Dynamic Resource Allocation in OFDMA Uplink for MAI Mitigation and Throughput Improvement

In OFDMA uplink, Multiple Access Interference (MAI) caused by synchronization offsets in user signals can considerably degrade user throughputs. In this paper, we investigate the problem of applying Dynamic Resource Allocation (DRA) to mitigate MAI and simultaneously improve user throughputs. Firstly, we introduce a novel solvable optimization problem (OP) following DRA to maximize the minimum user throughput. Using this proposed OP, we show the optimal resource allocation schemes can increase minimum user throughput up to 300% compared to existing DRA heuristics, and to 25% compared to the estimation and correction based approach. Secondly, by investigating the frequency assignment, we show an insight about the advantage of DRA in uplink, where better exploitation of frequency and multiuser diversities leads to efficient MAI mitigation and significant throughput gain. Finally, we introduce suboptimal OPs to reduce the complexity, among that the one based on chunking can significantly reduce the complexity and still provide comparable throughput gain.