Documente Academic
Documente Profesional
Documente Cultură
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Each additional bit placed on subchannel that requires the least incremental energy for its transport Incremental energy en (bn ) n (bn ) n (bn )
e (b ) g bn = log 2 1 + n n n
bit increment
Cant move a bit from one channel to another and reduce the total energy
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Always replace a bit distribution with a more efficient exhaustively search all single information unit changes at each step
Smallest energy to add a new bit Largest energy to subtract the bit Subtract bits from costly sub-channels and add to least costly sub-channels
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Efficientizing example
bn
bn + 1
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
If energy constraint violated subtract the most costly bit If energy less than max add the bit that costs the least to add
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Margin
N x
8 = 1.8 dB 5.32
Choose any bit distribution Make it efficient using (EF algorithm) Make it energy tight using (ET algorithm)
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Change the loading when channel changes LC is a natural candidate Keep the ET bit distribution and perturb based on channel changes
Channel partitioning
. . .
N-1(t) N(t) X1,k X2,k XN,k
x(t) h(t) +
. . .
N-1(T-t) N(-t)
t=T t=T
n1,k
(MIMO)
Qk
n2,k nN,k
+ +
Qk = Ik
Modal modulation
(-TH / 2, TH / 2)
n
T/2
1(T-t) 2(T-t)
. . .
N-1(t) N(t)
x(t) h(t) +
. . .
N-1(T-t) N(-t)
t=T t=T
yN-1,k yN,k
x(t) = y(t) =
xnn(t)
~(t) xn . [r(t) * n (t)] + n
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Eigenfunctions hard to compute and channel dependent As both N->inf and [-T/2, T/2]->(-inf,+inf)
Set of eigenvalues of any autocorrelation function is unique This set determines the performance of MM through SNR Eigen-functions are not unique is also a valid eigen-function for inf symbol period Corresponding eigenvalues are R(2n/T) No ISI on any tone since symbol period is infinite Each tone is AWGN channel SBS detector is MAP optimal are then eigen-functions even on finite [-T/2,T/2] Can use extra bandwidth in the design to make the channel look periodic
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
10
Can try to create a more sinc-looking symbols in time by filtering the tones in frequency domain
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
11
Digital realization
X0 m0 1/T x + x (t) (t) (LPF) y = Px + n h (t) n (t) + *(-t) y(t) (LPF)
A D C
Y0
X1
m1
Y1
D A C
XN-2
mN-2
* fN -2
YN-2
XN-1
mN-1
* fN -1
YN-1
yN-1 yN-2
p0 = 0 0 0
p1 p0 0
py py-1 p0
0 py p1
0 0 0 py
xN-1
x0 x-1 x-y
nN-1 nN-1 n0
y0
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
12
Vector coding
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
13
VC example
1+0.9D-1, N=8
SVD on Gives singular values Sub-channel SNRs Waterfilling shows only 7 dimensions can be used Sub-channel energies SNRs are then Total SNR VC capacity
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
14
Discrete multitone (DMT) and Orthogonal frequency Division Multiplexing (OFDM) DMT used on slowly time-varying channels
Optimize bn and En per sub-channel But uses same bn and En on all channels Used on one-way broadcast channels In vector coding, M,F channel dependent Make the channel circular and make M,F channel independent - simplify hardware implementation
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
15
CP DMT
xN ... xN 1 x0 ...
VC
0...0
x0 ...
xN ... xN 1
SVD can be replaced by eigen-decomposition (spectral factorization) A discrete form of modal modulation While SNRs are unique, many choices for M and F
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
16
DFT
Proof
qn is eigen-vector of P
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
17
DMT/OFDM implementation
X0
1 = N+v T
D A C
Y0
1 = N+v T (-t)
y(t)
(LPF)
A D C
X1
. . .
XN-2 XN-1
IDFT
x(t) (t)
(LPF)
n(t) h(t) +
Y1
DFT
. . .
