Spectrum Handoff Using Game Theory with Hypergraph Scheme in Distributed Cognitive Radio Networks

In this paper, a spectrum handoff scheme based on game theory and hypergraph in distributed Cognitive radio networks (CRNs) is proposed. Firstly, the hypergraph model is used to characterize potential spectrum occupation conflicts. By defining the potential cumulative conflict level (PCCL) function, the level of conflicts between secondary users (SUs) is quantified, and the cumulative handoff delay is obtained by combining the data transmission rates of SUs, which is taken as the optimization objective to maximize the CRNs throughput. Secondly, the spectrum handoff problem is characterized as a local altruistic game, which is proved to be an exact potential game (EPG) with at least one Pure Strategy Nash Equilibrium (PNE). Thirdly, a spatial adaptive play algorithm based on SUs' transmission rates (BTR-SAP) is proposed, which marks SUs' status with their transmission rates. Lastly, the performance of the proposed scheme is verified by simulation experiments. Simulation results show that the proposed scheme is superior to traditional SAP algorithm in terms of convergence speed and network throughput, and significantly improves spectrum utilization efficiency.