A Game Theoretic Framework for Cognitive Radio Networks Using Adaptive Channel Allocation Spectrum

Abstract— In this research, we propose a game theoretic framework to investigate the behavior of the cognitive radio networks for distributed adaptive channel allocation. We illustrate two separate objective functions for spectrum sharing games, which capture the benefit of selfish users and cooperative users, respectively. Based on utility definition for cooperative users, we determine that the channel allocation problem can be formed as the potential game, and converges to a deterministic channel allocation Nash equilibrium point. Alternatively, no regret learning implementation is proposed for both scenarios. Also, it is pointed to have similar performance with the possible game when cooperation is expected, but with a higher variability beyond users. The no regret learning formulation is beneficial to accommodate selfish users. Non-cooperative learning games have very low overhead for information interchange in the network. We point that cooperation-based spectrum sharing protocol improves the overall network performance at expense of an extended overhead needed for information exchange