Chapter 12 - 96 Problems and Solutions (Chapter 12) 1....

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 96 Problems and Solutions (Chapter 12) 1. Describe the 051 model. In which Iayer(s) does CDPD operate? [Solution] Application Layer: Provides access to the OSI environment for users; provides distributed information services. Presentation Layer: Provides application independence from data‘rep- resentation (syntax) differences. Session Layer: Provides the control structure for communication be- tween applications; manages all aspects of connections (sessions) between cooperating applications. Transport Layer: Provides reliable and transparent data transfer be- tween end points; provid end-to-end error recovery and flow control. Network Layer: Responsible for establishing, maintaining, and termi- nating connections. Provides upper layers with independence from the data transmission and switching tasks. Data Link Layer: Provides reliable data transfer across the physical link; manages frame synchronization in terms of error control and flow control. Physical Layer: Manages the movement of the unstructured bit stream over the physical medium. Handles all aspects of accessing the physical medium. CDPD operates on Physical Layer and Data Link Layer of the 081 model. . What are the differences between 081 and TCP/IP protocol models? Ex- plain clearly. [Solution] These are the differences between 031 and TCP/IP protocol models: a The 031 model consists of 7 layers, whereas the TOP/1P model con- sists only 5 layers. 0 In the TCP/IP protocol suite, the application layer serves the pur- pose of the three combined layers of application, presentation and session. The 081 model made a clear distinction between the top three layers. Each application in the TCP/IP suite has to indepen- dently implement the session and presentation layer functions. a The 081 suite was put forward before the protocols were invented whereas in the case of TCP/IP. the model was a description of the existing protocols. - o The 031 model supports connectionless and connection-oriented com- munication in the network layer, and only connection-oriented in the 97 transport layer. TCP/IP supports only connectionless communica- tion in the network layer, and provides a choice of both connectionless and connection-oriented communication in the transport layer. 3. Look at your favorite website and find the difl'erence between interior and exterior routing protoool. [Solution] An interior routing protocol is used within an autonomous system (AS), whereas an exterior routing protocol is used for routing between autonomous systems. Every AS is free to choose its own routing protocol to handle the routing of packets within itself, whereas exterior routing protocol is used to handle exterior routing. ' 4. “Basic RIP supports single subnet masking for each IP network”, give a practical example where this becomes a critical issue. ' [Solution] There may be systems that use different subnet masks for different subnets within a single network. 5. What are the difierences between path vector routing and shortest path routing? Explain clearly. [Solution] In the path vector routing (the routing protocol on which BGP is based) , every entry in the routing table contains the destination network, the next router and the path to reach the destination. The path usually contains an ordered list of autonomous systems that a. packet should travel, which in some cases may not be the shortest path. This differs from the shortest path routing protocol that just specifies the next router in the routing table entry for the packet to be forwarded to. 6. What is DHCP? How does DHCP supports dynamic address allocation? [Solution] DHCP (Dynamic Host Configuration Protocol) was proposed to reduce the problan of the limited IPv4 address space, by dynamically allocating IP addresses to hosts that need access to the Internet once in a. while. Thus, a permanent IP address does not need to be allocated. In DHCP, a heat first sends a DHCP discover packet on its network, specifying its hardware address. A DHCP server then sends a DHCP offer packet to the host, specifying the offered IP address. The host may receive multiple ofl'er packets. It then confirms its selection by broadcasting the selected I? address through a DHCP broadcast packet. The selected server sends a DHCP ACK packet, and the other servers then withdraw their offer. A lease time is associated with the IP address allocated and the host needs to renew the lease before its expiration time. 98 7. 