Analysis of the statistical time-access fairness index under a scheduler exploiting multiuser diversity

Recent studies report that in wireless networks, schedulers exploiting multiuser diversity can substantially enhance the maximum throughput of the overall system. In this paper, we focus on a scheduler which maximizes the throughput of the overall system by exploiting multiuser diversity and is combined with adaptive modulation and coding. We numerically study the statistical time-access fairness of the scheduler. The transient statistical fairness properties of schedulers exploiting multiuser diversity have not been sufficiently explored yet, although they are important in practice. We develop two numerical methods to understand the transient statistical fairness properties of the scheduler. Numerical results exhibit that under the scheduler, serious unfairness between users can remain for a relatively long period even if the users have the same average SNR.