Scribd
Upload
47 views3 pages

Institut Eur Ecom: Digital Communications Midterm Examination

The document is a midterm exam for a digital communications course containing 4 problems. Problem 1 involves analyzing a signal set including finding the average energy, an orthonormal basis, constellation, and minimum distance. Problem 2 involves analyzing a quaternary signal set transmitted over an AWGN channel including average energy, decision regions, probability of error bound, and coding gain compared to QPSK. Problem 3 analyzes a 4-PAM constellation including deriving likelihood functions, approximating a detection rule, and bounding the error probability. Problem 4 involves a non-coherent binary communication problem over a fading channel and asks for the maximum likelihood receiver and a bound on the error probability.

Uploaded by

Luis Manuel Capriles
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
47 views3 pages

Institut Eur Ecom: Digital Communications Midterm Examination

The document is a midterm exam for a digital communications course containing 4 problems. Problem 1 involves analyzing a signal set including finding the average energy, an orthonormal basis, constellation, and minimum distance. Problem 2 involves analyzing a quaternary signal set transmitted over an AWGN channel including average energy, decision regions, probability of error bound, and coding gain compared to QPSK. Problem 3 analyzes a 4-PAM constellation including deriving likelihood functions, approximating a detection rule, and bounding the error probability. Problem 4 involves a non-coherent binary communication problem over a fading channel and asks for the maximum likelihood receiver and a bound on the error probability.

Uploaded by

Luis Manuel Capriles
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
You are on page 1/ 3

Institut Eurecom

Digital Communications

Midterm Examination

Date: Dec. 16, 2013

Duration: 2 hours

Answer any 3 out of 4 questions

1 Problem 1

Consider the three signals,


(
E exp (t) , t [0, T /2]
p0 (t) = , p1 (t) = 0, p2 (t) = p0 (t T /2)
0 otherwise

(a) Find the value of for which the average energy of this signal set is E.

(b) Find an orthonormal basis for this set of signals.

(c) Draw a constellation representing this signal set on the basis you found

(d) What is the minimum distance in terms of E.

2 Problem 2


Consider the quaternary set of signals, xm = Eej2m/3 , m = 0, 1, 2, x3 = 0 transmit-
ted over an AWGN channel
y = Exm + z
where z is a zero-mean circularly-symmetric complex Gaussan random variable with vari-
ance N0 .

1
(a) What is the value of for which the average signal energy is E given that all signals
are equally likely?
(b) Draw the decision regions for this constellation
(c) What is the union-bound on the probability of error for this signal set assuming an
additive white-Gaussian noise channel model with power-spectral density N0 and
normalized signal energy?
(d) What is the coding gain/loss with respect to QPSK modulation?

3 Problem 3

Consider the 4-PAM contellation in Figure 1. Here the modulated symbols are mapped to
the 2-bit sequences (b0 b1 ) as shown using so-called Gray Coding.

(a) Assuming that we observe the transmitted symbols in white Gaussian noise with
variance N0 /2 as y = sm + z, 0 m 3, write the likelihood functions p(y|b0 )
and p(y|b1 ).
(b) What is the simplest detection rule for b0 ?
(c) Use the approximation log(ea + ebp ) max(a, b) to approximate the detection rule
for b1 as b1 = .5(1 + sign(|y| (2 Es /5))).
(d) Give a numerical expression (or even a bound) for the error probability for detecting
b0 .

01 00 10 11

q q q q
3 Es /5 Es /5 Es /5 3 Es /5

Figure 1: Constellation for Problem 3

4 Problem 4

Consider the binary communication problem consisting of vectors:



   
1 z
y = Eh + 1
x z2

2
where x {1, 1} (antipodal or BPSK modulation), zi are zero-mean independent com-
plex circularly symmetric Gaussian random variables with variance N0 , and h is a zero-
mean complex circularly symmetric Gaussian random variables with variance 1.

(a) What is the ML receiver (non-coherent) for this problem?

(b) Give a bound (or exact expression) for the error probability.

You might also like