERENCE
CAUSED BY FREQUENCY OFFSET
The ICI due to frequency offset in PCC-OFDM was analyzed
in [22]. In this section we present new results which demon-
strate the separate contributions of transmitter and receiver
processing in PCC-OFDM. A detailed analysis is given in
Appendix F. For the case where the difference in frequency
between the frequency of the receiver local oscillator and the
carrier frequency of the received signal is ∆f , and the phase
offset at the start of the i th symbol is θ0 [22], and assuming
no timing offset, a flat channel and no noise, then it can be
shown that
IV. PERFORMANCE OF CP-OFDM, PCC-OFDM AND
UFMC - SINGLE-USER CASE
We now show how the different properties of the three wave-
forms affect their sensitivity to timing and frequency offsets
and to noise in a single-user scenario, and demonstrate the
importance of the receiver weighting-and-adding operation
in the PCC-OFDM receiver.
A. PERFORMANCE IN AN AWGN CHANNEL
We first consider the case of single user in an additive white
Gaussian noise (AWGN) channel with no timing or fre-
quency offset between transmitter and receiver. Fig. 11 shows
BER results for the three waveforms as a function of Eb /N0
where Eb is the energy per bit and N0 the single-sided noise
spectral density. Results are given for 4-QAM, 16-QAM and
 i,i  N −1  64-QAM constellations. For each constellation size, PCC-
Y  1  X

j2πn (k − l + ∆f T ) 
 l,k  OFDM requires the lowest Eb /N0 of the three waveforms.
 i = exp  . (16)
 Xk  N  N This is because at the PCC-OFDM receiver the noise in


n=0
VOLUME 4, 2016 9

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

 i,i+1  
i,i+1 
  N −1
Y
 l,k + Yl,k+1 
   
1  X j2πn (k − l) j2π (−N + p + n) 
i+1 = exp 1 − exp  (14)
N N N

 Xk  
n=N −p



 i,i+1   
Z 0 0   N −1     0 0
  
 l ,k  1  X j2πn j2πn (2k − 2l ) j2π (−N + p + n) 
 i+1  =  1 − exp exp 1 − exp  (15)
 Dk  2N  N N N 
n=N −p 

different subcarriers is independent, so combining subcarrier

pairs at the receiver reduces the required Eb /N0 by 3 dB. The
disadvantage of PCC-OFDM is that the mapping of data onto
pairs of subcarriers reduces the spectral efficiency, so that for
a given constellation size PCC-OFDM carries only slightly
more than half the data of CP-OFDM. This means that to
compare systems of approximately equal spectral efficiency
the plots for 16-QAM PCC-OFDM should be compared with
the plots for 4-QAM CP-OFDM and UFMC in Fig. 11, Fig.
13 and Figs. 15-29. For CP-OFDM, Eb depends on the length
of the cyclic prefix. The results are for the case of N = 256
and NCP = 32. Similarly for FBMC, Eb depends on the
filter length and the results in Fig. 11 are for N = 256
and L = 33. The results for FBMC and CP-OFDM are
very similar because L = NCP + 1, so the overhead is
the same for each system. As expected, for each type of FIGURE 11. BER in an AWGN as a function of Eb /N0 for CP-OFDM,
PCC-OFDM and UFMC with 4-QAM, 16-QAM and 64-QAM constellations.
waveform the Eb /N0 required for a given BER increases as
the constellation size increases.
Fig. 12 demonstrates the importance of the weighting-and-
adding operation in a PCC-OFDM receiver. It shows the BER
performance for PCC-OFDM without weighting and adding,
that is the result if data estimation is based on the received
signal on only one of the two PCC subcarriers, and also with
weighting and adding of the subcarrier pair. For each size
of constellation it can be seen that the weighting-and-adding
operation reduces the required Eb /N0 by 3 dB.

