Resource Allocation for Downlink Multi-Cell OFDMA Cognitive Radio Network Using Hungarian Method

This paper considers the problem of resource allocation for downlink part of an OFDM-based multi-cell cognitive radio network which consists of multiple secondary transmitters and receivers communicating simultaneously in the presence of multiple primary users. We present a new framework to maximize the total data throughput of secondary users by means of subchannel assignment, while ensuring interference leakage to PUs is below a threshold. In this framework, we first formulate the resource allocation problem as a nonlinear and nonconvex optimization problem. Then we represent the problem as a maximum weighted matching in a bipartite graph and propose an iterative algorithm based on Hungarian method to solve it. The present contribution develops an efficient subchannel allocation algorithm that assigns subchannels to the secondary users without the perfect knowledge of fading channel gain between cognitive radio transmitter and primary receivers. The performance of the proposed subcarrier allocation algorithm is compared with a blind subchannel allocation as well as another scheme with the perfect knowledge of channel-state information. Simulation results reveal that a significant performance advantage can still be realized, even if the optimization at the secondary network is based on imperfect network information.