A User-Differentiation-based Resource Allocation Scheme for OFDMA Downlink Systems

This paper proposes a new user-differentiation-based resource allocation (UDRA) scheme for OFDMA downlink systems. The UDRA scheme can strike a good balance between quality-of-service (QoS) requirement guarantee and system throughput enhancement, whereas conventional schemes cannot explicitly and accurately control this tradeoff. The UDRA scheme first performs a utility-based resource allocation algorithm for urgent users to fulfill QoS requirement, then it performs a maximum-sum-rate resource allocation algorithm for regular users to enhance system throughput. To achieve the balance between QoS guarantee and throughput enhancement, the UDRA scheme further designs a fuzzy logic user identifier (FLUI) to intelligently differentiate users from urgent and regular. Simulation results show that the proposed UDRA scheme enhances the system throughput by 23.1%, 14.1%, and 9.7%, compared to the PSRA [1], ARRA [2], and U-TMCR [3] schemes, respectively, under a QoS requirement guarantee.