Communication Networks Opportunistic multiuser scheduling with reduced feedback load

Abstract

In this paper, we propose a reduced feedback opportunistic scheduling (RFOS) algorithm that reduces the feedback load while preserving the performance of opportunistic scheduling (OS). The RFOS algorithm is a modified version of our previously proposed algorithm, the DSMUDiv algorithm. The main difference is that RFOS consists of a probing process (search process) and a requesting feedback process based on a threshold. The threshold value is variable, and it depends on the probing process. To reduce the feedback rate, a quantised value indicating the modulation level is fed back, instead of the full value of the signal-to-noise ratio (SNR), which we call quantised SNR. The paper includes the closed-form expressions of the probing load, feedback load and spectral efficiency. In addition, we investigate the effect of the scheduling delay on the system throughput (STH). Under slow Rayleigh fading assumption, we compare RFOS algorithm with the DSMUDiv and optimal (full feedback load) selective diversity scheduling algorithms. Copyright © 2010 John Wiley & Sons, Ltd.