Sample

solution for the sample exam for FIT9135 Data Communications

S1 2016

Part B: Short Answer Questions (80 marks total)

Each question in the exam will have around 10 marks (some may have 9, some may have 12).

Question 1 (Fundamentals)
Answer the following:
a) Which layer in the OSI model is responsible for routing messages between different
networks?

The network layer

b) Which layers in the Internet model are responsible for error detection?

The data link and transport layers

m
er as
c) Describe, using a diagram, a client-server architecture.

co
eH w
o.
Client
rs e Server
ou urc
o

Presentation logic Data access logic

aC s

Application logic Data storage

vi y re
ed d

d) Explain the security implications of the SMTP email protocol.

ar stu

SMTP does not require authentication, which means that anyone can connect to an open
SMTP server and send an email to anybody else. The email FROM address is not verified,
is

so it is easy to impersonate other users.

SMTP can be transmitted through several intermediate servers, and each server could
Th

intercept or even modify the message.

SMTP can be transmitted over an unencrypted TCP/IP connection, which means that every
device on the entire circuit between sender and receiver can eavesdrop. Most SMTP servers
nowadays use SSL/TLS to encrypt the connection.
sh

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 Question 2 (Network architecture)
Consider the following network diagram:

e0
Router 3
Backbone e1

Router 1 Building C
e1

e0 Internet
Client 4

Client 3 server 1 server 2

www.fit9135.org mail.fit9135.org

m
Client 2 Client 6

er as
e1

co
e0

eH w
Router 2

o.
Client 1 rs e
Building A Building B Client 5
ou urc
The following table contains the hardware and network addresses of the devices:
o

Device Hardware Address Network Address

aC s

Client 1 00:05:69:AC:A7:EB 108.172.38.41

vi y re

Client 2 00:0C:29:BD:2F:03 108.172.38.42

Client 3 00:0C:29:9C:B7:07 108.172.38.43
Client 4 00:16:3E:F7:51:B0 108.172.38.44
ed d

Client 5 00:16:3E:29:A6:FC 108.172.35.25

ar stu

Client 6 00:16:3E:F7:51:B0 108.172.35.26

Server 1 00:50:56:3E:BA:DA 108.172.150.111
Server 2 00:50:56:92:F7:E5 108.172.150.112
Router 1 (e0) 00:16:3E:18:1E:96 108.172.38.101
is

(e1) 00:16:3E:01:9B:02 108.172.150.102

Th

Router 2 (e0) 00:16:3E:CD:22:01 108.172.35.103

(e1) 00:16:3E:FB:6C:D2 108.172.150.104
Router 3 (e0) 00:50:56:12:AE:14 108.172.150.105
(e1) 00:50:56:FB:0F:29 108.172.100.106
sh

a) What is the network address of building A, assuming the subnetmask 255.255.255.0?

108.172.38.0

b) What type of backbone network architecture is used here?

A routed backbone

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 c) Suppose that Client 2 (in building A) connects to Client 6 (in building B) using the telnet
protocol.
Assume that:
• All routers and servers involved know all network layer and data link layer
addresses.
• All clients involved know all network layer addresses (including their own and the
address of their gateway router), but do not know the data link layer addresses yet.
List and sketch the data link layer frames required to be exchanged between the two
computers and the routers to establish the telnet connection. For simplicity use only the last
byte of the network and data link layer addresses. Sketch the frames in the following
format:

Layer-2 destination, Layer-2 source, Layer-3 destination, Layer-3 source, payload

You do not need to give the detailed contents of the payload, a simple description of two or
three words is enough (such as “TCP SYN” or “HTTP 1.0 OK”).

m
This is an example including the actual telnet login. In the final exam, you won’t be asked to

er as
remember how exactly the protocol works.

co
eH w
broadcast,03,ARP "Who has 108.172.38.101"

o.
03,96,ARP "108.172.38.101 is 00:16:3E:18:1E:96"
rs e
96,03,108.172.35.26,108.172.38.42,TCP SYN
ou urc
D2,02,108.172.35.26,108.172.38.42,TCP SYN
B0,01,108.172.35.26,108.172.38.42,TCP SYN
broadcast,B0,ARP "Who has 108.172.35.103"
o

