Optimal Resource Allocation Framework for Downlink OFDMA System with Channel Estimation Error

In this paper, we study resource allocation in a downlink OFDMA system assuming imperfect channel state information (CSI) at the transmitter. To achieve the individual QoS of the users in OFDMA system, adaptive resource allocation is very important, and has therefore been an active area of research. However, in most of the the previous work perfect CSI at the transmitter is assumed which is rarely possible due to channel estimation error and feedback delay. In this paper, we study the effect of channel estimation error on resource allocation in a downlink OFDMA system. We assume that each user terminal estimates its channel by using an MMSE estimator and sends its CSI back to the base station through a feedback channel. We approach the problem by using convex optimization framework, provide an explicit closed form expression for the users' transmit power and then develop an optimal margin adaptive resource allocation algorithm. Our proposed algorithm minimizes the total transmit power of the system subject to constraints on users' average data rate. The algorithm has polynomial complexity and solves the problem with zero optimality gaps. Simulation results show that our algorithm highly improves the system performance in the presence of imperfect channel estimation.