B. PERFORMANCE WITH TIMING OFFSET

We now consider the effect of time offsets between the
transmitter and receiver for the single user case. Fig. 13
shows the BER in AWGN for a receiver offset of τ = 0.05
where τ is measured as a fraction of the OFDM symbol
period excluding the CP, and positive τ represents a delay
in the receiver timing relative to the transmitter. As N = 256
this represents a delay of 13 samples. Comparing Fig. 13 with
Fig. 11 it can be seen that this offset does not change the BER
for CP-OFDM. This is expected as the offset is less than the
CP length of 32 samples. In contrast the time offset increases
the BER for UFMC. This is most clearly seen from the 64-
QAM results where UFMC now has a higher BER than CP-
OFDM. This is because the receiver window is not aligned
with the maximum of the time domain envelope shown in
Fig. 7. For PCC-OFDM there is a very slight increase in BER
because the first part of each symbol is missed, so not all of
the received symbol energy is used to recover the data, but
10
FIGURE 12. BER in an AWGN as a function of Eb /N0 for PCC-OFDM with
and without weighting and adding in the receiver for 4-QAM, 16-QAM and
64-QAM constellations.
this effect is less than for UFMC as the time domain envelope
does not change so quickly. This result is consistent with the
effect of timing offset shown in Fig. 8 (c).
Fig. 14 again shows the importance of the weighting-and-
adding operation in the PCC-OFDM receiver. It shows that it
reduces the sensitivity to timing offset as well as providing a
3 dB improvement in SNR, it gives an improvement of up to
a 6 dB for 64-QAM with a timing offset of τ = 0.05.
Fig. 15 shows the effect of increasing τ to 0.2 which is
VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access
G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications
FIGURE 13. BER in an AWGN as a function of Eb /N0 for with time offset
τ = 0.05 for CP-OFDM, PCC-OFDM and UFMC with 4-QAM, 16-QAM and
64-QAM constellation.
FIGURE 14. BER in an AWGN as a function of Eb /N0 for with time offset
τ = 0.05 for PCC-OFDM with and without weighting and adding in the
receiver for 4-QAM, 16-QAM and 64-QAM constellations.
FIGURE 15. BER in an AWGN as a function of Eb /N0 for with time offset
τ = 0.2 for PCC-OFDM with and without weighting and adding in the receiver
for 4-QAM, 16-QAM and 64-QAM constellations.
VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
FIGURE 16. Eb /N0 required for waveforms to achieve a target BER of 10−2
as a function of normalized time offset τ .
greater than the length of the CP. For UFMC and CP-OFDM
the BER plateaus at levels about 10−2 whereas the BER
for PCC-OFDM is much lower, with 4-QAM and 16-QAM
showing relatively little degradation. We now explore the
effect of varying the timing offset. Fig. 16 shows how the
performance of each waveform varies as a function of time
offset, τ . It compares the required Eb /N0 for CP-OFDM,
PCC-OFDM and UFMC for a target BER of 10−2 for 4-
QAM and 16-QAM. For CP-OFDM the BER is constant
as long as the delay is within the cyclic prefix length, but
increases rapidly for τ < 0, or τ > 0.125. The plots
for CP-OFDM are not symmetric about τ = 0 because
the simulations are based on the use of a cyclic prefix (not
postfix), so for timing errors within the length of the CP
there is no ICI, or ISI. In contrast PCC-OFDM is much
less sensitive to time offset and degrades only slowly. Even
offsets of τ = ±0.2 require an increase of less than 2 dB
to maintain a BER of 10−2 . The performance of UFMC is
relatively constant over the smaller range −0.08 < τ < 0.08
but degrades rapidly for larger time offsets.
C. PERFORMANCE WITH FREQUENCY OFFSET -
SINGLE USER CASE
Frequency offsets between transmitter and receiver will also
degrade the BER performance. Fig. 17 shows the effect of
a normalized frequency offset, ∆f T = 0.05. Comparing
Fig. 17 with the AWGN results shown in Fig. 11, it can
be seen that frequency offset has negligible effect on PCC-
OFDM for all constellation sizes, while the performance of
CP-OFDM and UFMC are significantly degraded, with the
BER plateauing for 64-QAM constellations. The effect of
further increasing ∆f T is demonstrated in Fig. 18 showing
an even greater performance advantage of PCC-OFDM. Fig.
19 shows the effect of varying frequency offsets. Both UFMC
and CP-OFDM degrade rapidly as frequency offset increases,
while even normalized frequency offsets of 0.2 have little
effect on PCC-OFDM.
11
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

FIGURE 17. BER comparison of three waveforms for varying constellation

sizes affected by a normalized frequency offset, ∆f T = 0.05.
FIGURE 19. Eb /N0 required for waveforms to achieve a target BER of 10−2
as a function of normalized frequency offset, ∆f T .

FIGURE 18. BER comparison of three waveforms for varying constellation

sizes affected by a normalized frequency offset, ∆f T = 0.2.

(a)
A point to note is that both for large time offsets and
large frequency offsets 16-QAM PCC-OFDM outperforms 4-
QAM UFMC, and for small offsets has approximately equal
performance. This means that the loss in spectral efficiency of
PCC-OFDM due to the mapping of data onto two subcarriers,
can potentially be regained by using bigger constellations.

V. PERFORMANCE OF CP-OFDM, PCC-OFDM AND

UFMC - TWO-USER CASE
As we are ultimately interested in the application of PCC-
OFDM for uplink multiple access, we now investigate the
sensitivity of the three waveforms to time and frequency
offsets in a two-user scenario. A number of configurations
are possible, but here we calculate the BER for User 1 for
the case where the receiver is synchronized to User 1, so that (b)
there is no timing or frequency offset between User 1 and the FIGURE 20. Two user BER results for normalized timing offset of User 2,
receiver, but User 2 may have a timing or frequency offset. τ = 0.05 (a) guard band of 12 subcarriers, (b) no guard band.
Whereas in the previous section the ICI is ‘same user ICI’:
the ICI was caused by subcarriers allocated to one user, in
12 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access
G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications
(a)
(b)
FIGURE 21. Two user BER results for normalized timing offset of User 2,
τ = 0.05. User 2 received power 10 dB greater than User 1 (a) guard band of
12 subcarriers, (b) no guard band.
this section the impairment may also be due to ‘other user
ICI’, which is the ICI caused by subcarriers allocated to a
different user. Similarly for ISI. As can be seen in Fig. 8,
the level of ICI is strongly dependent on the spacing between
the subcarriers concerned, so we consider both cases, that is
where there is a guard band between the subcarriers allocated
to each user and cases where there is no guard band.
A. PERFORMANCE OF TWO-USER SYSTEM IN AWGN
CHANNEL
Simulations were performed for two users for 12 subcarrier
subbands with a 12 subcarrier guard band and an AWGN
channel. The BER results were identical to the single user
AWGN case shown in Fig. 11, and this was also the case
when there was no guard band. This was the expected result
for CP-OFDM and PCC-OFDM because in the absence of
time or frequency offset the subcarriers are strictly orthog-
onal. However there was also no observable degradation
for UFMC. In general the powers of the signals received
from different users will not be equal, to explore the effect
of a strong interfering signal, the received power of User
2 was increased by 10 dB. There was still no observable
VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
(a)
(b)
FIGURE 22. Two user BER results for normalized timing offset of User 2,
τ = 0.2 (a) guard band of 12 subcarriers, (b) no guard band.
degradation: the BER plots were the same as Fig. 11.
B. PERFORMANCE OF TWO-USER SYSTEM WITH
TIMING OFFSET
Fig. 20 shows the BER performance when User 2 has a time
offset of τ = 0.05. For a 12 subcarrier guard band comparing
Fig. 20(a) with Fig. 11 it can be seen that there is a very
slight degradation in the performance of 64-QAM UFMC.
Removing the guard band (Fig. 20 (b)) causes no observable
increase in degradation of UFMC. Fig. 21 shows the effect
of increasing the received power of User 2 by 10 dB. The
key point is that even with no guard band and a higher
power interfering signal there is no observable change in the
performance of PCC-OFDM. There is also no change for CP-
OFDM as this time offset is within the cyclic prefix. However
the performance of 64-QAM UFMC degrades significantly.
A slightly surprising result is that for this case the no guard
band 64-QAM UFMC has better performance than 64-QAM
UFMC with a guard band. This is because with no synchro-
nization errors the ICI in UFMC falls on the odd subcarriers
but timing offsets disrupt this. The effect of increasing τ to
13
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

(a)
(a)

(b)

FIGURE 24. Two user BER results for normalized frequency offset of User 2,
(b) ∆f T = 0.05 (a) guard band of 12 subcarriers, (b) no guard band.

