Advances in Wireless Communication 课堂笔记（中下）

上：Advances in Wireless Communication 课堂笔记（上）
中：Advances in Wireless Communication 课堂笔记（中）

Advanced modulation techniques, Iterative detection and decoding

Modulation in 5G

Mainly uses QAM schemes
Going back to structure of PDSCH for an M-ary modulation (M-QAM), the encoded interleaved rate-matched scrambled bits are splited into chunck of m bits and the mapped to the complex modulated symbols

Modulation coding scheme MCS

Modulation scheme	Modulation order
QPSK	2
16-QAM	4
64-QAM	6
256-QAM	8
release 17:1024-QAM	10

MCS table between 0 and 31 (5 bits representation)

MCS index	Modulation order	code rate (*1024)	spectral efficiency
0	2	120(lowest rate=120/1024)	0.234
…	…	…	…
27	8	948	7.4

Bounds on symbol error probability
M-ary modulation with signal set {$S_1(t),S_2(t),...,S_n(t)$} Let $E_{ij}$ be the event that $$\begin{gathered} P(y(t)|S_j(t))>P(y(t)|S_i(t)) \\ P(error|S_i(t))=P(\bigcup _{j=1,j\ne i}^M{E}_{ij}) \\ P(error|S_i(t))<\sum _{j=1,j\ne i}^{M}P(E_{ij}) \\ \Rightarrow \sum _{j=1,j\ne i}^{M}Q(D_{ij}/\sqrt{2N_o})\le (M-1)Q(\frac{D_{min}(i)}{\sqrt{2N_o}}) \end{gathered}$$
Union bound $P({\bigcup }_{i=1}^n{A}_i\le {\sum }_{i=1}^{n}P(A_i))$
$D_{min}(i)$

$D_{min}(i)=min{D}_{ij}1\le j\le m,j\ne i$

AWGN:

$P(E_{ij})=P(n\ge \frac{D_{ij}}{2})=Q(\frac{D_{ij}}{\sqrt{2N_o}})$
n: one component of Gaussian noise~$\mathcal{N}(0,\frac{N_o}{2})$

Also for linear bound
$P(error|S_i(t))\ge maxP(E_{ij})(i\ne j)=Q(D_{min}(i)/\sqrt{2N_o})$
For the average probability of error
$P_{av}(error)=\frac{1}{M}{\sum }_{i=1}^{m}P(error|S_i(t))\le (M-1)Q(\frac{D_{min}}{\sqrt{2N_o}})$

Bit-level demodulation
Example:
QPSK symbol($b_1,b_2$), received $y(y_1,y_2)$
$y=s_i+(n_1,n_2)$

gray-labling

ML decision for $b_1$
$b_1=0$ if $P(y|b_1=0)>P(y|b_1=1)$
$$\underset{S_1}{\underbrace{P(y|b_1=0,b_2=0)}}\underset{\frac{1}{2}}{\underbrace{P(b_2=0)}}+\underset{S_2}{\underbrace{P(y|b_1=0,b_2=1)}}\underset{\frac{1}{2}}{\underbrace{P(b_2=1)}}$$

Similarly
$P(y|b_1=1)=\frac{1}{2}P(y|S_3)+\frac{1}{2}P(y|S_4)$
Decide $b_1=0$ if $y_2\ge 0$
Decide $b_2=0$ if $y_1\ge 0$
P(error in $b_1$)=$P(n_2\le -1)=Q(\frac{1}{\sigma })$
this is the optimal labeling rule

An alternative which is worse:

Suppose $S_1$ is transmitted $P(correct{b}_2|S_1)=(1-Q(\frac{1}{\sigma }){)}^2+Q(\frac{1}{\sigma }{)}^2=1-2Q(\frac{1}{\sigma })+2Q(\frac{1}{\sigma }{)}^2$
P(error in $b_2)$=$$\begin{gathered} 2Q(\frac{1}{\sigma })-2Q(\frac{1}{\sigma }{)}^2>Q(\frac{1}{\sigma }) \\ \Rightarrow Q(\frac{1}{\sigma })>2Q(\frac{1}{\sigma }{)}^2 \\ \iff \frac{1}{2}>Q(\frac{1}{\sigma }) \end{gathered}$$

