Gaussian Process Learning for Opportunistic Scheduling in Wireless Networks
It is well known that opportunistic scheduling algorithms are throughput optimal under dynamic channel and network conditions. However, these algorithms achieve a hypothetical rate region which does not take into account the overhead associated with channel probing and feedback required to obtain the full channel state information at every slot. In this paper, the authors design a joint scheduling and channel probing algorithm takes into account the uncertainty of channels, estimation (regression), and optimization aspects in a holistic and structured manner. They adopt a correlated and possibly non-stationary channel model, which is more realistic than those used in the literature.