FIGURE 23. Two user BER results for normalized timing offset of User 2,
τ = 0.2. User 2 received power 10 dB greater than User 1 (a) guard band of
12 subcarriers, (b) no guard band. does PCC-OFDM show any degradation, and this is only
observable for the largest constellation.

0.2 is shown in Fig. 22 and Fig. 23. In all cases PCC-OFDM
substantially outperforms UFMC and CP-OFDM.
C. PERFORMANCE OF TWO-USER SYSTEM WITH
FREQUENCY OFFSET
Fig. 24 shows the BER performance when User 2 has a
frequency offset of ∆f T = 0.05. For a 12 subcarrier guard
band comparing Fig. 24 (a) with Fig. 11 it can be seen
that there is a very slight degradation in the performance of
64-QAM CP-OFDM. With no guard band (Fig. 24 (b)) the
performance of 64-QAM CP-OFDM degrades further and
64-QAM UFMC also shows some change. Fig. 25 shows the
effect of increasing the received power of User 2 by 10 dB.
The key point is that even with no guard band and a higher
power interfering signal there is no observable change in
the performance of PCC-OFDM. Fig. 26 shows the effect of
increasing the frequency offset. Now even with a guard band
(Fig. 26 (a)) the frequency offset degrades the performance
of CP-OFDM and UFMC. Fig. 27 shows that only for large
frequency offset and for a high power interfering signal
14
D. PERFORMANCE OF TWO-USER SYSTEM WITH TIME
AND FREQUENCY OFFSET
Finally we investigate the performance when User 2 has
both time and frequency offset. Fig. 28 shows that all three
waveforms perform well for small time and frequency offsets
even with no guard band. Comparing Fig. 24, which shows
the results for ∆f T = 0.05, τ = 0.05, with Fig. 11 it can
be seen that only 64-QAM CP-OFDM and 64-QAM UFMC
show any change. Similarly, Fig. 29 shows that even for
∆f T = 0.2, τ = 0.05 and with no guard band PCC-OFDM
shows no degradation.
VI. ADVANTAGES AND DISADVANTAGES OF
PCC-OFDM FOR 5G
In addition to the advantages of extreme insensitivity to time
and frequency described above, because PCC-OFDM is a
form of OFDM, it also retains many of the advantages of
CP-OFDM. For example, single tap equalizers can be used
in PCC-OFDM to correct for frequency selective fading,
and PCC-OFDM, like CP-OFDM is well suited to use in
VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access
G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications
(a)
(b)
FIGURE 25. Two user BER results for normalized frequency offset of User 2,
∆f T = 0.05, User 2 received power 10 dB greater than User 1 (a) guard
band of 12 subcarriers, (b) no guard band.
MIMO systems. Although not considered in the simulations
in this paper, PCC-OFDM has also been shown to be very
insensitive to Doppler spread [21]. The additional mapping
onto subcarriers at the transmitter and the weighting-and-
adding step at the receiver mean that PCC-OFDM adds
negligible complexity and latency relative to CP-OFDM and
has significantly lower complexity and latency than UFMC.
One important potential disadvantage of the form of PCC-
OFDM described in this paper is the potential loss in spectral
efficiency resulting from the mapping of data onto pairs of
subcarriers rather than onto single subcarriers. It is likely that
much of this loss can be recovered due to a combination of
the other properties of PCC-OFDM, including the fact that
no frequency guard band is required between users, and that
the improved performance may allow larger constellations to
be used. For low data rate applications a small loss in spectral
efficiency may be acceptable. For high data rate applications,
particularly in the downlink, PCC-OFDM with overlapped
symbol periods may offer a good solution [32].
A second potential disadvantage of PCC-OFDM is its
high PAPR, which is higher than CP-OFDM [44]. An in-
teresting area of future research would be to investigate
VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
(a)
(b)
FIGURE 26. Two user BER results for normalized frequency offset of User 2,
∆f T = 0.2 (a) guard band of 12 subcarriers, (b) no guard band.
PAPR mitigation schemes for PCC-OFDM. An alternative
solution might be to use no PAPR reduction. High PAPR is a
problem in many OFDM applications particularly high power
broadcast television and radio applications because, unless
linear amplifiers are used in the transmitters, the peaks result
in a distorted signal which contains out-of-band spectral
components which may interfere with other signals. The
disadvantage of linear amplifiers is that they are not power
efficient. However recent research [45] has shown that when
the power consumption of the digital signal processing is
considered there is no overall power advantage in applying
PAPR reduction techniques for low power applications. The
overall system advantages or disadvantage of PCC-OFDM
compared with other 5G solutions when all of the properties
of insensitivity to time and frequency offset and the conse-
quent reduction in signal processing required are considered
will be an interesting area of future research.
VII. CONCLUSIONS
This paper has evaluated PCC-OFDM as a contender for 5G
systems and compared it with well-known waveforms such as
CP-OFDM and UFMC. It has been shown that PCC-OFDM
performs substantially better than CP-OFDM and UFMC in
15
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

(a) (a)

(b) (b)

