582202 Introduction to Data Communication (4 cr)
Liisa Marttinen

Course examination 25.2.2009
Write the name of the course, date, your name, student number and signature on every separate sheet. Books and calculators are not allowed.

1. Addresses in the Internet [20 p]

1. What kind of addresses are used in the different layers of the Internet stack? (8 p)

2. How does the upper layer find out the address that is used in the service offering layer below? In other words when the receiver address used in the layer n is known, how is the receiver address used in the layer n-1 found out? (10 p)

3. How is the DHCP protocol related to addresses? (2 p)

2. Routing [20 p]
Packets (or datagrams) are routed in the Internet from the sender machine to the receiver machine.

1. What is a routing table, where is it located and how is it used? What is a routing algorithm and why is it needed? (6 p)

2. Explain the functionality of distance vector routing. (4 p)

3. Assume that the nodes B, C, D and E of the network in the picture have already got their routing tables calculated, when node A is added to the network. Show in detail how and why the contents of the routing table of the node A are changed when distance vector routing algorithm is used. (10 p)

picture

3. Congestion control [10 p]
TCP protocol is used with the maximum segment size (MSS) of 2 kB. The threshold value is 18 kB and the receiver window size is 20 kB.

1. Show with a diagram how in the very beginning of the TCP connection a user sends 100 kB of data. Assume that acknowledgments arrive in time and the receiver does not change the receiver window size. (5 p)

2. Suppose that the segment including the 50th kilobyte of the data sent is lost, but otherwise other segments arrive correctly to the receiver. Show using a diagram how the sender continues the sending after that assuming

   1. the TCP Reno version with fast retransmission and fast recovery is used. (3 p)

   2. the TCP Tahoe version without those is used. (2 p)