In general for M_PSK gray labeling is an ording if binary strings on the circle, such that two successive values differ in only on bit

8PSK gray coding and 16PSK gray coding

8PSK:

16PSK

liklihood ratios for bit $b_i$:
$$log\frac{P(y|b_i=0)}{P(y|b_i=1)}$$
this is LLR for bit $b_i$, which will be passed to the decoder

Bit-Interleaved coded Modulation (BICM)

16 QAM with gray mapping:

consider a symbol $x=(x_1,x_2,x_3,x_4),x_1$and $x_3$ observe identical and independence channels
The bits $x_2$ and $x_4$ also observe indentical an independence channels

In general BICM with M-ary modulation $M=2^m$
can be decomposed into m parallel channels

For 16-QAM, two types of channel, two from each type $x_1$ & $x_2$, $x_3$ & $x_4$ correlation

Coded modulation capacity and BICM capacity
channel capacity $c=$$\genfrac{}{}{0ex}{}{sup}{P_x}$ $I(x;y)$
$P_x$: input distribution

For AWGN with input power P, noise power N, bandwidth W $C=\frac{1}{2}log(1+P/W)=\frac{1}{2}log(1+SNR)$

Coded modulation capacity $I(X,Y)$
$X$ belongs to the constellation set of modulation with uniform distribution

$x=(x_1,x_2,x_3,x_4)$
$I(x_1;y)+I(x_2;y)+I(x_3;y)+I(x_4;y)\le I(x_1,x_2,x_3,x_4;y)$

For BICM, with $M=2^m$ (size of the constellation) in parallel channels with inputs $x_1,x_2,x_3,...,x_m$， $C_{BICM}={\sum }_{i=1}^{m}I(x_i;y)$

Coded modulation capacity and BICM capacity

BICM capacity (dash-dotted lines)
with Gray labeling, the difference between BICM capacity and coded modulation capacity is very small

Basics of classical multiple access techniques

Multiuser Communication

Two major scenarios
Broadcast Channel (downlink):

Multiple access channel (uplink):

Two early method of multiple access techniques

1. FDMA (frequency division multiple access)
2. TDMA (time division multiple access)

cross-talk or interference
3. implementation of filters at all possible frequency bands
4. crosstalk
guard band to reduce crosstalk

for TDMA, time synchronization is very challenging as the number of user grow

In both TDMA and FDMA, we allocate orthogonal resources either in time or frequency to users
$S_1(t)$ and $S_2(t)$ are orthogonal: $\int S_1(t)S_2(t)dt=0$

CDMA code division multiple access (CDMA)
In CDMA, each users is assigned a distinct signature sequence (or waveform) which the user employs to spread the signal

CDMA is a spread spectrum technique, where symbol rate is much smaller than the bandwidth
symbol rate is proportion to $\frac{1}{T}$

These techniques provide frequency diversity against “selective fading”
selective fading:

one of the most challenging parts of CDMA systems is the design of equalizes
$h(t)$:h(0),h(1),…,h(d) — all channel gain
$y(t)=x(t)\ast h(t),y[n]=x[n]\ast h[n]$

orthogonal frequency division multiplexing (OFDM)
$X_0,X_1,...,X_{N-1}$ denote N complex symbols (in frequency domain)
Taking the IFFT
$x_n=\frac{1}{\sqrt{N}}{\sum }_{k=0}^{N-1}{X}_k{e}^{\frac{j2\pi nk}{N}}forn=0,1,...,N-1$
$x_i$ s are in the time domain
Then $x_0,x_1,...,x_{N-1}$ are passed through a D/A (digital to analog) converter resulting in the baseband OFDM signal $x(t)$, $x(t)=\frac{1}{\sqrt{N}}{\sum }_{k=0}^{N-1}{X}_k{e}^{\frac{j2\pi k_t}{T_N}}0\le t\le T_N$ . Where $T_N$ is the duration of OFDM signal.
The sub-carrier frequencies are $f_i=i/T_N,i=0,1,...,N-1$ and the samples $x_0,x_1,...,x_{N-1}$ represent samples of $x(t)$ every $T_N/N$ second