FIGURE 27. Two user BER results for normalized frequency offset of User 2, FIGURE 29. Two user BER results for normalized frequency offset of User 2,
∆f T = 0.2, User 2 received power 10 dB greater than User 1 (a) guard band ∆f T = 0.2, and normalize time offset of User 2, τ = 0.05. (a) guard band of
of 12 subcarriers, (b) no guard band. 12 subcarriers, (b) no guard band.

also require an increase of only 2 dB. It has been shown that

FIGURE 28. Two user BER results for normalized frequency offset of User 2,
∆f T = 0.05, and normalize time offset of User 2, τ = 0.05. No guard band.
the presence of time and/or frequency offsets. Large offsets
that would make transmission with CP-OFDM or UFMC
impossible, cause very little degradation in PCC-OFDM. For
example time offsets of ±0.2 of a symbol period require
only a 2 dB increase in SNR to achieve a given target BER.
Similarly frequency offsets of ±0.2 of a subcarrier spacing
16
the weighting-and-adding step in the PCC-OFDM receiver
contributes significantly to the performance. Because of the
mapping of data onto two subcarriers, weighting and adding
improves the performance in AWGN by 3 dB but it improves
the performance by much more than this in the presence
of time and frequency offsets. The performance of PCC-
OFDM in a multiple access system using OFDMA has been
evaluated. It has been shown that because PCC-OFDM is
very well-localized in both the time and frequency domain
it is extremely robust in a multiuser system and as a result no
guard band is required between users. Only for very large
constellations, very large frequency offsets, and when the
power of the interfering user was 10 dB higher than that
of the user for which the BER was being calculated was
any degradation observable. In conclusion PCC-OFDM is a
very strong contender for application in 5G as it substantially
outperforms other better known waveforms in the presence
of timing and frequency offsets.
VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

APPENDIX. A - ANALYSIS OF CP-OFDM AND
PCC-OFDM IN AN AWGN CHANNEL
For the case of an AWGN channel and no timing or frequency
offset, then for both PCC-OFDM and CP-OFDM
Yki = Xki + Wki
where Wki is the white noise component of the k-th output
of the receiver FFT in the i-th symbol period. Wki are zero
mean identically distributed Gaussian random variables. So
the signal-to-noise ratio (SNR) of Yki is given by
average transmit power, the energy per bit for PCC-OFDM is
twice that for OFDM without a CP, because twice as many
subcarriers are used to transmit a given amount of data. For
CP-OFDM the energy used in the cyclic prefix has to be taken
into account. Both of these factors are included in the results
(19) presented in Section IV.
APPENDIX. B - ANALYSIS OF ICI IN OFDM DUE TO
TIMING OFFSET
To analyze the effect of timing offset inOFDM, we consider
n  o a sequence of two IFFT outputs xi = x i i i i
 0 , x1 , x2 · · · xN −1
E X i 2 i+1 i+1 i+1 i+1 i+1
and x  = x0 , x1 , x2 · · · xN −1 and one FFT input
i

k
SNR Yk = n 2 o . (20) q
yq = y0q , y1q , y2q · · · yN −1 . For q = i, and if the receiver
E W i  k is perfectly synchronized, the channel is flat and there is no
noise then, yq = xi , and Yq = Xi . For an unsynchronized
This is also the SNR for PCC-OFDM before combining. The
system with a time offset of p samples where p ≥ 1, and for
receiver output for PCC-OFDM after combining, and taking
NCP = 0
into account the PCC weighting at the transmitter is given by
yi = y0i , y1i , y2i · · · yN
i
 
i
Yki − Yk+1 −1
(25)
Zki = = xp · · · xiN −1 , xi+1
 i
· · · xi+1

2 0 p−1
Xki + Wki − Xk+1i i
− Wk+1
= (21) which is a function of both Xi and Xi+1 . The components of
2 yi and Yi due only to Xi are given by
1 i
Xk + Xk + Wki + Wk+1
i i



= .
yi,i = xip · · · xiN −1 , 0 · · · 0
 
2 (26)
So the SNR of Zki is given by
n and
2 o
E 2Xki  N −1
1 X i,i

−j2πln

SNR Zki = n Yli,i = √


2 o yn exp
E Wki + Wk+1
i  N n=0 N
n  o (22) NX−1−p   (27)
4E Xki 
2 1 i −j2πln
=√ xn+p exp .
= n 2 o . N n=0 N
E Wki + Wk+1
i 
From (4)
As the noise is white, the noise component on each received
N −1  
subcarrier is independent so 1 X i j2πk (n + p)
xin+p = √ Xk exp . (28)
n 2 o n  o n 2 o N k=0 N
E Wki + Wk+1 i  = E Wki 2 + E Wk+1 i 
n 2 o n 2 o Substituting this in (27) gives
= E Wki  + E Wki 
−1−p
NX N −1  
n 2 o 1 1 X i j2πk (n + p)
= 2E Wki  . Yli,i =√ √ Xk exp
N n=0
N k=0 N
(23)  
−j2πln
Substituting this in (22) gives × exp
n  o N
2 (29)
4E Xki 
i


SNR Zk =
and the component of Yli,i due to Xki is
n  o
2
2E Wki 
n  o
2 N −1−p
(24)
   
2E Xki  i,i 1 i X j2πk (n + p) −j2πln
Yl,k = Xk exp exp
= n  o
2 N N N
n=0
E Wki 
(30)
= 2 × SNR Yki .


which after some rearranging gives
This shows that the weighting and adding in the receiver   NX−1−p  
improves the SNR by a factor of 2, or equivalently 3 dB. i,i 1 i j2πpk j2πn (k − l)
Yl,k = Xk exp exp
The graphs in Section IV use the conventional form of BER N N n=0
N
curves in terms of energy-per-bit rather than SNR. For a given (31)
VOLUME 4, 2016 17

