Scribd
Upload
113 views2 pages

Channel Coding Assignment 4

This document provides the details for Assignment 4 of the ECEN 489 course, including 6 channel coding problems. Problem 1 involves finding the optimal input distribution and capacity for a discrete memoryless channel. Problem 2 involves a similar analysis for a channel with additive noise. Problem 3 analyzes another channel and finding the capacity when the output is quantized. Problem 4 analyzes a channel where the output is the product of the input and noise. Problem 5 analyzes a binary symmetric channel. Problem 6 is an optional bonus problem extending Problem 5 to different rates.

Uploaded by

AGM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
113 views2 pages

Channel Coding Assignment 4

This document provides the details for Assignment 4 of the ECEN 489 course, including 6 channel coding problems. Problem 1 involves finding the optimal input distribution and capacity for a discrete memoryless channel. Problem 2 involves a similar analysis for a channel with additive noise. Problem 3 analyzes another channel and finding the capacity when the output is quantized. Problem 4 analyzes a channel where the output is the product of the input and noise. Problem 5 analyzes a binary symmetric channel. Problem 6 is an optional bonus problem extending Problem 5 to different rates.

Uploaded by

AGM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

ECEN 489 – Fall 2019

Assignment 4 – Channel Coding Problem


Due: Friday, October 4, 2019 (in class)
NOTE 1: Show your work in details.
NOTE 2: Bonus question (Problem 6) is optional.

Problems:

1. Let X and Y be two random variables such that pY |X (y|x) is given by

pY |X (0|0) = 1, pY |X (1|0) = 0, pY |X (0|1) = p, pY |X (1|1) = 1 − p,

for 0 ≤ p ≤ 1. Consider a discrete memoryless channel (DMC) with input X and output Y .

(a) 5 pt – Find an optimal input distribution, i.e., a distribution pX which maximizes the
mutual information I(X; Y ).
(b) 5 pt – Find the channel capacity.

2. Let Y = X + Z (mod 5), where the random variable X takes on values x ∈ {0, 1, . . . , 5} and
the random variable Z takes on values z ∈ {0, 1, 2}. Suppose that X and Z are independent.
Suppose that
1 1 1
pZ (0) = , pZ (1) = , pZ (2) = .
2 4 4
(a) 10 pt – For a DMC with input X and output Y , find an optimal input distribution and
the channel capacity.
(b) 10 pt – For a DMC with input X and output (Y, Z), find an optimal input distribution
and the channel capacity.

3. Let X and Y be two random variables such that pY |X (y|x) is given by

pY |X (y|x) y=0 y=1 y=2 y=3


x=0 2/3 1/3 0 0
x=1 0 2/3 1/3 0
x=2 0 0 2/3 1/3
x=3 1/3 0 0 2/3

(a) 10 pt – For a DMC with input X and output Y , find an optimal input distribution and
the channel capacity.
(b) 10 pt – Define the function g(y) such that g(y) = 0 if y ∈ {0, 1}, and g(y) = 1 if
y ∈ {2, 3}. Let Z = g(Y ). Note that Z is a random variable. For a DMC with input X
and output Z, find an optimal input distribution and the channel capacity.

1
4. Let Y = XZ, where X and Z are two independent and identically distributed (i.i.d.) random
variables, each taking on values −1 and 1.

(a) 10 pt – For a DMC with input X and output Y , find an optimal input distribution and
the channel capacity.
(b) 10 pt – For a DMC with input X and output (Y, Z), find an optimal input distribution
and the channel capacity.

5. Consider a binary symmetric channel (BSC) with input X and output Y , where

pY |X (1|0) = pY |X (0|1) = 0.1, pY |X (0|0) = pY |X (1|1) = 0.9.

Suppose that the input distribution pX maximizes the mutual information I(X; Y ). Let n =
25,  = 0.2, and R = I(X; Y ). Suppose that d2nR e codewords X n (1), X n (2), . . . , X n (d2nR e),
each of which is a binary sequence of length n, are randomly generated according to the
optimal input distribution pX .

(a) 5 pt – Find all -typical sequences xn and all -typical sequences y n .


(b) 5 pt – Find all pairs (xn , y n ) such that xn and y n are jointly -typical. How many pairs
(xn , y n ) are in the jointly typical set?
(c) 5 pt – For a fixed codeword xn (i), what is the probability that xn (i) and the random
sequence Y n (generated according to pY ) are jointly -typical?
(d) 5 pt – For a fixed received word y n , what is the probability that the random sequence
X n (generated according to pX ) and y n are jointly -typical?
(e) 10 pt – What is the average probability of error for the designed code when jointly
typical decoding is used? Can you design a different decoding that yields a smaller
average probability of error?

6. (Bonus) Repeat the parts (d)-(f) of Problem 5 for each of the following cases, and compare
the results with those in Problem 5.

(a) 10 pt – R = I(X; Y ) − 0.1.


(b) 10 pt – R = I(X; Y ) + 0.1.

You might also like