10) With suitable examples? Explain the differences between a connection- oriented and connectionless protocols. [Solution] A connection-oriented protocol needs the client and the server to establish a connection with the help of control packets, before the transmission of any data packets. This is termed as a handshake between the client and the server. The handshake procedure serves to negotiate certain parameters such as determining the flow control so as to prevent the receiver from being overwhelmed, establishing the sequence numbers between the two parties etc. A connectionless protocol does not involve any handshaking mechanism between the two parties before data transfer. The sender can directly send a data packet to the receiver irrespective of whether the receiver is ready to accept it or not. The advantage of a connectionless protocol is that it does not require the overhead involved with establishing a connection, thereby improving the time required for the first packet to be received by the receiver. This is useful when the sender needs to intermittently send a few padoets, and so the time spent on establishing the connection is not justified. What are the disadvantages of using wireline TCP over wireless networks? [Solution] The main disadvantage of using wireline TCP over wireless networks is that wireline TCP attributes loss of packets during packet transmission to congestion in the network. However, this may not be the case in wireless networks, where packet losses occur mainly due to the physical nature of the medium such as attenuation, thermal effects and interference in the air medium. The wireline TCP thus goes into congestion control mechanism in these cases when there is no need to do this. This further reduces the throughput. . Explain the significance of “initial sequence number” in TOP? [Solution] The ‘Initial Sequence Number (ISNY is the number associated with the first byte that the sender transmits. This is established during the three- way handshake. Each party specifies the ISN that it would use for the first byte. The sequence number is increased for every succeeding byte. The ISN is chosen randomly, instead of zero, to prevent any confirsion because of delayed segments from previous connections. What are the inherent characteristics of wireless networks that require changes in existing TOP? [Solution] 11. 12. 13. 14. 15. 99 High packet error rate, high latency, frequent link breakage; due to mo— bility are some of the inherent diaractefistics of wirel networks which require changes to be made in the existing TCP in order for it to be used in a wireless medium. What are the particular advnntages and disadvantages of using a split TCP approach for wirelas networks? {Solution} The advantage of the split TCP approach for wireless networks is that it hides the mobility of the receiver from the sender. Its disadvantage is that the TOP connection between the sender and the receiver gets split at the intermediate BS and does not maintain end to d. What are the problems faced by designers of wireless TCP stack when using link layer protocols? [Solution] In wireless TCP scenario, some of the link layer protocols uses connection information from the TCP layer (service awareness). This breaks the layer independence in TOP/1P stack. What makes the fast-retransmission approach desirable in improving TCP performance over wirelms networks? [Solution] In the Fast-Retransmission approach. whenever the receiver completes a handofl’. it sends a certain number of acknowledgements to the sender. The TOP at the sender then reduces its window size and retransmits packets starting from the first missing packet, for which the duplicate acknowledgement has been sent. Thus TCP does not need to enter slow start. When is the reliable link layer useful in enhancing TCP performance? [Solution] If the link layer provides almost in—order delivery and TCP retransmisn sion timeout large enough to tolerate additional delays to the link layer retransmits. What is the operational difference between standard ACKs used in con- ventional TCP and SACKs used in wireless TCP? What improvement in performance does it provide for wireless networks? [Solution] Conventional TCP uses cumulative acknowledgements wherein an ACK for a particular segment can be used cumulatively to acknowledge all the preceding segments. In SACKs used in wireless TOP, the ACKs can be used to acknowledge segments selectively rather than cumulatively. The 100 16. 17. 18. 19. advantage of this is that if a particular segment in a transmission is cor- rupted, then the sender needs to retransmit only that particular segment, thereby saving precious bandwidth that would have been wasted due to the retransmissions of the succeeding segments that have been received correctly. Both I-TCP and M—TCP are split TCP approaches to improving the per- formance of wireline TCP over wireless networks. What is the difference between these two approaches? [Solution] In M—TCP, the receiver can make the sender enter the persist mode by advertising a zero window size in presence of frequent diseonnections. In I~TCP, all the support needed for handling the mobility related prob- lems, is built into the wireless side of the interaction. Handofi between two different MSRs is supported on the wireless side without having to reestablish the connection at the new MSR. Even though explicit bad state notification (EBSN) appears to be a very pragmatic approach for improving TCP performance over