Unit home

help3002

Lecture recordings
via LMS

Schedule

Unit outline

Resources

Textbooks

Extra reading

Look after yourself!

Sample mid-semester test questions

Good luck,

Chris McDonald.
March 2024.

Layered protocols

1. Name the seven layers of the OSI Networking Reference model, from the lowest to the highest.

2. List 2 advantages and 2 disadvantages of strictly conforming to a layered networking standard, such as the OSI/ISO networking model.

3. Data encryption of network traffic can, arguably, be employed in a number of layers of the OSI/ISO Reference model. If you were to employ encryption in exactly 2 layers, which layers would you choose, and why?

4. A system uses an n-layer protocol hierarchy. Applications generate messages of M bytes each. At each of the layers, an h byte header is added. What fraction of the network bandwidth is filled with headers?

Physical Layer and Data Link Layer

5. Knowing that some typical checksums are only 16 bits long, it appears that a corrupted data frame has a 1 in 2¹⁶ chance of having the same checksum as its original data frame. Discuss this idea with respect to the design of checksum algorithms.

6. Local area networks (LANs) employ sophisticated signal encoding techniques, in preference to the naive (0=low, 1=high) encoding technique. List 2 distinct reasons for employing the sophisticated encoding techniques.

7. Under what circumstances would the use of Data-Link Layer frame pipelining be unecessary (a waste of either time or effort)?

8. What is the primary motivation for piggy-backed acknowledgements in a Data-Link layer protocol?

9. In some protocol stacks, such as the Internet Protocol (IP) and Asynchronous Transfer Method (ATM) networks, the checksum of network-layer information covers only the header, and not the data. Why do you suppose that this design was chosen?

10. Consider a stop-and-wait protocol that uses only negative acknowledgments (NACKs), and no positive acknowledgments (ACKs). Describe what timeouts would need to be scheduled. Explain why an ACK-based protocol is usually preferred to a NACK-based protocol.

11. Within a sliding window Data-Link layer protocol, is it meaningful for the receiver's maximum window size to be larger than the sender's maximum window size? Why or why not?

Local Area Networks

12. What is the role of the preamble in IEEE 802 protocols?

13. Explain the need for both a minimum frame length and a maximum frame length in the IEEE 802.3 Ethernet specification.

14. Outline the difference between frequency division multiplexing and time division multiplexing.

15. Why must IEEE 802.11 wireless networks employ collision avoidance instead of simply collision detection?

16. Does the IEEE 802.3 protocol employ multi-frame pipelining? Briefly explain your answer.

17. In an effort to reduce the likelihood of intruders employing passive scanning to assess the security of an IEEE802.11 wireless network, the owner of a wireless access point enables a facility so that only registered MAC addresses may use the access point. Explain why this measure would be ineffective against a determined intruder assessing the network's security.

Network Layer

18. Given that Data Link Layer protocols already employ sequence numbers, provide 2 distinct motivations for Network Layer protocols also employing sequence numbers.

19. Why doesn't each node in a virtual circuit use the same number to identify the circuit?

20. Briefly, in a sentence or two, distinguish between the terms flow control and congestion control.

21. Briefly describe the intended effects of employing the leaky bucket and token bucket flow control algorithms.

22. As a brute-force form of congestion control, the Network Layer may choose to deliberately discard some packets when its maximum permissible memory allocation is reached. Which packets are the best candidates for removal? What effects will their removal have on the total network's congestion?