B0,01,ARP "108.172.35.103 is 00:16:3E:CD:22:01"

aC s

01,B0,108.172.38.42,108.172.35.26,TCP SYN/ACK
vi y re

02,D2,108.172.38.42,108.172.35.26,TCP SYN/ACK
03,96,108.172.38.42,108.172.35.26,TCP SYN/ACK
96,03,108.172.35.26,108.172.38.42,TCP ACK
D2,02,108.172.35.26,108.172.38.42,TCP ACK
ed d

B0,01,108.172.35.26,108.172.38.42,TCP ACK
ar stu

01,B0,108.172.38.42,108.172.35.26,TCP "Login:"
02,D2,108.172.38.42,108.172.35.26,TCP "Login:"
03,96,108.172.38.42,108.172.35.26,TCP "Login:"
96,03,108.172.35.26,108.172.38.42,TCP "guidot"
is

D2,02,108.172.35.26,108.172.38.42,TCP "guidot"
Th

B0,01,108.172.35.26,108.172.38.42,TCP "guidot"
01,B0,108.172.38.42,108.172.35.26,TCP "Password:"
02,D2,108.172.38.42,108.172.35.26,TCP "Password:"
03,96,108.172.38.42,108.172.35.26,TCP "Password:"
sh

96,03,108.172.35.26,108.172.38.42,TCP "secret"
D2,02,108.172.35.26,108.172.38.42,TCP "secret"
B0,01,108.172.35.26,108.172.38.42,TCP "secret"

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 Question 3 (Addressing)

Use the Moodle quiz to revise your knowledge on addressing and subnetting.

Question 4 (TCP/IP)
Regarding the network and transport layers of the TCP/IP protocol suite, answer the following:
a) Why does the TCP/IP protocol stack provide both TCP and UDP as transport layer
protocols?

TCP is a connection-oriented protocol, which provides a reliable communications channel

using continuous ARQ and automatic segmentation of large messages. UDP on the other
hand provides connectionless, unreliable messaging, but has the advantage that its header is
much smaller because it needs to transmit less information. We use UDP when only short
messages are needed or in real-time situations where error correction using ARQ would be
useless.

m
er as
b) Explain the main differences between the OSPF and RIP routing protocols.

co
eH w
OSPF is a link-state protocol, i.e. it uses information about the quality of a link (e.g.

o.
bandwidth) to make routing decisions, and OSPF routers exchange enough information so
rs e
that every router can build a topological map of the entire network and perform a shortest-
ou urc
path algorithm for routing messages.
RIP, on the other hand, uses distance-vector routing, i.e., it uses only the number of hops as
the distance metric. Routers exchange their routing tables instead of the link-state database,
o

and therefore don’t have a topological map of the entire network.

aC s

Both protocols are internal routing protocols.

vi y re
ed d

c) When would you use static routing?

ar stu

In small networks that are easy to configure and don’t change often.
is
Th

d) Explain two advantages of a Virtual LAN switch compared to a router for implementing a
backbone network.

1. VLAN switches are faster, i.e., have lower latency than routers. This is important for
sh

applications like real-time videoconferencing or high-performance database and file

servers.
2. Using VLAN, we can place devices on the same network based on their function, not their
location. E.g., a server can be in a dedicated server room with air conditioning, but still be
placed inside the same subnet as the clients it is serving.

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 e) Describe briefly how TCP determines the optimal transmission speed.

TCP starts sending single messages, waiting for each one to be acknowledged. If that is
successful, it increases the transmission speed (sending two, four etc messages before
waiting for an acknowledgement). When no ACK arrives, the protocol assumes that some
packets had to be dropped because the link capacity was reached. It then drops back to a
slower speed. This process is performed continually throughout the entire session.

m
er as
co
eH w
o.
rs e
ou urc
o
aC s
vi y re
ed d
ar stu
is
Th
sh

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 Question 5 (Wireless LAN)
a) Name two standards used for securing wireless networks, and indicate which one of the two
should be used and why.

