Research On The Spectrum Sharing Technologies Of Cognitive Radio Based On The Game Theory

By allowing cognitive user opportunity access to the spectrum under-utilized by licensee users,cognitive system architecture can effectively improve the spectrum utilization and system throughput.Cognitive radio technology is regarded as one of the most promising technologies in the fifth generation of mobile communications.It is particularly critical to study the efficient and effective spectrum sharing model for the purpose of effectively use of cognitive radio.In this paper,the latest progress of cognitive radio technology is summarized,game theory used in spectrum sharing is investigated,as well.This paper introduces the current mainstream technologies of spectrum sensing,switching,sharing and effective cognitive radio system network architecture,at the same time,introduces the classical game theory model.The research is mainly focused on the non-cooperative game model and the application of game theory and optimization theory in the field of spectrum sharing application solutions.Comprehensive development trend of the current network model,this paper formulates a novel network consists of public spectrum and private spectrum,allow cognitive users exclusively access to the leased licensee spectrum,and then transfer part of their communication load to the public spectrum.In order to maximize their utility of communication,cognitive users need to make reasonable load distribution strategy and spectrum leasing strategy,leading a complex game problem among cognitive users.In this paper,the optimal strategy selection scheme for cognitive users under this network architecture is fully studied through the non-cooperative game model and optimization theory.Nash equilibrium strategy in this non-cooperative game model is proved,the dynamic arrival process of user equilibrium strategy and the corresponding algorithm are formulated,as well.The convergence characteristics of the dynamic adaptive algorithm is verified through numerical simulations.Meanwhile,based on the policy declared by FCC,a geo-location database assist cognitive network are formulated for the effective use of TV frequency band,the best response of cognitive users is well-designed.The equilibrium characteristics of the network is fully studied.The simulation verification and system performance analysis are carried out,shows that these network frameworks can effectively extract the characteristics of the future network architectures,describe the behavior characteristics of cognitive users.These proposed network models can effectively save energy consumption and achieve a fair and effective spectrum allocation method,provide an important reference for the new network sharing mode in the future.