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

so
−1−p
 i,i    NX 
Y   1 
j2πpk

j2πn (k − l) 
 l,k   APPENDIX. C - ANALYSIS OF ICI IN PCC-OFDM DUE TO
 i  =  exp exp
 Xk   N N N

n=0
 TIMING OFFSET WITH TRANSMITTER WEIGHTING ONLY
(32) For PCC-OFDM Xk+1 = −Xk and the l-th output of the
receiver FFT is the sum of the components due to Xk and
Noting that |exp (j2πpk/N )| = 1, this can be simplified to
 i,i  N −1−p Xk+1 , and is given by (34). This can be simplified to give
Y    (35) and (36).
 l,k  1  X j2πn (k − l) 
 i = exp . (33)
 Xk  N  N


n=0

i,i i,i
Yl,k + Yl,k+1
−1−p
 NX −1−p
 NX
(34)
     
1 i j2πpk j2πn (k − l) 1 i j2πp (k + 1) j2πn (k + 1 − l)
= X exp exp − X exp exp
N k N n=0
N N k N n=0
N

 −1−p
 NX    
i,i i,i 1 j2πpk j2πn (k − l) j2π (n + p)
Yl,k + Yl,k+1 = Xki exp exp 1 − exp (35)
N N n=0
N N

 i,i N −1−p
 Y + Y i,i 
    
 l,k l,k+1  1  X j2πn (k − l) j2π (n + p)
= exp 1 − exp (36)

Xki  N N N
  
 
n=0

APPENDIX. D - ANALYSIS OF ICI IN PCC-OFDM DUE TO mance depends on how Dk0 varies as a function of Z l0
TIMING OFFSET WITH BOTH TRANSMITTER AND and as the time offset is varied. Denoting the components
RECEIVER PROCESSING of Zi due Di by Zi,i and the component of Zli,i 0 due to
When both the transmitter mapping, and the receiver Dki 0 as Zli,i
0 ,k 0 , and using Zl0 = (Y2l0 − Y2l0 +1 )/2 and
weighting-and-adding operation are included the perfor- Dk0 = X2k0 = −X2k0 −1 gives

1  i,i  1 
Zli,i
0 ,k 0 = Y2l0 ,2k0 + Y2li,i0 ,2k0 +1 − Y2li,i0 +1,2k0 + Y2li,i0 +1,2k0 +1 (37)
2 2
0
 NX−1−p !
j2πn (2k 0 − 2l0 )
   
1 i j2πp2k j2π (n + p)
= D 0 exp exp 1 − exp (38)
2N k N n=0
N N
NX−1−p !
j2πp2k 0 j2πn (2k 0 − 2l0 + 1)
     
1 i j2π (n + p)
− D exp exp 1 − exp (39)
2N k N n=0
N N

which after some rearrangement gives

−1−p 
 NX
j4πpk 0 j4πn (k 0 − l0 )
      
1 i j2πn j2π (n + p)
Zli,i
0 ,k 0 = D 0 exp 1 − exp exp 1 − exp (40)
2N k N n=0
N N N

18 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

and so

 i,i  N −1−p  
Z 0 0  j4πn (k 0 − l0 )
    
 l ,k  1  X j2πn j2π (n + p)
 i = 1 − exp exp 1 − exp . (41)

 Dk0  2N  N N N


n=0

APPENDIX. E - ANALYSIS OF ICI DUE TO derived in a similar way. In this case

INTERSYMBOL INTERFERENCE
yi,i+1 = 0 · · · 0, xi+1 · · · xi+1
 
0 p−1 (42)
Expressions for the components of interference caused by
intersymbol interference from the i + 1 th symbol can be and it can be shown that for OFDM

 −1
 NX  
i,i+1 1 i+1 − (N − p) k j2πn (k − l)
Yl,k = X exp exp (43)
N k N N
n=N −p

and

 i,i+1   
Y  NX−1  
 l,k 
 1  j2πn (k − l) 
 i+1  =  exp . (44)
 Xk  N   N 
n=N −p 

Similarly, for PCC-OFDM encoding at the transmitter only, it can be shown that

 
i,i+1 i,i+1 1 i+1 −j2π (N − p) k
Yl,k + Yl,k+1 = Xk exp
N N
 
N −1     (45)
X j2πn (k − l) j2π (−N + p + n)
× exp 1 − exp 
N N
n=N −p

so

 i,i+1  
i,i+1 

Y −1
 NX
+ Y
   
 l,k l,k+1  1 
 j2πn (k − l) j2π (−N + p + n) 
i+1 = exp 1 − exp . (46)
 N N N

 Xk 
n=N −p



Performing a similar analysis for PCC-OFDM with both denoting the component of Zli,i i+1
0 due to Dk 0 as Zli,i+1
0 ,k 0 , it can

transmitter mapping and receiver weighting-and-adding, and be shown that

−j2π (N − p) 2k 0
 
1 i+1
Zli,i+1
0 ,k 0 = Dk exp
2N N
N −1 (47)
j2πn (2k 0 − 2l0 )
      
X j2πn j2π (−N + p + n)
× 1 − exp exp 1 − exp
N N N
n=N −p

VOLUME 4, 2016 19

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

and

 i,i+1   
Z 0 0   NX−1     0 0
  
 l ,k  1  j2πn j2πn (2k − 2l ) j2π (−N + p + n) 
 i+1  =  1 − exp exp 1 − exp . (48)
 Dk  2N   N N N 
n=N −p 

and
 i,i   
Y  N −1 
APPENDIX. F - ANALYSIS OF ICI CAUSED BY  l,k  1  X j2πn (k − l + ∆f T ) 
 i = exp . (55)
FREQUENCY OFFSET  Xk  N  N


n=0
The ICI due to frequency offset in PCC-OFDM was analyzed
in [22]. Here we present a more detailed analysis which an- From (55) it can be seen that the magnitude of ICI due
alyzes the separate contributions of transmitter and receiver to frequency offset depends only on (k − l), the difference
processing in PCC-OFDM. between the subcarrier indices and ∆f T , where ∆f T is the
Consider the case where the difference in frequency be- frequency offset normalized to the subcarrier spacing.
tween the frequency of the receiver local oscillator and the For PCC-OFDM, considering only the transmitter weight-
carrier frequency of the received signal is ∆f and the phase ing,
offset at the start of the i-th symbol is θ0 [22]. Assuming no
timing offset, a flat channel and no noise, then i,i i,i
Yl,k + Yl,k+1
N −1  
yi,i = y0i , y1i · · · , yni · · · yN
 i

