Delay quality-of-service driven resource allocation for relay-based multiuser OFDMA networks

This paper focuses on the delay-QoS driven resource allocation problems for multiuser orthogonal frequency division multiple access (OFDMA) relay-based networks. We develop an optimal strategy, with respect to the joint subchannel assignment and power control, to maximize the throughput of the systems, under both delay-QoS guarantees and average power constraints. First, the concept of effective capacity, which characterizes the maximum arrival rate that the system can support with delay requirements, is introduced, and accordingly treated as the maximum throughput under these constraints. Then the optimal algorithm is exploited by dividing such intractable problem into two individual steps with rigorous proofs, to avoid prohibitive computational burden caused by Mixed Integer Nonlinear Programming (MINLP). Numerical results demonstrate that our derived scheme delivers substantial performance improvements comparing with other algorithms, which is feasible and efficient in real-time applications.