The Research Of Spectrum Allocation Based On Game Theory In Cognitive Radio

With the development of wireless communication technology, the demand of mobilecommunications and broadband wireless access services has increased dramatically, and thephenomenon of lacking spectrum resource is becoming more and more prominent. As a result，thespectrum allocation in cognitive radio has become a research hot point in the world．Because thecognitive users are intelligent, the spectrum can be shared among primary and second users. Inorder to make full use of spectrum, an economic model is required for the primary users and thesecondary users. Game Theory is the effective method for analyzing spectrum allocation. There is aprofound significance to the research and development of cognitive radio by using Game Theory toanalyze the spectrum.First of all, this thesis analyzes several Game models briefly, and introduces Nash Equilibriumand Bayesian Nash Equilibrium in details. Then this thesis introduces a spectrum allocationalgorithm based on the auction mechanism. By analyzing the Cournot model and Stakelberg model,we can easily find out that these existing algorithms are based on complete information, and theyare not suitable for the practical application.Based on incomplete information Game, this thesis analyzes the spectrum allocation by usingBayesian Game. Firstly, considering the noise is random, we propose a new algorithm based onBayesian Game in this thesis. In cognitive radio system, the transmit power is viewed as one factorof primary users’ cost function. And this algorithm has a broader application in the situation ofnoise uncertain. The experimental results on spectrum allocation demonstrate the effectiveness ofthe proposed scheme. Meanwhile, congsidering the impact of signal-to-noise to the transmissionrate, the spectrum allocation of secondary user based on dynamic Bayesian Game algorithm isproposed in this thesis. In the new algorithm, the signal-to-noise is viewed as one of influencingfactor of the secondary users’ revenue function, so it has a profound significance to the research ofsecondary users. The simulation results show that the spectrum allocation algorithm based onBayesian Game is effective on the spectrum allocation.