(49) 1 i
X j2πn (k − l + ∆f T )
−1 = exp (jθ0 ) Xk exp
N n=0
N
where N −1  
1 X j2πn (k + 1 − l + ∆f T )

j2πn∆f T
 − exp (jθ0 ) Xki exp
yni = exp (jθ0 ) xin exp . (50) N n=0
N
N (56)
So
which can be simplified to
N −1 i,i i,i
1 X Yl,k + Yl,k+1
Yli,i = √ exp (jθ0 ) xin
N n=0 N −1  
 (51) 1 X j2πn (k − l + ∆f T )

j2πn∆f T
 
−j2πln = exp (jθ0 ) Xki exp
× exp exp . N n=0
N
N N   
j2πn
And using (4) gives × 1 − exp .
N
(57)
N −1 N −1  
1 X 1 X i j2πkn
Yli,i = √ exp (jθ0 ) √ Xk exp Therefore,
N n=0 N k=0 N
 i,i
 Y + Y i,i 
  N −1  
j2πn (k − l + ∆f T )
   
j2πn∆f T −j2πln  l,k l,k+1  1  X
× exp exp . =  exp
Xki  N N

N N 
n=0
(52)    
j2πn
× 1 − exp .

Rearranging gives N 
(58)
N −1 N −1  
1 X X j2πn (k − l + ∆f T )
Yli,i = exp (jθ0 ) i
Xk exp . Now considering PCC-OFDM, including the receiver
N n=0 k=0
N
weighting-and-adding step, gives
(53)
So, for conventional OFDM where the subcarriers are inde- 1  i,i 
pendently modulated Zli,i
0 ,k 0 = Y2l0 ,2k0 + Y2li,i0 ,2k0 +1
2 (59)
N −1 1  i,i 
− Y2l0 +1,2k0 + Y2li,i0 +1,2k0 +1 .
 
i,i 1 i
X j2πn (k − l + ∆f T )
Yl,k = exp (jθ0 ) Xk exp 2
N n=0
N
(54) Substituting (57) in (59), and with some rearrangement gives
20 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

[16] B. Farhang-Boroujeny, “OFDM versus filter bank multicarrier,” IEEE

N −1   2 Signal Processing Magazine, vol. 28, no. 3, pp. 92–112, 2011.
1 X j2πn [17] G. Wunder, P. Jung, M. Kasparick, T. Wild, F. Schaich, Y. Chen,
Zli,i
0 ,k 0 =
i
exp (jθ0 ) X2k 0 1 − exp S. Ten Brink, I. Gaspar, N. Michailow, A. Festag et al., “5GNOW:
2N n=0
N
non-orthogonal, asynchronous waveforms for future mobile applications.”
0 0
 
j2πn (2k − 2l + ∆f T ) IEEE Communications Magazine, vol. 52, no. 2, pp. 97–105, 2014.
× exp [18] V. Vakilian, T. Wild, F. Schaich, S. ten Brink, and J.-F. Frigon, “Universal-
N filtered multi-carrier technique for wireless systems beyond LTE,” in IEEE
(60) Globecom Workshops (GC Wkshps), 2013, pp. 223–228.
[19] T. Tang, Y. Zhang, J. Jiang, and P. Zhang, “A MIMO-PCC-OFDM system
and for multipath fading channels with carrier frequency offset,” in Inter-
national Symposium on Communications and Information Technologies,
 i,i 
Z 0 0 
 N −1   2 2006, pp. 511–515.
 l ,k  1  X j2πn [20] X. Li and J. Armstrong, “An Improved MIMO PCC OFDM System with
 i = 1 − exp Reduced Sensitivity to Imperfect CSI,” in IEEE Wireless Communications
 X2k0  2N  N

n=0 and Networking Conference (WCNC), 2010, pp. 1–6.
 (61)
j2πn (2k 0 − 2l0 + ∆f T )  [21] J. Armstrong, P. M. Grant, and G. Povey, “Polynomial cancellation coding