WEP and WPA2. WEP should not be used any more because it is not secure, its encryption
can be broken within a few seconds. WPA2 is much more secure, in particular if the
password used for encryption is strong. Therefore WPA2 should be used in practice.

b) Assume that you have to provide Wireless LAN access for a two-story office building that is
50 meters long and 25 meters wide. An access point can cover an area of 20 by 20 meters.
i. How many access points do you need to cover the whole building?

ii. Which channels are available, assuming the AP uses IEE802.11n 2.4 GHz?

There are 13 channels, but only channels 1, 6 and 11 should be used to avoid

m
overlap.

er as
co
eH w
iii. If all devices that are planned to be used in the wireless network are compatible with

o.
IEEE802.11n 5 GHz, what would be reasons for switching to that standard?
rs e
ou urc
The 5 GHz band provides much more bandwidth and therefore more channels to
choose from, which avoids interference from neighbouring APs. In the 5 GHz band,
we can also bundle two channels together, effectively more than doubling the
o

potential transmission speed, without using up all the spectrum for neighbouring
aC s

APs.
vi y re
ed d
ar stu
is
Th
sh

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 Question 6 (Network Security)
Answer the following:
a) Explain briefly how a TCP SYN-flood attack works.

The attacker sends lots of TCP SYN packets to the target, to which the target replies with a
SYN/ACK, but the attacker never sends an ACK. The target needs to allocate memory for the
pending connection, which means that the server either crashes when it runs out of memory,
or it has to stop accepting new connections above a certain limit. In both cases, the server
becomes unavailable. This makes TCP SYN flooding a denial-of-service attack.

m
er as
co
eH w
b) Explain the three primary goals in network security.

o.
rs e
Confidentiality – protect data from unauthorised disclosure (e.g. credit card numbers).
ou urc
Integrity – protect data from unauthorised modification or destruction (e.g. bank balance,
student records).
Availability – ensure continuous operation of systems so that they are available for
o

customers and employees.

aC s
vi y re

c) Provide a list of at least six assets that should be secured in a data communications network.
ed d
ar stu

1. File server with application software licenses

2. Database with customer records
3. Client computers
4. WiFi access points
is

5. Racks with backbone switches

Th

6. Company web server

sh

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/
 Question 7 (The Internet)
Answer the following:
a) Explain how the Internet is a “network of networks”.

The internet is composed of so-called Autonomous Systems, which are networks operated
and owned by individual organisations such as ISPs, universities or corporations. All ASs
use the same network protocols (TCP/IP) and all devices must have globally unique
addresses. Each AS is independent in the sense that messages within the AS are routed
internally, and the AS can operate completely autonomously. The Internet is a network of
networks because ASs are connected via their border gateway routers, which can route
messages through other ASs to the final destination.

b) Describe the concept of an Internet Exchange Point (IXP).

m
er as
An IXP is essentially a high-speed switch where different ASs are connected with one

co
another. Typically, these ASs belong to Internet Service Providers, who have mutual

eH w
agreements to exchange traffic. IXPs often don’t charge for this service, but are instead

o.
owned and operated by the participating IXPs.
rs e
ou urc
o
aC s
vi y re
ed d
ar stu

c) Explain how ADSL works, in particular what the “A” stands for and why it needs to be “A”.

ADSL stands for Asymmetric Digital Subscriber Line. It is a common technology for
accessing the internet. ADSL uses the existing twisted pair copper telephone lines. The
is

customer uses a DSL modem, and the other end of the telephone line is connected to a
Th

DSLAM at e.g. the telephone exchange.

DSL is based on frequency division multiplexing, where a small part of the spectrum is used
for voice calls, a larger part is used for the downstream, and most of the bandwidth is
reserved for the upstream (therefore “asymmetric”). The reason for the asymmetry is that
sh

cross talk at the DSLAM is very high (far away from the transmitter, and lots of cables close
together), which limits the possible transmission rate.

*** END OF THE EXAM ***

https://www.coursehero.com/file/30844401/Sample-exam-sample-solution-FIT9135-S1-2016pdf/

Powered by TCPDF (www.tcpdf.org)