The block diagram of an OFDM system

A major advantage of OFDM comparing with CDMA sis its resilience to ISI(inter-symbol interference) because the data rate on each sub-channel is very low ($B/N<<B_c$, coherence bandwidth), $B_c$ proportional to $1/D$, where D is delay spread. Channel equalization is much simpler since channel gain can be consider flat at each sub-channel

Non-orthogonal multiple access techniques

Cyclic prefix for OFDM:

$$\begin{gathered} \underset{\text{data symbols in time domain}}{\underbrace{x_0,x_1,...x_{N-1}}}\genfrac{}{}{0ex}{}{IFFT}{\leftarrow }\underset{\text{data symbols in frequency domain}}{\underbrace{X_0,X_1,...,X_{N-1}}} \\ y[n]=x[n]\ast c[n]=\sum _{k=n-\mu +1}^{n}x[k]c[n-k] \end{gathered}$$

channel delay spread is at most $\mu T_N/N$
In the frequency domain, see $X\cdot C$

If we consider the convolution of $x_{N-\mu },...,x_{N-1},x_0,x_1,....x_{N-1}$ with $c_0,c_1,...,c_{\mu }$ and remove the last $\mu$ samples, we get a linear convolution which is equivalent to circular convolution
Therefore in the frequency domain, we have ${\hat{x}}_k=c_k\times x_k$
$c_k$: FFT of the sequence ($\underset{\mu +1}{\underbrace{c_0,c_1...c_{\mu }}},\underset{N-\mu +1}{\underbrace{0,0,...}}$)

Therefore the effect of channel, $c(t)$ can be completely removed by a simple equalization in the frequency domain

Uplink AWGN channel

consider the discrete time model with two users $y=x_1+x_2+n,n\sim \mathcal{N}(0,N_0)$ users have powewr constraint
user K power is $P_K$
$R_1<\frac{1}{2}log(1+P_1/N_o)$— maximum capacity user 1 can achieve, $R_1$ : rate for user 1 ($R_1+R_2$) sum rate
user 1 power $P_1$ user 2 power $P_2$

Capacity region pairs of achievable $(R_1,R_2)$

{user1-$P_1⨁$ user2-$P_2$} — superuser of power $P_1+P_2$ and rate $R_1+R_2$
$R_1+R_2<\frac{1}{2}log(1+\frac{P_1+P_2}{N_0})$

How to achieve corner points A B
SIC: successive interference cancellation to achieve A. Decode $x_1$ assuming $x_2$ is noise. Cancel out $x_1$ and decode $x_2$, assuming $x_1$ is known and cancelled.

what rate $R_1$ we can get?
$R_1<\frac{1}{2}log(1+\frac{P_1}{P_2+N_o})$
$log(1+\frac{P_1}{P_2+N_o})+log(1+\frac{P_2}{N_o})=log(1+\frac{P_1+P_2}{N_o})$, so we can achieve A, B

How can we achieve power between A B ?
time sharing. $\lambda$ fraction of time operate at A. ($1-\lambda$) fraction of time operate at A
achieve $\lambda A+(1-\lambda )B$
Any convex combination of A and B is achievable

The difference with an orthogonal method (FDMA or OFDM) for instance, $\alpha$ fraction of Bandwidth is given to user one and ($1-\alpha$) fraction is given to user 2

$C=Wlog\frac{P}{N_{o}W}$ (bit/second)
$C_1=\alpha Wlog\frac{P_1}{\alpha N_{o}W}$
$C_2=(1-\alpha )Wlog\frac{P_1}{(1-\alpha )N_{o}W}$
in the discrete domain
$R_1=\frac{\alpha }{2}log\frac{P_1}{\alpha N_o}$
$R_2=\frac{(1-\alpha )}{2}log\frac{P_1}{(1-\alpha )N_o}$