YN-2 YN-1
18
19
Rx/Tx Windowing
Rectangular window Raised cosine 5% Raised cosine 25%
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
20
N=8
21
Example 1+0.9D-1 channel MMSE-DFE, SNR=7.6 dB, 3ff, 1fb tap, 4 mac/sample VC, N=8, SNR=8.1 dB, 7*8/9=6.2MAC/sample DMT, N=8, SNR=7.6 dB, 8pt FFT/IFFT, 2.7MAC/sample
N=16, 3.8MAC/sample, SNR=8.8 dB DFE needs 10FF taps, 1FB tap, SNR=8.4 dB, 11MAC/sample
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
22
Split
1.5 Mbps
Split
ADSL
Telco CO
internet
138 kHz
frequency
1/T=2.208 MHz (CP = 40 samples) Each time domain symbol 2*256+40=552 samples
Hermitian symmetry creates real signal transmitted from 0-1.1MHz First 2-3 tones near DC not used avoid interference with voice Tone 256 also not used, 64 reserved for pilot
1.1 MHz
POTS
psd
UP Up
10 kHz 0 Hz
Down Down
23
40 mW
200 mW
800 mW
5.15 GHz
5.35 GHz
5.725 GHz
5.825 GHz
not used
Symbol length = 80 samples, CP=16 Symbol rate 250kHz (T=4uS, T=50ns), CPguard=0.8us
6 9 12 18 24 36 48 54 BPSK BPSK 4QAM 4QAM 16QAM 16QAM 64QAM 64QAM
20 MHz
bn
1/2 3/4 1 3/2 2 3/2 3 9/2
bn
1/4 3/8 1/2 3/4 1 3/4 3/2 9/4
b
24 36 48 72 96 144 192 216
4 pilots
48 data tones
FCC demands flat spectrum so no energy-allocation The only knob is data rate selection
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
24
Data link control layer (DLC) Physical radio layer (PHY) Digital Guard Viterbi Demodulation interval decoder + + descrambler deinterleaver extraction + FFT Scrambler + forward error correction (FEC) coder Interleaver + modulation IFFT + guard interval insertion A/D Analog Analog front-end
D/A
Synchronization
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
25
Transceiver architecture
Pilot insertion Cyclic prefix Windowing Interleaver Upconvert Scrambler Descrambler Encoder Viterbi decoder Mapper Demapper DAC
FFT/IFFT
Channel estimator
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
Deinterleaver
Remove prefix
Synchronizer
AGC& ADC
[2]
26
Convolutional Encoder
Puncturer
Interleaver
Signal Mapper
IFFT
Cyclic Extend
RF
Preamble
D/A
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
27
Scrambling
Need to randomize incoming data Enables a number of tracking algorithms in the receiver Provides flat spectrum in the given band
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
28
Interleaver
Block size is the #of coded bits in OFDM symbol (NCBPS) Two-step permutation Adjacent coded bits mapped
Onto nonadjacent sub-carriers Alternate between less and more significant bits in the constellation avoid long runs of low reliability LSBs
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
29
Convolutional Encoder
g0=1338
Input Data
Stolen Bit
Tb
Tb
Tb
Tb
Tb
Tb
g1=1718
64-state (constraint length K=7) code Viterbi algorithm applied in the decoder
6.973 Communication System Design
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
30
Signal mapper
Data divided into groups of (1,2,4,6) bits and mapped to a constellation point (i.e. a complex number) Gray-coded constellation mapings
31
Pilot insertion
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
32
Spectral mask
Cannot use last 5 tones on each side Does not use extra windowing
Power Spectral Density (dB) Transmit Spectrum Mask (not to scale) -20 dBr
-28 dBr
fc
9 11
20
30 Frequency (MHz)
33
Receiver architecture
I and Q from analog front-end Rx data to MAC
ADC
FIR
Remove DC of fset
Rotate
FFT
Channel
correct
Deinterleave
Viterbi
FIR
Frequency lock
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
34
t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 GI2
T1
T2
Signal Detect, Coarse Freq. Channel and Fine Frequency RATE AGC, Diversity Offset Estimation Offset Estimation LENGTH Selection Timing Synchronize
to timing control
training symbol pilots pilots long1 training from fft long2 training
angle adjust
magnitude adjust
fir
complex inverse
rotate
data
35
Synchronizer
FFT
IJ*
Plateau Detector
Ig.1(.) Jc(k)
Combine
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
36
Channel estimator
Equalizer Sync. + FFT Reference Sign Correction X(i,k) = H(i-D,k) Y(i,k) H(i-D,k) X(i,k) Residual Phase Correction Soft Demapper Deinterleaver Soft Viterbi
Channel Estimator Delay Buffer (D) HREF (k) Pilot Sign Correction Zero Forcing Y(i-D,k) H(i-D,k) = X(i-D,k) Mapper X(i-D,k) Interleaver Conv. Encoder Estimation Buffer (Circular Buffer)
Y(i-D,k)
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
37
[1]
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
38
References
[1] T.H. Meng, B. McFarland, D. Su and J. Thomson "Design and implementation of an all-CMOS 802.11a wireless LAN chipset," Communications Magazine, IEEE vol. 41, no. 8 SN - 0163-6804, pp. 160-168, 2003 [2] M. Krstic, K. Maharatna, A. Troya, E. Grass and U. Jagdhold "Implementation of an IEEE 802.11a compliant low-power baseband processor," Telecommunications in Modern Satellite, Cable and Broadcasting Service, 2003. TELSIKS 2003. 6th International Conference on vol. 1, no. SN -, pp. 97-100 vol.1, 2003. [3] J. Thomson, B. Baas, E.M. Cooper, J.M. Gilbert, G. Hsieh, P. Husted, A. Lokanathan, J.S. Kuskin, D. McCracken, B. McFarland, T.H. Meng, D. Nakahira, S. Ng, M. Rattehalli, J.L. Smith, R. Subramanian, L. Thon, Y.-H. Wang and R. Yu "An integrated 802.11a baseband and MAC processor," Solid-State Circuits Conference, 2002. Digest of Technical Papers. ISSCC. 2002 IEEE International vol. 1, no. SN -, pp. 126451 vol.1, 2002. [4] E. Grass, K. Tittelbach-Helmrich, U. Jagdhold, A. Troya, G. Lippert, O. Kruger, J. Lehmann, K. Maharatna, K.F. Dombrowski, N. Fiebig, R. Kraemer and P. Mahonen "On the single-chip implementation of a Hiperlan/2 and IEEE 802.11a capable modem," Personal Communications, IEEE [see also IEEE Wireless Communications] vol. 8, no. 6 SN - 1070-9916, pp. 48-57, 2001.
Cite as: Vladimir Stojanovic, course materials for 6.973 Communication System Design, Spring 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY].
39