INTRODUCTION TO WIRELESS AND CELLULAR COMMUNICATIONS

NPTEL JUL – NOV 2019

Assignment – 5
1. A mobile communication system uses a DBPSK modulation scheme (Tx sends 0 at the
beginning of the transmission). Bit 0 is transmitted with phase and Bit 1 is transmitted with
phase. If the following binary stream of data 0,0,1,1,0,1 has to be conveyed to the receiver.
Which of the following is the correct bit stream that is sent to the transmitter after the
Differential modulation?
a. 0,0,0,1,0,0,1
b. 0,0,0,1,0,1,1
c. 1,1,1,0,1,1,0
d. 1,1,1,0,1,0,0
Answer: (a)

Let the differentially modulated bit stream be represented by {𝑏𝑘 }

𝑏0 = 0 , 𝑏𝑘 = 𝑏𝑘−1 ik

𝑏1 = 00 = 0
𝑏2 = 00 = 0
𝑏3 = 01 = 1
𝑏4 = 11 = 0
𝑏5 = 00 = 0
𝑏6 = 01 = 1

2. Consider a system that uses coherent QPSK modulation and detection scheme with 20
us symbol period for communication. The channel has a coherence time of 10 ms. If 100
symbols are being used for the purpose of channel estimation find the data rate?
a. 160 kbps
b. 100 kbps
c. 80 kbps
d. 40 kbps
Answer: (c)

Solution:

Symbol period = 20us implies bandwidth = 50kHz

Because of QPSK, total bit rate = 2*50 = 100kbps

But every 10ms, 100 symbols (=200bits; because of QPSK) are being used for channel
estimation and are unavailable for data transmission.
So, 200bits/20ms = 20kbps is used for channel estimation.

Data rate = 100kbps-20kbps = 80kbps

3. Which of the following can result in catastrophic error propagation?

a. Coherent modulation and coherent detection.
b. Differential modulation and differential detection.
c. Differential detection and coherent modulation.
 d. Coherent modulation and differential detection.
Answer: (d)

Solution:

As discussed in lecture videos, coherent modulation and differential detection leads to

catastrophic error propagation. That is, if there is an error in detection of 1 bit, because
of considering differences while demodulation, the error will propagate to future bits
catastrophically.

4. In QPSK, assuming 𝐸𝑏 /𝑁0 =8dB, evaluate 𝑃𝑏 (BER) using exact expression.

a. 3.17 × 10−5
b. 1.92 × 10−4
c. 2.87 × 10−7
d. 8.16 × 10−4
Answer: (b)

Solution:

Assuming Grey coding, BER for QPSK is given as,

2𝐸𝑏
𝐵𝐸𝑅 = 𝑄 (√ ) = 𝑄 (√2 ∗ 100.8 ) = 1.9 × 10−4
𝑁0

5. In the previous question, find the 𝑃𝑠 (SER) using exact expression. One QPSK symbol
consists of two BPSK bits. Use this definition and probability concepts to calculate the
Symbol error rate.
a. 6.34 × 10−5
b. 3.84 × 10−4
c. 5.74 × 10−7
d. 1.63 × 10−3
Answer: (b)

Solution:

P(symbol is in error) = P(first bit in error) + P(second bit in error) – P(both bit in error)
= P(first bit in error) + P(second bit in error) – P(first bit in error)* P(second bit in error)

So,
2
2𝐸𝑏 2𝐸𝑏
𝑆𝐸𝑅 = 2𝑄 (√ ) − 𝑄 (√ ) = 3.8 × 10−4
𝑁0 𝑁0

6. If 𝑃𝑒 denotes the probability of bit error then the probability of bit correctly recovered is (1 −
𝑃𝑒 ). What is the probability that two independent bits are recovered without any error?
a. 1 − 𝑃𝑒2
b. 1 − 2𝑃𝑒
 c. 𝑃𝑒 (1 − 𝑃𝑒 )
d. (1 − 𝑃𝑒 )2
Answer: (d)

Solution:

P(both bits received correctly)

= P(first bit received correctly)* P(second bit received correctly)
= (1 − 𝑃𝑒 )2

7. Calculate the BER of Differential BPSK if SNR = 8 dB. Assume only AWGN with no fading.
a. 8.5 × 10−4
b. 1.92 × 10−4
c. 9.1 × 10−4
d. 1.5 × 10−2
Answer: (c)

Solution:

BER for DBPSK is,

𝐵𝐸𝑅 = 0.5 exp(−𝑆𝑁𝑅) = 9.1 × 10−4

8. Calculate the BER of Coherent BFSK if average SNR = 10dB. Assume Rayleigh fading
environment.
a. 0.015
b. 0.044
c. 0.050
d. 7.83× 10−4
Answer: (b)

BER of coherent BFSK assuming Rayleigh fading is,

Γ
𝐵𝐸𝑅 = 0.5 (1 − √ ) = 0.044
2+Γ
Where, Γ is SNR.

9. What is the Rice factor (k) for a channel with no Line of Sight (LOS) component?
a. 0
b. 1
c. 2
d. 3
Answer: (a)

Solution:

Rice factor = Power in LOS component/Power in NLOS component

Since, there is no LOS component, k=0.

10. How does the Rice factor (k) change with a decrease in the power of Line of Sight (LOS)
component?
 a. Increase
b. Decrease
c. Remains constant
d. Cannot say
Answer: (b)

Solution:

Rice factor = Power in LOS component/Power in NLOS component

So, k will decrease with decrease in the power of Line of Sight (LOS) component

11. Which of the following gives rise to statistics similar to that of a Rayleigh pdf?
a. Rician distribution with rice factor (k) = 0
b. Rician distribution with rice factor (k) = 1
c. Nakagami-m distribution with m = 1/2
d. Both (a) and (c)
Answer: (a)

Rayleigh fading is Rice fading with no LOS component, so Rician with k=0 is the
correct answer.
Also, Nakagami-m with m=1 is Rayleigh fading (𝑚 = (𝑘 + 1)2 /(2𝑘 + 1) = 1).

