Advanced Networking

SECTION A
[ 40 marks ] Answer ALL questions in this section

A1.
[40]

(a) What is meant by full-duplex and half-duplex communication? Provide a suitable example of an application which uses full-duplex and half-duplex communication. [6] Full-duplex is the connection is bi-directional and information can flow in both directions at the same time [2] Example: telephone [2] Half-duplex is the communications that is bi-directional, but one-way at a time.[2] Example: - walkietalkie

(b)

Twenty Four voice channels are to be multiplexed and transmitted over twisted pair.

(i) What is the bandwidth required for FDM assuming 4KHz is needed per voice channel? [5] (ii) Assuming a bandwidth efficiency of 1bps/Hz and a data rate of 64Kbps per voice channel, what is the bandwidth required for TDM? [6]

(i) 4Khz * 24 [1] = 96Khz [4] (ii) (24 * 64Kbps) * 1bps/Hz [3] = 1536Hz or 1.536Khz [3]

(c)

Describe the functional specification of the RS232C? [6]

The specification is divided into 4 groups, [2] data, control, ground and timing [2] Describes the functionality of each pin is described for every group. [2]

(d) Noise cause distortion is data communications. Most types of noise can be avoided, however, there is one type that cannot be eliminated? Name and explain this type of noise.[2] Thermal noise[1] . It is caused by random motion of electrons [1] and it is a function of temperature[1]

(e) Explain what is meant by delay distortion. What problems can this type of impairment cause? How is this problem overcome?[9] The propagation velocity varies on the media causing different components of the same signal to be received at different times[3] It may cause signals to be misinterpreted by the receiver[2](errors) It can be overcome by equalizing the velocity throughout the band.[2] This can be achieved by using equalizers [2]

(f)

Describe two the disadvantages of the bus topology?[6] Any damage to the shared media(cable) would bring the entire network down.[3]

It has a more difficult access control problem because of the single media used in a multi-access environment. [3]

SECTION B
[ 60 marks ] Answer ANY 2 questions in this section

B1
[30 marks]

(a) Briefly explain the circuit switching technique. [3] Call Set-Up - A station dials up another a if the receiving accepts the call a dedicated physical path is established between the two stations. [1] Data Transfer - Once the dedicated path is established, data can be transferred between the 2 stations. [1] Call Termination Once data transmission ceased the path is released so that it may be utilized by other parties.[1]

(b) Explain the concept of space division multiplexing and where would this technique be used?[3] Each channel consists of a separate physical connection[1] that make up a communication facility[1]. This technique is used in the star topology where every node has a separate cable linking it to the hub[1]

(c) Explain what a guard band is and why and where it is used [4] Guard bands are unused bandwidth[1] that separate sub-channels multiplexed by frequency[1]. This is done ensure there is no overlap between the frequencies.[2].

(d) How is statistical Time Division multiplexing different from synchronous TDM. Under what circumstance(s), would each be suitable. [6]

Synchronous TDM would be appropriate in a situation where all users are actively Sending because all stations will be given time slots equally to send their data[2] Statistical TDM is suitable where only one or several users are active[2].

Reason : Additional addressing overhead is needed to identify the owner of a particular time-slot[2]

(e) Does attenuation affect microwave transmission? Explain your answer. [2] Yes[1], rainfall is the cause. The raindrops absorb the microwaves [1]

(f) Consider a transmission that carries 3 bits per signaling element and that the shortest signaling element is 0.8 milliseconds. Calculate both the signaling rate and the data rate of this transmission. Show all workings. [4] Baud = 1 / (0.8 x 10 3) => 1250 baud [2] Bit rate = baud rate x no. of bits per element = 1250 x 3 [2] = 3750 bps

(g) Briefly describe some characteristics of a Local Area Network. [5] Limited Geographical location [1] High data rates [1] Dedicated communications channel [1] Peer to peer communications [1] Low error rate [1] Owned by a single entity [1] Any five of the above points

(h) What is the maximum line utilization of an asynchronous transmission assuming there is only one start and one stop bit per character? State any assumptions you wish to make. Show all workings. [3] Assume a 8 bit character + 1 start and 1 stop bit = 10 bits[1] (8/ 10) * 100% [1] = 80%[1] Note : other assumptions may have different results Awards up to full credit for logical assumptions

