Optimum Throughput Constrained Opportunistic Scheduling in Cellular Data Networks

This paper considers the problem of scheduling multiple users in the downlink of a time-slotted cellular data network. For such a network, opportunistic scheduling algorithms exploit the time-varying radio channel and improve the system performance. This paper introduces a new optimum scheduling algorithm that maximizes the sum of average user performance subject to certain minimum and maximum performance constraints. The proposed algorithm, which is named as throughput constrained opportunistic scheduling (TCOS), is an extension of the framework proposed by Liu et al. For memory less fading channels, TCOS performs identically to another optimum algorithm. For fading channels with memory, TCOS significantly improves system performance. The TCOS with the minimum and maximum performance constraints offers better service differentiation among different classes of users than opportunistic scheduling algorithm provides with only minimum performance constraints