12. What is the most appropriate multipath model for cellular transmission, where a dominant
line-of-sight plus many weak reflections are present?
a. Rayleigh fading
b. Rician fading
c. Nakagami fading
d. None of the above
Answer: (b)

Solution:

Rician fading is suitable for such a environment.

13. Consider a GMSK system with SNR = 15 dB. What should be the SNR in dB of a BPSK
system such that the Bit Error Rate remains same as that of the GMSK system? Assume
AWGN channel.
a. 8.45
b. 9.65
c. 10.21
d. 15
Answer: (c) (not considered for evaluation)

Solution:

Wrong options; not considered for evaluation.

For a given BER,

𝑆𝑁𝑅𝐺𝑀𝑆𝐾 = 𝑆𝑁𝑅𝐵𝑃𝑆𝐾 + 1.55𝑑𝐵
This can be derived from the BER expressions as explained in the lectures.
So the answer should be 13.45dB.
14. An m x n interleaver takes mn bits as input (code-bits), inserts them column-wise into an
m x n matrix, and the output bits (Tx bits) are produced by reading the elements of this
matrix row-wise. Consider a 4 x 5 interleaver at the transmitter. Which of the following
blocks can deinterleave (reverse-process of interleaving) the Tx bits at the receiver.
a. Any i x j interleaver such that ij = 20
b. 4 x 5 interleaver
c. 5 x 4 interleaver
d. None of the above
Answer: (c)

Solution:

If the interleaver is of size (𝑖 × 𝑗) , a (𝑗 × 𝑖) interleaver can act as a de-interleaver.

15. Let’s say 12 code-bits (c0, c1, c2, ..., c11) are passed through a 3 x 4 interleaver and the
output bits (t0, t1, t2, ..., t11) are transmitted using BPSK modulation. Due to momentary
bad-channel condition, bits t2, t3, t4 are corrupted by the channel. What are the code-bits
that are obtained incorrectly at the receiver after deinterleaving?
a. c2, c3, c4
b. c1, c4, c8
c. c1, c4, c7
d. c6, c9, c1
Answer: (d)

Solution:

𝑐0 𝑐3 𝑐6 𝑐9 𝑡0 𝑡1 𝑡2 𝑡3
[𝑐1 𝑐4 𝑐7 𝑐10 ] = [𝑡4 𝑡5 𝑡6 𝑡7 ]
𝑐2 𝑐5 𝑐8 𝑐11 𝑡8 𝑡9 𝑡10 𝑡11

(𝑡2 , 𝑡3 , 𝑡4 ) = (𝑐6 , 𝑐9 , 𝑐1 )

So, the answer is (c).

16. A t-error correcting (n, k) FEC (Forward Error Correction) decoder takes in ‘n’ code-bits at
atime and corrects upto ‘t’ bit errors, if present in the n-bit codeword. An m-bit burst error
occurs when the channel occasionally goes into a fading below acceptable level,
corrupting m consecutive bits. Consider a 2-error correcting (7,4) FEC decoder at the
receiver. Let’s say 28 bits were transmitted into the channel and received at the receiver.
What is the maximum length of burst error that can be definitely corrected at the receiver,
if there is no interleaver?
a. 28
b. 8
c. 14
d. 2
Answer: (d)

Solution:
 As there is no interleaver, a 2-error correctinf (7,4) FEC can only correct a burst error
of length 2 (max.), by the definition.

17. How will the answer change for the above question if we introduce a 7 x 4 interleaver at
the transmitter and corresponding deinterleaver at the receiver? What will be the
maximum length of burst error that can surely be corrected at the Rx now?
a. 28
b. 8
c. 14
d. 2
Answer: (b)

Solution:

Let the transmitted bits be (𝑡0 , 𝑡1 , … , 𝑡27 ). When it is passed through the de-interleaver,
(Write it into the columns of 4 × 7 matrix and read it out from rows). The de-interleaver
output will be (𝑡0 , 𝑡4 , 𝑡8 , 𝑡12 , 𝑡16 , 𝑡20 , 𝑡24 , 𝑡1 , … ) .
One can see that if there is a burst error of length 8 (Say (𝑡0 , 𝑡1 , … , 𝑡7 ) are in error), the
8 errors are distributed in blocks of 2 ((𝑡0 , 𝑡4 ),(𝑡1 , 𝑡5 ),(𝑡2 , 𝑡6 ),(𝑡3 , 𝑡7 )), each of which can
be corrected by FEC.

18. A BPSK modulation transmits a symbol ‘+1’ or ‘-1’ in one period. A constellation diagram
is a plot of the complex modulated symbols in the complex z-plane (real-part/In-Phase vs
imaginary-part/Quadrature). Shown below is the constellation of 100 BPSK symbols (+1,-
1,-1,+1,-1, …) that were sent into the channel at the transmitter.

How will this constellation look for the 100 symbols at the receiver, if these BPSK symbols
are affected by a Doppler shift? (Assume that Doppler shift is the only impairment => No
channel/other impairments, not even path-loss!). Which of the below plots best represents
the effect of Doppler shift on BPSK?
a.

b.

c.

d.
Answer: (c)

Solution:

No Path-Loss + No noise -> All the received symbols will have the same power
 They all will lie on the unit circle

Doppler introduces phase shift on the unit circle.

So the answer is c)