Birla Institute of Technology & Science, Pilani

First Semester 2009-2010

Note:
1. First attempt PART-A. Collect a separate answer book to write PART-B after
submitting PART-A answer book. There is no individual time limit for PART-A and
PART-B. Mode of examination is Close Book.
2. Answer all sub parts of the questions all together.

Expected Time: 90 min PART-A Max Marks: 40

Q.1 Answer the following. [Transport Layer and Network Layer] [5x3=15M]
i) An organization is granted the block 211.17.180.0/24. The administrator wants to
create 32 subnets.
a) What is the subnet mask in x.y.z.t/n notation?
b) What is the range of addresses in first and last subnet?
c) How many addresses are in each subnet?

ii) You are hired to design a reliable byte-stream protocol that uses a sliding window
like TCP. This protocol will run over a 100-Mbps network. The RTT of the network
100 ms, and the maximum segment lifetime is 60 seconds. How many bits would
you include in the Advertised Window field of your protocol header?

iii) Suppose one IPv6 router wants to send a datagram to another IPv6 router, but
the two are connected together via an intervening IPv4 router. How the datagram
will be sent using tunneling?

iv) Draw the multicast tree for the network given in Fig.1 using RPF (Reverse Path
Forwarding) algorithm.

Fig. 1

v) In the network shown below in Fig.2, we would like a TCP connection to fully
utilize the bottleneck link. What should we make the size of buffer B? Assume that
the advertised receiver window is very large. The link latencies given are one-way.

Fig.2

Page 1 of 4
Q.2 Answer the following. [Link Layer] [5x3=15M]

i) A node using Hamming error detection with even parity received the following bit
stream: 111001001111  (Bit no. one). What were the data bits sent by the sender?
Assume that not more than one bit is in error.

ii) Write four desirable characteristics of a broadcast channel. Which of these

iii) In CSMA/CD, after the fifth collision, what is the probability that a node chooses a
random value k=4? What will be the corresponding delay (in sec) for k=4 on a 10
Mbps Ethernet.

iv) Assume that hosts A and B have a TCP connection established. Assume that the
two hosts are separate by one router (i.e., they are one hop apart). Why does host A
not directly use the MAC address of host B when constructing its packets to send to
host B?

v) Consider six wireless stations, A through F, arranged in a straight line in the order
A to F, communicating using the MACA (Multiple Access using Collision Avoidance)
protocol. Assume that each station can reach only to its nearest neighbors. How
many stations can transmit simultaneously? Justify your answer.

Q.3 Answer the following. [Physical and Application Layer] [5x2=10M]

i) Draw the time amplitude graph of the Manchester and Differential Manchester
encoding schemes for the bit stream 00101100.

ii) In a digital transmission, the sender clock is 0.3 percent faster than the receiver
clock. How many extra bits per second does the sender send if the data rate is 2
Mbps?

iii) Consider a noiseless channel with a bandwidth of 3 KHz is transmitting at 12,000

iv) How does an email client discover the IP address of an email server for bits-
pilani.ac.in? Mention the protocol and the query parameters.

v) What is the common characteristic of Virtual Circuit network (e.g., ATM) and
circuit-switched network (e.g. Telephone Network)? What is the main difference
between them?

End of Part A

Page 2 of 4
 Expected Time: 90 min PART-B Max Marks: 40

Q.1 The following frame transition diagram (Fig.1) shows an exchange of Ethernet
frames between two computers, A and B connected via a 10BaseT hub. Each frame
sent by computer A contains 1500 Byte of Ethernet payload data, while each frame
sent by computer B contains 40 Byte of Ethernet payload data. [3+1+2=6M]

Fig.1
a) Calculate the average utilization of the media during this exchange.

b) Is the exchange of frames best described as Full Duplex, Half Duplex, or Simplex?

c) What is the throughput of the transfer from A to B measured at the TCP layer?

Q.2 In a network that has a maximum packet size of 128 bytes, a maximum
packet lifetime of 30 seconds, and an 8-bit sequence number, what is the maximum
data rate per connection? [4M]

Q.3 Consider three RTT samples (in milli sec): 150, 200 and 210 in that order. Assume
initial RTT= 200 ms, initial DevRTT = 50 ms, β = 0.25 and α = 0.125.
Answer the following questions.
a) Show the initial Timeout (that is, the value before processing the three samples.)

b) Give the RTT and Timeout after processing the three sample RTTs. [2+4=6M]

Q.4 Bridges are used to interconnect LAN segments. Use of multiple bridges can be lead
to loops. The Fig.2 shows the interconnection of six LAN segments (LAN A, LAN B,
LAN C, LAN D, LAN E, and LAN F) using five bridges (B1, B2, B3, B4, and B5).
Run the spanning tree algorithm and identify the route-bridge, route ports and
designated ports for the given network. [6M]

Fig. 2
Page 3 of 4
Q.5 Consider a source A that uses TCP/Reno to transfer a file of 9 KB to destination B.
Each packet has a payload of 1KB. Assume that only two packets are lost during the
transfer: the 4th data packet and the acknowledgment of the 6th data p acket. Draw
the time diagram of the entire file transfer. Assume that retransmission timeout
(RTO) is 4*RTT. How many RTTs does it take to transfer the entire file? Ignore the
connection setup, processing and transmission times. [7M]

Q.6 a) The Fig.3 shows a network with 8 routers, where each link has an integer weight
shown above the link. Apply the Dijkstra routing algorithm to find the shortest paths
from the node u to all the remaining nodes. Show each step clearly.

Fig.3

b) Suppose node u sends a large amount of traffic to node v, consuming most of the
bandwidth on the (u,v) link. The traffic from u to w also traverses the (u,v) link. To
alleviate congestion on the (u,v) edge, the network administrator plans to change
the weight of the (v,w) edge to divert the traffic from u to w onto another path.
What is the minimum weight the (v,w) edge would need to force all of the traffic
from u to w to take a different path? What is the maximum weight the link could
have while still ensuring the traffic from v to w goes directly over the (v,w) link?

c) Why would a network administrator change the link weight, rather than simply
relying on the end hosts (connected to node u) to adapt their sending rates in the
presence of congestion? [4+2+2=8M]

Q.7 Consider a router inside one AS that receives a packet for a network inside another
AS. How the router decides which router to send it to in its own AS so that it can
reach the other AS. [3M]

---------------------------------

Page 4 of 4