Learning Based Spectrum Sharing Algorithms by Using Coopetition Game Approach

摘要

As the rapid development of wireless communication, the scarcity of spectrum become more and more prominent. Recently, Cognitive Radio (CR) technology is proposed to overcome the problem of low spectrum utilization brought by static spectrum allocation. In this paper, a new CR spectrum sharing scheme is developed by employing the dynamic coopetition game model. Coopetition game approach is an effective tool to describe CR sharing problems. The proposed scheme dynamically senses the spectrum based on the global game and efficiently re-allocates the available spectrum according to the trust-proportional bargaining approach. The main goal of our game based scheme is to greatly improve the sensing performance while maximizing the overall system capacity. Under widely different CR network situations, the proposed scheme is so dynamic and flexible that it can adaptively respond to current CR network conditions.