Advanced Wireless Networks - 4G Technologies

REFERENCES 291

REFERENCES

[1]T. Henderson and R.H. Katz, Transport protocols for Internet-compatible satellite networks, IEEE JSAC, vol. 17, no. 2, 1999, pp. 326–344.

[2]T.V. Lakshman and U. Madhow, The performance of TCP/IP for networks with high bandwidth–delay products and random loss, IEEE/ACM Trans. Networking, 1997, vol. 5, no. 3, pp. 336–350.

[3]S. Floyd and V. Jacobson, Random early detection gateways for congestion avoidance, IEEE/ACM Trans. Networking, vol. 1, no. 4, 1993, pp. 397–413.

[4]L. Brakmo and L. Peterson, TCP Vegas: end to end congestion avoidance on a global internet, IEEE JSAC, vol. 13, no. 8, 1995, pp. 1465–1480.

[5]C. Barakat and E. Altman, Analysis of TCP with several bottleneck nodes, IEEE GLOBECOM, 1999, pp. 1709–1713.

[6]W.R. Stevens, TCP/IP Illustrated, vol. 1. Reading, MA: Addison-Wesley, 1994.

[7]R. Caceres and L. Iftode, Improving the performance of reliable transport protocols in mobile computing environments, IEEE J. Select. Areas Commun., vol. 13, 1995

pp.850–857.

[8]A. Desimone, M.C. Chuah and O.C. Yue, Throughput performance of transport layer protocols over wireless LAN’s, in Proc. IEEE GLOBECOM’93, December 1993,

pp.542–549.

[9]K. Fall and S. Floyd, Comparisons of Tahoe, Reno, and Sack TCP [Online], March 1996. Available FTP: ftp://ftp.ee.lbl.gov

[10]V.K. Joysula, R.B. Bunt and J.J. Harms, Measurements of TCP performance over wireless connections, in Proc. WIRELESS’96, 8th Int. Conf. Wireless Communications, Calgary, 8–10 July 1996, pp. 289–299.

[11]T.V. Lakshman and U. Madhow, The performance of TCP/IP for networks with high bandwidth–delay products and random loss, IEEE Trans. Networking, vol. 3, 1997,

pp.336–350.

[12]P.P. Mishra, D. Sanghi and S.K. Tripathi, TCP flow control in lossy networks: Analysis and enhancements, in Computer Networks, Architecture and Applications, IFIP Transactions C-13, S.V. Raghavan, G.V. Bochmann and G. Pujolle (eds). Amsterdam: Elsevier North-Holland, 1993, pp. 181–193.

[13]A. Kumar, Comparative performance analysis of versions of TCP in a local network with a lossy link, IEEE/ACM Trans. Networking, vol. 6, no. 4, 1998, pp. 485–498

[14]A. Bakre and B.R. Badrinath, I-TCP: indirect TCP for mobile hosts, in Proc. 15th Int. Con. on Distributed Computing Systems, Vancouver, June 1995, pp. 136–143.

[15]A. Bakre and B.R. Badrinath, Indirect transport layer protocols for mobile wireless environment, in Mobile Computing, T. Imielinski and H.F. Korth (eds). Kluwer Academic: Norwell, MA 1996, pp. 229–252.

[16]H. Balakrishnan, S. Seshan and Randy Katz, Improving reliable transport and handoff performance in cellular wireless networks, Wireless Networks, vol. 1, no. 4, 1995.

[17]H. Balakrishnan, V.N. Padmanabhan, S. Seshan and R. Katz, A comparison of mechanisms for improving TCP performance over wireless links, in ACM SIGCOMM’96, Palo Alto, CA, August 1996, pp. 256–269.

[18]B.R. Badrinath and T. Imielinski, Location management for networks with mobile users, in Mobile Computing, T. Imielinski and H.F. Korth (eds). Kluwer Academic: Norwell, MA 1996, pp. 129–152.

292 ADAPTIVE TCP LAYER

[19]K. Brown and S. Singh, A network architecture for mobile computing, in Proc. IEEE INFOCOMM’96, March 1996, pp. 1388–1396.

