Final Examination,

2013 / 2014
Module Title Introduction to Computer Networks and Communications

Module Leader Dr. Amal ElNahas Semester

One
Equipment allowed (for example
calculator – dictionaries) Calculators

Instructions to Students

 You should attempt all questions

 The exam paper is 3 pages long, and is in 1 section.
 The approximate allocation of marks is shown in brackets by the
questions.

This examination is ____2_____ hours long.

[Turn over]
Q1 Consider a host A that wants to send a 1 Mbyte packet to host B.
Assume that A and B are connected via router R. Link AR connects A
to R, and link RB connects R to B. Link AR is 1 km long and link RB
is 2 km long. Suppose the capacity of each link is 10 Mbytes/s, the
propagation speed is 2*10*8 m/s and the processing delay at R is 10
msec. Find after how long will host B receive the packet.

Solution:
Total delay = delay from A to R + delay from R to B.
Delay from A to R = transmission delay + propagation delay
= 1/10 + 1000/2*10^8
= 100.005 msec.
Delay at router R = 10 mec (given)
Delay from R to B = 1/10 + 2000/2*10^8
= 100.01 msec.

Total delay = 100.005 + 10 + 100.01 = 210.015 msec.

[Total: 10 marks]

Q 2 An Internet Mail server accepts incoming TCP packets destined to

port 25.
i. How does it distinguish between different packets belonging
to different TCP sessions (remember that different sessions
can originate from the same host)? [4 marks]

Solution:
The combination of a client’s IP address and its source port
is enough to uniquely discriminate between different
sessions arriving at the same Mail server

ii. How does TCP distinguish between different packets in the

same session? [4 marks]

Solution:
Within the same session, each byte is represented by a
unique sequence number.

iii. Why is TCP well suited for applications like HTTP and POP
(email)? [4 marks]

Solution:
 These applications require reliability and do not have any
limitations with respect to timeliness of data delivery. Thus the
overhead (or delay) incurred from possible retransmissions that
help ensure reliability do not degrade the performance of the
application significantly.

iv. For what type of applications is UDP well suited?

[4 marks]
Solution:
UDP is best suited for applications that do not require
reliability, and therefore do not need to “pay” the overhead
for retransmissions. Examples include DNS and streaming
media applications.

[Total: 16 marks]

Q3 A sender and a receiver are communicating using a TCP-based

protocol in which data is sent immediately after connection. The
bandwidth of the link is 100Mbps, the round-trip time is 10ms and the
maximum packet/segment size is 1500 bytes (ignore header
overhead). Draw a packet exchange diagram showing how long it
takes to transfer 15000 bytes (consider opening and closing the
connection)
[Total: 12 marks]

Solution:
This is a TCP based protocol, so we use slow-start.
1st RTT: SYN + SYN/ACK
2nd RTT: 1500 bytes of data + ACK
3rd RTT: 3000 bytes of data + two ACKs
4th RTT: 6000 bytes of data + four ACKs
5th RTT: 3500 bytes of data + three ACKs
6th RTT: FIN handshake

Q 4 a) The following are estimates of the population of major regions

of the world: Africa 900 million; South America 500 million; North
America 400 million; East Asia 1500 million; South and Central Asia
2200 million; Russia 200 million; Europe 500 million. Suppose each
region is to be assigned 100 IP addresses per person. Is this possible?
If not, what would you recommend?
[6 marks]
 Solution:
The total number of IPv4 addresses is: 2^32 which is approximately
4.29 billion. The above world population estimate totals 6.2 billion, so
it is not possible to assign an individual address to each person. IPv6
is required to provide 100 addresses per person.

b) Consider a subnet with prefix 128.119.40.128/26.

i. Give an example of one IP address (of form
xxx.xxx.xxx.xxx) that can be assigned to this network.
[2 marks]
ii. Suppose an ISP owns the block of addresses of the form
128.119.40.64/26. Suppose it wants to create four subnets
from this block, with each block having the same number of
IP addresses. What are the prefixes (of form a.b.c.d/x) for
the four subnets?
[4 marks]

Solution:
i. Any IP address in range 128.119.40.128 to 128.119.40.191

ii. Four equal size subnets: 128.119.40.64/28, 128.119.40.80/28,

128.119.40.96/28, 128.119.40.112/28
[Total: 12 marks]

Q 5 Suppose a router receives an IP packet containing 600 data bytes

and has to forward the packet to a network with maximum
transmission unit of 200 bytes. Assume that IP header is 20 bytes
long. Show the fragments that the router creates and specify the
relevant values in each fragment header (i.e., total length, fragment
offset, and MF and DF flags).

Solution:
IP packet = 600 data bytes, MTU = 200 bytes, IP header = 20
header bytes.
Maximum possible data length per fragment = MTU – IP header =
200 – 20 = 180 bytes.
The data length of each fragment must be a multiple of eight bytes,
therefore the maximum number of data bytes that can be carried
per fragment is 22*8=176 bytes.
The data packet must be divided into 4 frames, as follows:
176 + 176 + 176 + 72 = 600 (data size)
196, 196, 196, 92 (total size)
 The sequence of frames and packet headers is shown below:
Total length Id Mf Fragment Offset
Original Packet 620 x 0 0
Fragment 1 196 x 1 0
Fragment 2 196 x 1 22
Fragment 3 196 x 1 44
Fragment 4 92 x 0 66

[Total: 16 marks]

Q 6 The nodes participating in the Link State algorithm in one network

are broadcasting the following link-state packets. Based on these
packets,

i. Draw the network topology and assign link costs.

[6 marks]

ii. Run the Link State (Dijkstra) algorithm to determine the

shortest path from D to A.
[10 marks]

[Total: 16 marks]

Solution:
Q7 Draw the signal encodings of the given bit-stream : 101110001, for
the encoding schemes RZ and NRZ-I
[18 marks]

Solution: