Research On Railway Cognitive Spectrum Sharing Based On Game Theory

With the advancement of digitalization,informationization and intelligence construction of railways,the demand for wireless communication services has continued to expand,and the development trend of multimedia and broadband has been presented.However,both the existing GSM-R and the upcoming next-generation railway mobile communication system LTE-R face the problem of shortage of radio spectrum resources,which will greatly limit the carrying capacity of the system,so that the railway wireless communication demand cannot be fully met.The market-driven cognitive radio spectrum sharing mode provides a feasible way to solve the problem of spectrum shortage: the railway wireless communication network leases the temporarily idle spectrum resources of other radio networks(systems)to make up for the shortage of licensed spectrum.There are usually multiple primary/secondary users in spectrum sharing,and the same kind of users tend to be competitive relationships and game theory is the most effective mathematical tool for modeling and analyzing such competitive relationships.Based on game theory,this thesis studies the market-driven cognitive radio spectrum sharing problem in railway wireless communication.The main work and innovations are as follows:(1)Two main obstacles to the application of cognitive radio in railway communication are summarized:(1)the high mobility of the train leads to time-varying non-stationary channel characteristics and severe doppler frequency shift,which makes spectrum sensing degraded in performance or even unavailable;(2)in the general spectrum sharing mode of "opportunistic utilization of spectrum holes",the interference of secondary users to primary users is usually difficult to avoid completely.In addition,spectrum resources are important and scarce,and primary users often lack the willingness to share the spectrum.Pointed out a feasible solution that can overcome the above two obstacles,namely the market-driven cognitive radio spectrum sharing mode.(2)Aiming at the railway cognitive spectrum sharing scenario with multiple primary networks and a single secondary network,a spectrum sharing algorithm of primary network game based on the approval degree is proposed.Firstly,considering the selection preference of the secondary network under the horizontal differences of spectrum and the reputation degree of the primary network,which are two important factors affecting spectrum sharing.By introducing the approval mechanism,the spectrum sharing model based on Bertrand game is improved.A new utility function of secondary network is proposed,so that the new game is formed,and the existence of its Nash equilibrium is proved.Then the spectrum leasing framework is constructed,the algorithm cycle and phase are defined,the approval update mechanism is proposed,and the dynamic updating of the approval parameters is realized.The purpose is to reflect the willingness of the secondary network as a buyer in different algorithm cycles,and punish the dishonest behavior of the primary networks,so as to optimize the spectrum sharing environment.Finally,the algorithm is extended to the case of dynamic game and its local steady state is discussed.The simulation results show that the algorithm adapts to the actual situation and improves the rationality,fairness and effectiveness of spectrum sharing.(3)Aiming at the railway cognitive spectrum sharing scenario with a single primary network and multiple secondary users,a spectrum sharing algorithm of secondary user game based on the priority is proposed.The problem of spectrum sharing between primary network and secondary users is modeled as a transaction problem in monopoly market: the primary network acts as the only spectrum seller to obtain the maximum benefit by formulating the best spectrum price,and all secondary users allocate the spectrum leased by primary network in a competitive way.The strategy of each secondary user is to determine its own spectrum request bandwidth,with the goal of maximizing its own utility.Firstly,the satisfaction function of the secondary network and the income function of the primary network are designed,and the optimal spectrum pricing function of the primary network is derived.Then the priority mechanism of the secondary user is proposed.The calculation method of the priority weight(including the importance weight and the real-time weight)based on the Saaty scale is given,and the utility function of secondary user with the priority parameter is designed.Finally,the corresponding dynamic game algorithm is proposed and its stability is analyzed.Simulation results show that the algorithm can achieve reasonable,effective and flexible sharing of spectrum.