× exp . of OFDM to reduce intercarrier interference due to Doppler spread,” in
N  IEEE Global Telecommunications Conference, vol. 5, 1998, pp. 2771–
REFERENCES
[1] M. Shafi, A. F. Molisch, P. J. Smith, T. Haustein, P. Zhu, P. Silva,
F. Tufvesson, A. Benjebbour, and G. Wunder, “5G: A tutorial overview of
standards, trials, challenges, deployment, and practice,” IEEE Journal on
Selected Areas in Communications, vol. 35, no. 6, pp. 1201–1221, 2017.
[2] P. Banelli, S. Buzzi, G. Colavolpe, A. Modenini, F. Rusek, and A. Ugolini,
“Modulation formats and waveforms for 5G networks: Who will be the
heir of OFDM?: An overview of alternative modulation schemes for
improved spectral efficiency,” IEEE Signal Processing Magazine, vol. 31,
no. 6, pp. 80–93, 2014.
[3] Z. E. Ankarali, B. Peköz, and H. Arslan, “Flexible Radio Access Beyond
5G: A Future Projection on Waveform, Numerology, and Frame Design
Principles,” IEEE Access, vol. 5, pp. 18 295–18 309, 2017.
[4] Q. Bodinier, F. Bader, and J. Palicot, “On Spectral Coexistence of CP-
OFDM and FB-MC Waveforms in 5G Networks,” IEEE Access, vol. 5,
pp. 13 883–13 900, 2017.
[5] A. Ijaz, L. Zhang, M. Grau, A. Mohamed, S. Vural, A. U. Quddus, M. A.
Imran, C. H. Foh, and R. Tafazolli, “Enabling massive IoT in 5G and
beyond systems: PHY radio frame design considerations,” IEEE Access,
vol. 4, pp. 3322–3339, 2016.
[6] Y. Medjahdi, S. Traverso, R. Gerzaguet, H. Shaiek, R. Zayani, D. Demmer,
R. Zakaria, J.-B. Doré, M. B. Mabrouk, D. Le Ruyet et al., “On the road to
5G: comparative study of physical layer in MTC context,” IEEE Access,
vol. 5, pp. 26 556–26 581, 2017.
[7] N. Michailow, M. Matthé, I. S. Gaspar, A. N. Caldevilla, L. L. Mendes,
A. Festag, and G. Fettweis, “Generalized frequency division multiplexing
for 5th generation cellular networks,” IEEE Transactions on Communica-
tions, vol. 62, no. 9, pp. 3045–3061, 2014.
[8] G. Fettweis, M. Krondorf, and S. Bittner, “GFDM-generalized frequency
division multiplexing,” in IEEE 69th Vehicular Technology Conference,
2009, pp. 1–4.
[9] S. N. Premnath, D. Wasden, S. K. Kasera, N. Patwari, and B. Farhang-
Boroujeny, “Beyond OFDM: Best-effort dynamic spectrum access using
filterbank multicarrier,” IEEE/ACM Transactions on Networking (TON),
vol. 21, no. 3, pp. 869–882, 2013.
[10] F. Schaich and T. Wild, “Waveform contenders for 5G - OFDM vs. FBMC
vs. UFMC,” in 6th International Symposium on Communications, Control
and Signal Processing (ISCCSP), 2014, pp. 457–460.
[11] B. Farhang-Boroujeny and H. Moradi, “OFDM inspired waveforms for
5G,” IEEE Communications Surveys & Tutorials, vol. 18, no. 4, pp. 2474–
2492, 2016.
[12] J. Armstrong, “OFDM for optical communications,” Journal of Lightwave
Technology, vol. 27, no. 3, pp. 189–204, 2009.
[13] T. Jiang and Y. Wu, “An overview: Peak-to-average power ratio reduc-
tion techniques for OFDM signals,” IEEE Transactions on Broadcasting,
vol. 54, no. 2, pp. 257–268, 2008.
[14] S. H. Han and J. H. Lee, “An overview of peak-to-average power ratio
reduction techniques for multicarrier transmission,” IEEE Wireless Com-
munications, vol. 12, no. 2, pp. 56–65, 2005.
[15] B. E. Priyanto, H. Codina, S. Rene, T. B. Sorensen, and P. Mogensen,
“Initial performance evaluation of DFT-spread OFDM based SC-FDMA
for UTRA LTE uplink,” in IEEE 65th Vehicular Technology Conference.
IEEE, 2007, pp. 3175–3179.
VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
2776.
[22] J. Armstrong, “Analysis of new and existing methods of reducing in-
tercarrier interference due to carrier frequency offset in OFDM,” IEEE
Transactions on Communications, vol. 47, no. 3, pp. 365–369, 1999.
[23] Y. Zhao and S.-G. Haggman, “Sensitivity to Doppler shift and carrier
frequency errors in OFDM systems-the consequences and solutions,” in
IEEE 46th Vehicular Technology Conference, vol. 3, 1996, pp. 1564–1568.
[24] ——, “Intercarrier interference self-cancellation scheme for OFDM mo-
bile communication systems,” IEEE Transactions on Communications,
vol. 49, no. 7, pp. 1185–1191, 2001.
[25] J. Shentu and J. Armstrong, “Blind estimation of time and frequency offset
for PCC-OFDM,” in IEEE 53rd Vehicular Technology Conference, vol. 1,
2001, pp. 702–706.
[26] J. Shentu, K. Panta, and J. Armstrong, “Effects of phase noise on per-
formance of OFDM systems using an ICI cancellation scheme,” IEEE
Transactions on Broadcasting, vol. 49, no. 2, pp. 221–224, 2003.
[27] W. T. Ng and V. K. Dubey, “Analysis of PCC-OFDM systems for general
time-varying channel,” IEEE Communications Letters, vol. 9, no. 5, pp.
394–396, 2005.
[28] A. S. Alaraimi and T. Hashimoto, “Performance analysis of polynomial
cancellation coding for OFDM systems over time-varying Rayleigh fading
channels,” IEICE Transactions on Communications, vol. 88, no. 2, pp.
471–477, 2005.
[29] K. Panta and J. Armstrong, “Spectral analysis of OFDM signals and its
improvement by polynomial cancellation coding,” IEEE Transactions on
Consumer Electronics, vol. 49, no. 4, pp. 939–943, 2003.
[30] T. Hashimoto, A. S. Alaraimi, and C. Han, “ICI Mitigation Schemes for
Uncoded OFDM over Channels with Doppler Spreads and Frequency
Offsets–Part I,” J. Commun., vol. 7, no. 9, pp. 676–685, 2012.
[31] ——, “ICI Mitigation Schemes for Uncoded OFDM over Channels with
Doppler Spreads and Frequency Offsets-Part II: Asymptotic Analysis,”
JCM, vol. 7, no. 9, pp. 686–700, 2012.
[32] J. Armstrong, T. Gill, and C. Tellambura, “Performance of PCC-OFDM
with overlapping symbol periods in a multipath channel,” in IEEE Global
Telecommunications Conference, vol. 1, 2000, pp. 87–91.
[33] Q. Inc., “R1-162199: Waveform Candidates,” 2016.
[34] G. J. Foschini and M. J. Gans, “On limits of wireless communications in
a fading environment when using multiple antennas,” Wireless personal
communications, vol. 6, no. 3, pp. 311–335, 1998.
[35] E. Telatar, “Capacity of multi-antenna Gaussian channels,” European
Transactions on Telecommunications, vol. 10, no. 6, pp. 585–595, 1999.
[36] S. Wang, J. S. Thompson, and P. M. Grant, “Closed-form expressions for
ICI/ISI in filtered OFDM systems for asynchronous 5G uplink,” IEEE
Transactions on Communications, vol. 65, no. 11, pp. 4886–4898, 2017.
[37] S. Wang, J. Armstrong, and J. S. Thompson, “Waveform performance
for asynchronous wireless 5G uplink communications,” in IEEE Personal,
Indoor, and Mobile Radio Communications (PIMRC), 2016, pp. 1–6.
[38] G. Kongara and J. Armstrong, “Implementation of PCC OFDM on a
software defined radio platform,” in 26th International Telecommunication
Networks and Applications Conference (ITNAC), 2016, pp. 295–299.
[39] ——, “Performance evaluation of PCC OFDM on a software defined radio
platform,” in International Conference on Computing, Networking and
Communications (ICNC), 2017, pp. 287–292.
21
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2949792, IEEE Access

