DECEMBER 2022

FINAL EXAMINATION

ECSE308F22: INTRODUCTION TO COMMUNICATION SYSTEMS AND NETWORKS

DATE & TIME OF EXAM: December 9/2022, 09:00-12:00

EXAMINER: Tho Le-Ngoc ASSOC. EXAMINER: Benoît Champagne

CLOSED BOOK ☒ OPEN BOOK ☐

PRINTED ON BOTH SIDES OF THE PAGE ☒ SINGLE-SIDED ☐

EXAM:
ANSWER BOOKLET REQUIRED: YES ☒ NO ☐
EXTRA BOOKLETS PERMITTED: YES ☒ NO ☐
ANSWER ON EXAM: YES ☐ NO ☒
SHOULD THE EXAM BE: RETURNED ☒ KEPT BY STUDENT ☐

CRIB SHEETS: PERMITTED ☒ one 8.5”x11” double-sided sheet

DICTIONARIES: TRANSLATION ☐ REGULAR ☐ NOT PERMITTED ☒

CALCULATORS: PERMITTED (Non-Programmable) ☒

ANY SPECIAL Attempt all twenty (20) questions.

INSTRUCTIONS:

1) (6 points) Describe the operations of non-persistent, 1-persistent and p-persistent CSMA. Explain how collision
detection (CD) can be implemented in Ethernet but cannot be successfully realized in WLAN, and the differences
between CSMA/CA and CSMA/CD.
2) Consider a bus system with 4 stations equally spaced at 5 m from one another. The data rate of the bus is
100Mbps and the propagation speed is 100 km/s.
a) (3 points) Calculate the total time taken to send a 15,000-bit packet from any station to its adjacent station,
measured from the beginning of transmission to the end of reception.
b) (3 points) Calculate the total time taken to send a 5,000-bit packet from a station at one end to the station
I
www.py by lines
at the other end, measured from the beginning of transmission to the end of reception.
3) (6 points) Explain the differences and similarities between the bridge and router and their operations,
advantages and disadvantages.
4) (6 points) A switching network has N =2 hops between two given end-devices, message length of L = 2000 bits,
data rate of B = 2048bps on all links, fixed packet size of P = 1024 bits including header of H = 16 bits, call setup
time of S=0.2 seconds, and propagation delay per hop of D = 0.004 seconds. Compute the end-to-end delay for
circuit switching, virtual circuit packet switching, and datagram packet switching. Assume that
acknowledgments and processing delay at the nodes are negligible.
5) (4 points) A transport-layer message consisting of 2200 bits of data and 250 bits of header is sent to a network-
layer, which appends another header of 100 bits. The resulting message (including data and headers) is then
segmented for packet transmission to the destination via two networks. The data link layer uses a 25-bit packet
header, and the maximum frame length is 1200 bits (including headers). How many bits, including headers, are
delivered to the destination? Explain your derivation in sufficient detail.

ECSE308F22- FINAL EXAM 1

 125 4 18 11 20 H
any
6) Briefly define the channel capacity and its key factors. Consider an additive white Gaussian noise (AWGN)
channel with a bandwidth of 100 kHz and an available signal-to-noise ratio (SNR) of 15dB.
a) (2 points) Using Shannon theorem, calculate the maximum theoretical transmission data rate over this
channel?
b) (2 points) If the available SNR is increased to 20dB, find the required channel bandwidth to provide the
same channel capacity found in a). Briefly comment on the result.
c) (2 points) If the bandwidth is decreased to 75kHz, find the required minimum SNR in dB to provide the
same channel capacity found in a). Briefly comment on the result.
7) (4 points) Explain the main advantages and disadvantages of the text-based protocols, e.g., HTTP, email, DNS.
8) (8 points) A network consists of 3 nodes N1, N2, N3 connected by 3 links with associated link costs: N1-N3=4,
N2-N3=1, N1-N2=2. Using Bellman–Ford algorithm to find the least-cost paths. Show how the algorithm work
by computing the node tables in various iterations.
9) (4 points) A frame of 120 bits is transmitted over a channel with a bit error probability of 10 and statistically
independent bit errors. Find the probability that the received frame contains no error and the probability that
the received frame contains at least one bit error.
10) (4 points) Encode the data pattern 111101 using a cyclic redundancy check (CRC) error-detecting scheme with

I
𝑃 𝑋 𝑋 1.
11)
A
(4 points) Briefly describe the Internet checksum operation and derive the Internet checksum for the data
segment 0011 F202 F4EE F5E6.
12) Briefly explain the differences between TDMA, Otter Ks
I FDMA. An AWGN channel with a total bandwidth of 244kHz is
77 shared by 4 users using QPSK modulation at rate of 120 kbps. Using Nyquist criterion for transmission for no
Inter-Symbol Interference (ISI), calculate the minimum bandwidth required by each user, and show
a) (3 points) the corresponding frequency division for 4 users in case of FDMA with derived frequency-slot
and frequency guard band sizes
39 12 2 60
b) (3 points) the corresponding time division for 4 users in case TDMA using a time frame of 10ms, with
derived timeslot and guard-time sizes
13) (8 points) Explain the TCP Tahoe congestion control procedure. Assuming TCP Tahoe congestion control
procedure in use with an initial very large ssthresh and the connection continuing until a timeout (indicating
congestion) at the 3rd RTT, plot the cwnd versus RTT=0,1,2,3,...,8,9.
14) (4 points) Explain the principle of frequency reuse and frequency reuse factor in the context of a cellular system.
Derive the number of channels available per cell for a frequency reuse factor of 4 and 7 in a cellular system
having a total of 33 MHz of bandwidth and using two 25-kHz one-way links to provide one two-way voice
channel.
15) (8 points) A network consists of 6 nodes: A,B,C,D,E,F connected by 10 links with associated link costs: AB=1,
AC=2, AD=5, BC=2, BD=3, BE=1, CD=3, DE=1, DF=5, EF=2 . Using Dijktra’s algorithm to find the least-cost path
from node E to all other nodes. Show how the algorithm work by computing the Dijktra table.
16) (8 points) Consider a token bucket used for traffic management with a token replenishment rate R=2 Mbps
(note: bps: bits per second) and a bucket size B=200 kbits. The coming traffic for transmission is queued until
sufficient tokens are available and then transmitted. Determine the continually sustainable data rate over a
relatively long period of time for this flow. Suppose that the bucket is initially full of tokens when traffic comes
as 5 bursts of 300 kbits, 400 kbits, 250 kbits, 200 kbits, 20 kbits in 5 consecutive 100ms-intervals. Show the
possible outputs over the 5 intervals. Justify all your answers.
17) (4 points) A single video source transmits on a 256-kbps link with a throughput of 200 kbps. The data
experiences a delay jitter of 2s. What size of delay buffer is required at the destination to eliminate the jitter?
What is the utilization over the link?
18) (4 points) For a cellular network with the pathloss exponent of 3.8, what is the required minimum frequency

I
reuse factor to maintain a signal-to-interference ratio (SIR) of 17dB or higher?
19)
v
(4 points) Explain the possible hidden and exposed station problems in WLAN and the use of RTS and CTS to
u
b
resolve these problems. Explain the reasons for NAV and its operation.
20)
a u
(4 points) Explain the reasons and operation of the binary backoff procedure in CSMA. For a round-trip time of
15 𝜇𝑠, when can the device attempt for (re)-transmission after the 4th, 11th, 16th collision?

ECSE308F22- FINAL EXAM 2