wireless net- works, what is the most significant disadvantage? [Solution] The main disadvantage of EBSN is that it requires modification to TCP code at the source. Can any of the methods (e.g., I-TCP, M~TCP, SACK, EBSN, etc.) be used to improve performance of TCP over wireless ad hoe networks? Suggest any ways by which this can be done. [Solution] ' ' - The above protocols are basically designed for a single hop infrastructure wireless network. To make it applicable for wireless ad-hoc networks, they needs to be modified to support the multihop wireless links with support for mobility. _How many interations are needed to calculate shortest path to all nodes from node 3? find the shortest distance to each node and what is the path used for each one of th? 20. 21. 101 Figure for Problem 12.19. [Solution] 2 iterations are required to calculate shortest paths from node 3. The shortest paths from node 3 are: Given figure in Problem 12.19 as the connectivity graph of anetwork. you are allowed to go through only two steps of Bellman-Ford algorithm at each node so that their complexity (and hence the time required) can be kept to a low value. What is the impact on shortest path calculations? Comment on the accuracy of the procedure? [Solution] Since the Bellman-Ford algorithm managm to find the shortest path to n—hop nodes in n iterations, the given graph will be correctly evaluated. The procedure will not be accurate for larger networks. However, running the procedure on all nodes might produce correct routing by using other distributed methods. What kind of security measures used in diiferent layers of TCP/IP? Ex- plain. [Solution] The TCP/IP protocol stack has inbuilt security mechanisms at the various levels. These are as follows: TOP: The TC? suite uses the SSL/TLS security suite. This protocol is summa- rized as follows: Secure Socket Layer/Transport Layer Security (SSL/TLS): The SSL/TLS protocol divides the data into records. These records are of four types: user data, handshake messages, alerts and change cipher specs. User A first contacts user B to initiate dialogue. B sends his certificate to A. After verifying the certificate, A extracts B’s public key. It then picks a random number R, encrypts it with B‘s public key and sends it to B. Both now compute the session key using this random number. 1?: 102 22. 23. The IP layer uses the IP Security protocol (IPSec) for security purposes. This protocol, coupled with the Internet Key Exchange (IKE) protocol, makes the IP layer security quite robust. A Security Association (SA) is established between the two end parties. The IKE protocol uses a Diffie Hellman key exchange process to ensure integrity of the manage: passed. The security is additionally protected by an optional certificate that either party may request from each other. Having established a shared session key using IKE, the IPSec protocol comes into play. The IPSec protocol has two components, the Authentication Header (AH) and the Encapsulating Security Payload (ESP). AH provides integrity protection only while ESP provides encryption and/or integrity protection. In most cases, only one of the two is used. IPSec operates in two modes, the tunnel and the transport mode. In the tunnel mode, the original IP packet is kept intact and a new IP header is added to the packet. The transport mode adds the IPSec infonnation between the 1P header and the remainder of the packet. What are the advantages of IPv6? Discuss if an IPv6 network can support IPv4 packets and how? [Solution] Advantages of IPv6: o The 128-bit address space of IPv6 can support 2128 IP addresses, instead of the 232 addresses provided by IPv4, which is supposed to solve the problem of limited address space once and for all. - 1va has a fixed header size, while the IPv4 had h variable header slze. - 0 IPv6 has fewer fields in the basic header, which results in faster packet processing. 0 In IPv6, the routers do not perform any fragmentation, unlike IPv4. o H’v6 supports authentication and encryption. ' An IPv6 network can also support packets with IPv4 addressing that might be needed for the transition period. It might happen that the two and systems support IPv6. but the packet needs to travel through an intermediate IPv4 system. In this case, the-packet needs to have an IPv4 address. The sender can encapsulate the IPv6 packet within an IPv4 packet (tunneling) and it sets the protocol number. Here, the end system uses aoompatible IP address. For e.g., the IPv4 address 2.13.17.14 beoom 0::020D:110E. IPv6 supports resource allocation. Explain how this is achieved? [Solution] 103 IPv6 can support resource allocation by the use of the flow label field in the IP header. The flow label can be used to support transmission of real time audio and video, which require resources such as high bandwidth, large bufi'ers, long processing time, etc. ' ...
View Full Document

Page1 / 8

Chapter 12 - 96 Problems and Solutions (Chapter 12) 1....

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online