Delay statistics and throughput performance for multi-rate wireless networks under multiuser diversity

Abstract — An analytical framework for radio link level perfor-mance evaluation under scheduling and automatic repeat request (ARQ)-based error control in a multi-rate wireless network is presented. The multi-rate transmission is assumed to be achieved through adaptive modulation and coding (AMC) in a correlated fading channel. The analytical framework, which is developed based on a vacation queueing model, can be applied to any scheduling scheme as long as the evolution of the joint service/vacation and channel processes can be determined. The exact statistics of queue length and delay are obtained and the radio link level throughput is calculated under both saturated and non-saturated buffer scenarios. As an example of using the general analytical model, we analyze the performance of max-rate (MR) scheduling scheme which exploits multiuser diversity and compare its performance with the round-robin (RR) scheduling scheme. Although the MR scheduling always results in higher throughput than the RR counterpart, we observe that the RR scheduling offers better delay performance than the MR scheme under light traffic load conditions. The usefulness of the presented analysis is highlighted by illustrating its applications for cross-layer design and packet-level admission control under delay constraints. After all, this analytical framework would be very useful for comprehensive analysis of radio link level scheduling schemes and hence for design and engineering of radio link control protocols. Index Terms — Adaptive modulation and coding (AMC), finite state Markov channel (FSMC), automatic repeat request (ARQ), wireless scheduling, cross-layer design, admission control. I