Combined resource allocation for multi-user OFDM systems

Abstract

For the problem of dynamic resource allocation in an adaptive multi-user orthogonal frequency division multiplexing (OFDM) system, this paper adopts the framework of cross-layer to propose a novel resource allocation strategy on combined sub-carriers, bits and power to minimize power consumption of base station under total bits constraint. The distribution of sub-carriers concerns not only channel state information (CSI), but also user fairness, priority and different quality of service (QoS) requirements as well through information exchange with media access control (MAC) layer. Meanwhile, by introducing net utility function based on Game theory, the process of bit loading and power allocation on each sub-carrier can be transformed into a set of parallel net utility function solving problems. Also, the proposed re-loading algorithm helps to implement adaptive modulation and coding (AMC) on hardware through limited granular bit loading. Simulation proves better performances on users' fairness, priority, overall power consumption, and algorithm complexity.