G. Kongara et al.: A Comparison of CP-OFDM, PCC-OFDM and UFMC for 5G Uplink Communications

[40] G. Kongara, J. Armstrong et al., “Implementation of PCC-OFDM on a LEI YANG received the B.E. and Ph.D. degrees in
software-defined radio testbed,” Australian Journal of Telecommunica- electrical and computer system engineering from
tions and the Digital Economy, vol. 5, no. 2, p. 17, 2017. the Monash University in 2008 and 2012, respec-
[41] K. Sathananthan, R. Rajatheva, and S. B. Slimane, “Analysis of OFDM tively. In 2011, He joined Nokia Networks as a
in the presence of frequency offset and a method to reduce performance senior algorithm and system engineer and worked
degradation,” in IEEE Global Telecommunications Conference, vol. 1. on 4G/5G key technologies such as CoMP, MIMO
IEEE, 2000, pp. 72–76. etc. Since 2017, he has been involved in new
[42] K. Sathananthan, C. R. Athaudage, and B. Qiu, “A novel ICI cancellation
5G waveform and optical wireless communication
scheme to reduce both frequency offset and IQ imbalance effects in
research at Monash University.
OFDM,” in Ninth International Symposium on Computers And Commu-
nications Proceedings, vol. 2. IEEE, 2004, pp. 708–713.
[43] T. Van Waterschoot, V. Le Nir, J. Duplicy, and M. Moonen, “Analytical
expressions for the power spectral density of CP-OFDM and ZP-OFDM
signals,” IEEE Signal Processing Letters, vol. 17, no. 4, pp. 371–374,
2010.
[44] Q. Liu, Z. Zhou, W. Zou, S. Kan, and Y. Ye, “Peak-to-Average Power
Ratio Analysis of PCC-OFDM Signals,” in 2009 IEEE 70th Vehicular
Technology Conference Fall (VTC 2009-Fall). IEEE, 2009, pp. 1–5.
[45] S. Boumard, M. Lasanen, O. Apilo, A. Hekkala, C. Cassan, J.-P. Verdeil,
J. David, and L. Pichon, “Power consumption trade-off between power
amplifier OBO, DPD, and clipping and filtering,” in 26th International
Teletraffic Congress (ITC). IEEE, 2014, pp. 1–5.

JEAN ARMSTRONG received the B.Sc. (First

GAYATHRI KONGARA received her Ph.D in
Electrical and electronic engineering from the
University of Canterbury in 2010. She was a
research engineer at Wireless Research Center
(WRC), NZ until 2013 and joined Monash Univer-
sity in 2014. She specializes in the design and im-
plementation of signal processing algorithms for
5G wireless communication systems. Her research
interests include MIMO receiver implementation,
Physical layer for 5G and software defined radios.
CUIWEI HE received the B.E. degree in telecom-
munication engineering from Central South Uni-
versity of Forestry and Technology, Changsha,
China, in 2010, the M.E. degree in digital sig-
nal processing from University of the Ryukyus,
Okinawa, Japan, in 2013, and the Ph.D. degree
in optical wireless communication from Monash
University, Melbourne, Australia, in 2017, where
he has been a Research Fellow with the Depart-
ment of Electrical and Computer Systems Engi-
neering, since September 2017. His research interests are digital signal
processing, optical wireless communication, multiple-input multiple-output
(MIMO) technology and optical orthogonal frequency division multiplexing
(OFDM).
Class Honours) degree in electrical engineering
from the University of Edinburgh, Edinburgh,
U.K., and the M.Sc. degree in digital techniques
from Heriot-Watt University, Edinburgh, U.K. She
is a Professor at Monash University, Melbourne,
Victoria, Australia, where she leads a research
group working on topics including optical wire-
less, radio frequency and optical fiber communica-
tions. She has published numerous papers includ-
ing over 70 on aspects of OFDM for wireless and optical communications
and has six commercialized patents. Before emigrating to Australia, she
worked as a Design Engineer at Hewlett-Packard Ltd., Scotland. In Aus-
tralia, she has held a range of academic positions at Monash University, the
University of Melbourne and La Trobe University.
During her career, she has received numerous awards including a Carolyn
Haslett Memorial Scholarship, a Zonta Amelia Earhart Fellowship, the Peter
Doherty prize for the best commercialization opportunity in Australia (joint
winner), induction into the Victorian Honour Roll of Women, the 2014
IEEE Communications Society Best Tutorial Paper Award, and the 2016
IET Mountbatten Medal. She is a Fellow of the IEEE and IEAust. She has
served on the Australian Research Council (ARC) College of Experts and
the ARC Research Evaluation Committee.

22 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.