Low-overhead uplink scheduling through load prediction for WiMAX real-time services

As WiMAX achieves increasing deployment, the large overhead in its uplink scheduling when providing real-time services has become a major challenge. In this study, the authors present effective, low-overhead scheduling algorithms for WiMAX uplink scheduling. The authors adaptively predict users´ load and select a small set of active users to be served. This addresses the major source of overhead in WiMAX uplink scheduling: the Markovian arrival process information elements (IEs) and media access control layer (MAC) service data unit (SDUs) subheader overheads grow with the number of active users. The authors introduce additional novel techniques, including piggybacking, to reduce MAC overhead. The authors implement their algorithms and conduct extensive evaluations. The results show that their algorithms not only provide quality-of-service guarantees, but also substantially reduce the scheduling overhead compared with existing schemes.