Extension to K-user uplink
$y=x_1+x_2+...+x_k+n$
power of user i is $P_i$
we can form a superuser for any subset S ⊂ { 1 , . . . , k } S\subset \{1,...,k\} S⊂{1,...,k} of user. For any S ⊂ { 1 , . . . , k } ,    ∑ i ∈ S R i < 1 2 l o g ( 1 + ∑ i ∈ s P i N o ) S\subset \{1,...,k\},\;\sum_{i\in S}R_i<\frac12log(1+\sum_{i\in s}\frac{P_i}{N_o}) S⊂{1,...,k},∑i∈S​Ri​<21​log(1+∑i∈s​No​Pi​​)($R_i$: rate for user i). $2^k-1$ constrains must be satisfied

$$\begin{gathered} R_1<\frac{1}{2}log(1+\frac{P_1}{N_o}) \\ {R}_2<\frac{1}{2}log(1+\frac{P_2}{N_o}) \\ {R}_3<\frac{1}{2}log(1+\frac{P_3}{N_o}) \\ {R}_1+R_2 \\ {R}_1+R_3 \\ {R}_2+R_3 \\ {R}_1+R_2+R_3 \end{gathered}$$
6(=3!) corner point correspond

to one of the 6 possible ordering of user 1 2 3

SIC decode user 1 assuming $x_2,x_3$ are noise, decode $x_2$ give $x_1$(cancel out) and assuming $x_3$ as noise and finally decode $x_3$ given $x_1$ and $x_2$(cancel out)
$$\begin{gathered} R_1<\frac{1}{2}log(1+\frac{P_1}{P_2+P_3+N_o}) \\ {R}_2<\frac{1}{2}log(1+\frac{P_2}{P_3+N_o}) \\ {R}_2<\frac{1}{2}log(1+\frac{P_3}{N_o}) \end{gathered}$$

The sum-capacity(maximum of Sum-rate) is given by
$$C_{sum}=\frac{1}{2}log(1+\frac{{\sum }_{k=1}^k{P}_k}{N_o})$$
If the users powers are equal then sum-capacity is the same for orghogonal and non-orthogonal cases

Downlink AWGN channel

$$\begin{gathered} y_1=h_{1}x+n_1 \\ {y}_2=h_{2}x+n_2 \\ {n}_1\&n_2\sim \mathcal{N}(0,N_o) \\ |h_1|\le |h_2| \end{gathered}$$

Superposition coding, interference cancellation methods

strategy super position coding

$x=x_1(\text{signal intended for user 1})+x_2(\text{signal intended for user 2})$
user 1 treats $x_2$ as noise $R_1<\frac{1}{2}log(1+\frac{P_1|h_1{|}^2}{P_2|h_1{|}^2+N_o})$
user 2 decodes $x_1$ and cancel it out then decode $x_2$, $R2<\frac{1}{2}log(1+\frac{P_2|h_2{|}^2}{N_o}$
visual super position coding(user 1 and user 2 both use QPSK):

for the orthogonal case, suppose a fraction of bandwidth is user 1 and $1-\alpha$ fraction is allocated for user2
rate of user 1: $R_1=\frac{\alpha }{2}log(1+\frac{P_1|h_1{|}^2}{\alpha N_o})$
rate of user 2: $\frac{1-\alpha }{2}log(1+\frac{P_2|h_2{|}^2}{(1-\alpha )N_o})$

In general, with the ordering
$|h_1|\le |h_2|\le ...\le |h_k|$
$x=x_1+x_2+...+x_k$

Then the boundary of the achievable capacity region of the k-user downlink AWGN is given by the parameterized rate tuple
$$R_k=\frac{1}{2}log(1+\frac{P_k|h_k{|}^2}{N_o+({\sum }_{j=1}^k{P}_j)|h_k{|}^2})k=1,2,...,K$$
where $P={\sum }_{j=1}^k{P}_j$ is the power splits among the users

Multiuser fading channels

The single user case:
Two types of fading $\{\begin{array}{l}\text{1) slow fading}\\ \text{2) fast fading}\end{array}$
generic model for fading $y=hx+n,h:\text{channel gain}$

typical model for the distribution of $h$ is the Rayleigh distribution

The received SNR: $|h{|}^{2}SNR-(transmitSNR)$
R: transmission Rate
Communication is successful is $R<\frac{1}{2}log(1+|h{|}^{2}SNR)$

probability of outage: P o u t ( R ) = P { 1 2 l o g ( 1 + ∣ h ∣ 2 S N R ) ≤ R } P_{out}(R)=P\{\frac 12log(1+|h|^2SNR)\le R\} Pout​(R)=P{21​log(1+∣h∣2SNR)≤R} — depends on the distribution of h