[20]K. Brown and S. Singh, M-UDP: UDP for mobile networks, ACM Comput. Commun. Rev., vol. 26, no. 5, 1996, pp. 60–78.

[21]R. Caceres and L. Iftode, Improving the performance of reliable transport protocols in mobile computing environments, IEEE Selected Areas in Commun., vol. 13, no. 5, 1994, pp. 850–857.

[22]R. Ghai and S. Singh, An architecture and communication protocol for picocellular networks, IEEE Personal Commun. Mag., no. 3, 1994, pp. 36–46.

[23]K. Seal and S. Singh, Loss profiles: a quality of service measure in mobile computing, J. Wireless Networks, vol. 2, 1996, pp. 45–61.

[24]S. Singh, Quality of service guarantees in mobile computing, J. Comput. Commun., vol. 19, no. 4, 1996, pp. 359–371.

[25]S. Floyd and V. Jacobson, Random early detection gateways for congestion avoidance, IEEE/ACM Trans. Networking, vol. 1, no. 4, 1993, pp. 397–413.

[26]J. Liu and S. Singh, ATCP: TCP for mobile ad hoc networks, IEEE J. Selected Areas Commun., vol. 19, no. 7, 2001, pp. 1300–1315.

[27]http://www-nrg.ee.lbl.gov/ecn-arch/

[28]G. Holland and N. Vaidya, Analysis of TCP performance over mobile ad hoc networks, in Proc. ACMMobile Communications Conf., Seattle, WA, 15–20 August 1999, pp. 219–230.

[29]D.B. Johnson and D.A. Maltz, Dynamic source routing in ad hoc wireless networks, in Mobile Computing, T. Imielinski and H. F. Korth (eds). Norwell, MA: Kluwer Academic, 1996, pp. 153–191.

[30]K. Chandran, S. Raghunathan, S. Venkatesan and R. Prakash, A feedback-based scheme for improving TCP performance in ad hoc wireless networks, in Proc. 18th Int. Conf. Distributed Computing Systems, Amsterdam, 26–29 May 1998, pp. 474–479.

294 CROSSLAYER OPTIMIZATION

but with lower capacity. However, in practice, there is still the problem that the TCP sender may not be fully shielded from wireless link losses. This can lead to the TCP congestion control mechanism reacting to packet losses, thus resulting in redundant retransmissions and loss of throughput. However, once ECN-enabled TCP is deployed, where the ECN bit can be used to mark packets to indicate congestion, there is a means of differentiating between congestion-related loss and wireless channel-related loss.

Thus, the channel need not be smoothed because the ECN mechanism provides a means of explicitly indicating congestion. In Chapter 9 it was demonstrated that, in a single user environment, if packets are marked based solely on congestion information, there is no significant degradation of TCP performance due to the time-varying nature of the wireless channel as compared with wired networks. Such an approach is an example of where a crosslayer view of physical layer information (channel conditions) is used at the network layer to significantly improve network-layer throughput performance.

As a second example consider a conventional cellular system with a fixed base station and a number of mobile users. Data flows (packets) arrive from the core network to the cell base station and are destined for the mobile users, with the packets for each user being queued temporarily at the base station (a separate queue is maintained for each user). The objective of the base station is to schedule these packets to various mobiles in a timely manner. Owing to the asymmetric nature of the traffic we can restrict ourselves to just considering the forward link problem, where the direction of the data flow is from the base station to the mobile user. An efficient way to use the cellular spectrum is to dedicate some bandwidth for data transport and allocate this bandwidth in a dynamic manner among various data users. For instance, time could be divided into fixed-size time slots and users allocated time slots dynamically. The simplest scheduling algorithm is the round-robin mechanism, where users are periodically allocated slots irrespective of whether there is data to be sent to a particular mobile user. Note that this simple TDM scheme still leads to some wasted bandwidth during inactivity. Furthermore, it suffers from a more subtle problem: this approach is independent of the channel state (a channel-state-independent scheduling mechanism).

