Research On Spectrum Sharing Algorithm Of Cognitive Network Based On Game Theory

At present,most of the spectrum resources are idle under the fixed allocation mode,and the spectrum utilization ratio is low,which can not meet the increasing spectrum demand.Cognitive radio(CR)technology has been proposed to solve the contradiction between the shortage of spectrum resources and the low utilization rate.CR can search the spectrum resources in the current radio environment can search the spectrum resources in the current radio environment and realize the dynamic use of the spectrum,so as to improve the spectrum utilization.Spectrum sharing in CR technology mainly includes the quality of communication service(Qo S)of authorized users,the efficiency of authorized users to rent spectrum,the number of cognitive users to rent spectrum,and the efficiency of cognitive users to obtain spectrum.In the spectrum sharing scenario,how to allocate the spectrum and how to compete among cognitive users under the condition of existing spectrum is the key to solve the spectrum sharing problem.Game theory is a mathematical tool to help analyze decision selection problems.Game theory can be used to study and analyze the problems in spectrum sharing scenarios.In this thesis,based on game theory,we model spectrum sharing scenarios,analyze Nash equilibrium,game strategy and user benefit in spectrum sharing games,and propose a cognitive network spectrum sharing algorithm based on sub-user grouping.The results show that the proposed algorithm can improve the rationality and effectiveness of spectrum allocation while ensuring the stability of the system.The main contents and achievements of the thesis are as follows:(1)In the spectrum sharing model based on game theory,the pricing method and benefit function design adopted by authorized users should not only consider the number of leased spectrum,but also consider the influence of other factors such as channel quality.Otherwise,the design of authorized user benefit function is unreasonable,which makes authorized user have low willingness to rent spectrum and can not share spectrum resources.A non-cooperative game model is established based on the cost function and benefit function of authorized user system under the premise of satisfying the service quality of the main user.Through the game algorithm,the spectrum sharing model composed of an authorized user system and several cognitive users is simulated and analyzed,and the influence of channel quality,cost weight,learning factor and other factors on spectrum sharing is obtained.The simulation results show that the Nash Equilibrium is achieved when the cognitive user benefit reaches the maximum,and the different functions of different factors in the spectrum sharing game are proved.(2)In this thesis,cognitive users are grouped according to the requirements of spectrum resources and business types,and the average spectrum efficiency factors of each group are obtained,and then the cognitive user benefit function is improved.The spectrum competition between cognitive users is modeled as dynamic Cournot game.The simulation results show that the proposed algorithm can improve the rationality and effectiveness of spectrum allocation while ensuring the stability of the system.