For example: for Rayleigh fading channel(pdf of $h=\frac{h}{2}{e}^{-\frac{h^2}{2{\delta }^2}},\delta$ is the parameter of fading), $P_{out}(R)=1-exp(-\frac{2^R-1}{SNR})$
at high SNR, $P_{out}(R)\sim \frac{2^R-1}{SNR}$(decays out $\frac{1}{SNR}$)

Notions of capacity:$\epsilon$ — outage capacity $P_{out}(R)\le \epsilon$
Let F be the cdf of $|h{|}^2$, then outage =$\frac{1}{2}log(1+F^{-1}(1-\epsilon )SNR)$

Another notion of capacity ergodic capacity ${\int }_n\frac{1}{2}log(1+|h{|}^{2}SNR)f_H(h)dh$
$f_H(h)$: pdf of h

distribution of h across time is $f_H(h)$ ergodic capacity: average of capacity across time as the # of time-frames grows large

Slow fading: the communication delay (time frame) is relatively short compared with channel coherence time

In fast fading: $y=hx+n$
$h:$ i.i.d for each symbol transmission

for transmitting L symbols, outage probability P o u t ( R ) = P { 1 L ∑ l = 1 L 1 2 l o g ( 1 + ∣ h l ∣ 2 S N R ) < R } P_{out}(R)=P\{\frac1L\sum_{l=1}^L\frac12log(1+|h_l|^2SNR)<R\} Pout​(R)=P{L1​∑l=1L​21​log(1+∣hl​∣2SNR)<R}

uplink fading channel:
slow fading, suppose we have k users
P o u t u p l i n k ( R ) = P { 1 2 l o g ( 1 + S N R ∑ k ∈ S ∣ h k ∣ 2 ) < ∣ s ∣ R    f o r    s o m e    S C    { 1 , 2 , . . . , k } } P_{out}^{uplink}(R)=P\{\frac12log(1+SNR\sum_{k\in S}|h_k|^2)<|s|R \;for\; some \;SC \;\{1,2,...,k\}\} Poutuplink​(R)=P{21​log(1+SNRk∈S∑​∣hk​∣2)<∣s∣RforsomeSC{1,2,...,k}}
suppose all users are transmitting at rate $R$
At low SNR, this can be approximates
P o u t u p l i n k = P { 1 2 ∣ h k ∣ 2 S N R < R    f o r    s o m e    k ∈ { 1 , 2 , . . . , k } } ≈ k P o u t P_{out}^{uplink}=P\{\frac12|h_k|^2SNR<R \;for\; some \;k\in\{1,2,...,k\}\}\approx kP_{out} Poutuplink​=P{21​∣hk​∣2SNR<Rforsomek∈{1,2,...,k}}≈kPout​
Ergodic capacity (for both slow and fast fading)
$$\begin{gathered} C_{sum}=E[\frac{1}{2}log(1+\sum _{k=1}^k|h_k{|}^{2}P)] \\ \sum _{k=1}^k|h_k{|}^{2}P:SNR,N_o \\ {R}_1+R_2+...+R_k<\frac{1}{2}log(1+\sum _{k=1}^k|h_k|P_o) \\ \sum _{k=1}^k|h_k|P_o:effectiveWRofthesuperuser \end{gathered}$$

How does this compose with the sum-capacity of uplink without fading?
(Assuming $E[|h_k{|}^2]=1$)
By Jensen’s inequality:
$$E[log(1+\frac{{\sum }_{k=1}^k|h_k{|}^{2}P}{N_o})]\le log(1+\frac{{\sum }_{k=1}^k|h_k{|}^{2}P}{N_o})=log(1+\frac{kP}{N_o})$$
$log(1+\frac{kP}{N_o})$: sum-capacity without fading

未完，接Advances in Wireless Communication 课堂笔记（下）