A channel-state-dependent scheduling algorithm can provide significant gains. As an illustration consider a wireless system consisting of three users as in Figure 10.1. For this example, we consider access time to be slotted, and the channels to be constant over a time slot. For demonstration, let us assume that the channels are either ON or OFF, equally likely, and the channels are independent of each user. Thus, in this system, there are eight possible (instantaneous) channel states for the three independent users, ranging from (ON, ON, ON) to (OFF, OFF, OFF). When a user’s channel is ON, one packet can be transmitted successfully to the mobile user during the time slot. The system is assumed to be a TDM system; thus, the base station can transmit to only one user on each time slot. The associated scheduling problem is to decide which user is allowed access to the channel during each particular time slot. A naive scheduling rule would be to employ the well-known roundrobin mechanism. In such a scheme, the users are periodically given access to the channel, with each user getting one-third of the slots over time. As the channel of each user is equally likely to be ON or OFF in each time slot, it follows that, over time, on average, each user will get a data rate of (1/2)(1/3) = 1/6 packets/slot.

On the other hand, suppose the base station uses knowledge of the instantaneous channel state. Then a simple policy would be to schedule and transmit to a user whose channel is in the ON state. If more than one user’s channel is in the ON state, the scheduler could pick a user randomly (equally likely) among those users whose channels are ON, and send

INTRODUCTION 295

Scheduler

Data from the Internet

Figure 10.1 Channel state-dependent scheduling.

data to the selected user. This simple example assumes that all users have identical traffic demands. For the case where some users have greater needs than others, high-demand users would be assigned channel access with greater likelihood. In this case data is not sent by the base station if and only if all users’ channels are OFF (which occurs on average only one-eighth of the time). Thus, the total data rate achieved by this state-dependent rule is 1 − 18 = 78 packets/slot. As this rule is symmetric across users, it follows that on average, the data rate per user is (1/3)(7/8) = 7/24 packets/slot, which is almost twice the throughput as the round-robin scheme that provided 16 packets/slot. In addition, the base station does not radiate power during bad channel conditions, thereby decreasing interference levels in the wireless network.

This gain achieved due to channel-state-dependent scheduling is called the multi-user diversity gain. This example illustrates the significance of multi-user diversity gain in network scheduling. However, a central question is how to design online algorithms that achieve this gain, while also supporting diverse QoS requirements for various users for realistic channel scenarios. While this question is not yet completely answered, there has been extensive research on various aspects of the problem [2].

Figure 10.3 GOP structure of MPEG-4 PFGS (progressive fine granularity scalable) scalable video. (Reproduced by permission of IEEE [50].)

packets among the classes. At the video application layer, each video packet is characterized based on its loss and delay properties, which contribute to the end-to-end video quality and service. Then, these video packets are classified and optimally mapped to the classes of link transmission module under the rate constraint [57, 58]. The video application layer QoS and link-layer QoS are allowed to interact with each other and adapt along with the wireless channel condition. The objective of these interaction and adaptation is to find a satisfactory QoS tradeoff so that each end user’s video service can be supported with available transmission resources.

Figure 10.3 shows a group-of-picture (GOP) structure of MPEG-4 PFGS ( progressive fine granularity scalable) [56]. Suppose that there are video frames and there are video layers in one GOP. Therefore, the loss of any video portion will affect the end-to-end video quality due to the interdependency within the encoding structure. Each video layer is packetized into several fixed-size packets before transmission. Each video packet cannot contain video data across video layers or video frames. Packets from the same video layer are put onto the same priority class. Furthermore, suppose that the video playback frame rate at the end user is fixed at F frames/s. If the mobile terminal starts to play back the first video frame of a GOP at time Tp, video frame n in the same GOP should be received and be ready to be displayed before time Td (n) = Tp + [(n − 1)/F] for uninterrupted playback.

Let p = ( p1, p2, . . . , pM ) be the mapping policy from M video layers to K priority classes, where p j {0, 1, . . . , K } is the priority class that video layer j is transmitted. pj = 0 represents the fact that video layer j is abstained from transmission. The overall expected distortion from the mapping scheme p can be derived using the dependent structure of scalable video [50, 57], as shown in Figure 10.3, which can be expressed as follows:

j=0