B2
[30 marks]

(a) Describe the characteristics of message switching [5] message send as a single block[1] subject to various forms of delays making it unsuitable of interactive traffic[1] minimal addressing overheads as only one header and one trailer is needed for the entire message[1]. uses the store and forward method of data transmission[1] if error occurs entire message has to be re-sent[1] other characteristics may also be accepted subjected to one mark pr characteristic

(b) Packet switching differs slightly from message switching. Describe the differences and similarities between them?[12]

Message Packet Low addressing overheads as High addressing overhead as each packet entire message contains only one has to contain full addressing information header and one trailer. [2] [2] If error occurs the entire message Only affected packets need to re-sent. [1] has to be re-sent [1] Subject to substantial delays and Minimal delays and as such is suitable for as such is not be suitable for interactive traffic [2] interactive traffic [1] Message will arrive in sequence Packets are individually routed and hence at receiving end. [1] may arrive at the destination in the wrong order and has to be re-as assembled. [2]
[2 marks for each difference] other relevant points may earn credit

(c) Packet switching requires control and address bits to be added to each packet. This causes considerable overhead in packet switching. In circuit switching, a transparent circuit is established. No

extra bits are needed and therefore line utilization for circuit switching is better than for packet switching. Do you agree with this statement? Explain your answer.[5] NO[1] the statement ignores overhead incurred during the call set up and call termination phases[2]. Also, for circuit switching bandwidth is dedicated between the 2 parties and may result in 0% utilization should no transmission takes place after call set-up(periods of silence)[2].In contrast packet switching allows link resources to be shared amongst several connections and hence provides very good bandwidth utilization.[1] [Maximum 5 marks]

(d) What is the problem associated with parallel transmission. Briefly explain this problem and why it occurs. Does this problem also affect serial transmission? Explain .[8] Skew[1] Separate cables carry separate bits of a character simultaneously.[1] Over large distances the bits lose their timing relationship. [2] Thus bits from the same character arrive at different times causing errors[2] No it does not apply to serial transmission as all bits sent contiguously over one cable[2]

B3
[30 marks]

(a) What is the IEEE standard for contention networks? [2] IEEE 802.3 [2]

(b) Describe how do contention-based networks operate. [8] All nodes are attached d to a shared media and transmission is bi-directional.[2] Nodes are constantly listening to the media. If a node wishes to transmit data it will wait for the media to be idle and then start transmitting it data. (CSMA).[2] After sending data the node continues to listen to the media. If a collision occurs a jam signal is sent out to inform all nodes that a collision has occurred [2](CD). The node will then back-off for a random interval and then re-transmit its data.[2].

Other relevant points may earn credit [Maximum 8 marks]

(c) What is the full form of the acronym CSMA/CA. What type media access scheme does this protocol use. Explain how this type of media access scheme works. Give an example of a network that employ this technique. [7]

Carrier sense multiple access with collision avoidance[1] It is based on the distributed round-robin technique[1]. Each station listens to the carrier while a transmission is in progress. After the transmission ends, each station waits for a period of time, based on its position in a logical list. [2] If no other station has started transmitted by the time a particular stations time has elapsed, it may begin to transmission.[2] Example:- ISDN basic rate interface [1]

(d) Explain the stop and wait protocol. [5] The sender sends out a block and then waits for an acknowledgement from the receiver[1]. Upon receiving the block the receiver sends out a positive acknowledgement if the block is error-free. The sender then sends out the next block[1]. If errors are a detected a negative acknowledgement is sent and the sender resends the original block.[2]. Also, if no acknowledgements are received after a fixed interval the sender will re-send the block.[1]

(e) What is the sliding window protocol used for? Describe, how this protocol works? [8] This protocol is used for flow control[1] The sender maintains a list of all blocks sent but not yet acknowledged.[2] The receiver maintains a list of all blocks it can accept[1] The window size is the flow control parameter[1] Example :Explanation Total 3 marks

[ Total 100 marks ]

- END OF PAPER -