The relationship between rate-based utility functions and fairness in wireless networks has been shown in Chapter 8. Rate-based scheduling schemes, which apply the channel state information (CSI) and rate-based utility functions, do not take traffic burstiness into account. In this chapter, utility functions with respect to average delays is used for designing channel- and queue-aware scheduling, which is highly advantageous to data transmission with a low latency requirement.

This chapter is organized as follows. In Section 9.1, we introduce the background and motivations of this work. In Section 9.2, we briefly introduce how to extend scheduling schemes existing in single-carrier systems into the corresponding multi-channel scheduling schemes. In Section 9.3, we develop maximum delay utility (MDU) scheduling based on maximizing the total utility in terms of average waiting time. In Section 9.4, we state the maximum stability region and develop the results regarding stability. In Section 9.6, we use delay transmit diversity and adaptive power allocation to further improve the system performance. Finally, in Section 9.7, we compare several multi-carrier scheduling schemes using simulation.

Introduction

It is increasingly clear that most information traffic would be delivered based on IP networks because of the efficient bandwidth use and the low-cost infrastructure construction. Thus, the queue state information, such as queue length and packet delay, which is a reflection of traffic burstiness, should be utilized in scheduling packets. On the other hand, since the queue state information is tightly connected with quality of service (QoS), wisely controlling queues is one of the most effective ways for QoS provisioning. As compared to channel-aware scheduling, joint channel- and queueaware scheduling would be more beneficial to wireless resource allocation and QoS provisioning.