Research On Dynamic Spectrum Management Technology Based On Game Theory

As a scarce resource, spectrum has always been the focus of wireless communications. With the rapid development of new wireless technology, many optional wireless access technologies has emerged and will coexist for a long time in prospective communication networks. Frequency, space, time, and coding reuse have been proposed for the scarcity problem of spectrum, but the frequency bandwidth as the bottleneck is still widespread. With the rapid development of global mobile internet services, the rapid growth of broadband users and doubled network traffic, the scarcity problem of spectrum resources will become more and more serious, and the conflict between limited frequency resources and the growing services will be increasingly prominent. However, the frequency resources have different degrees of idleness in time and space dimension because of the fixed spectrum allocation system, even in some more crowded bands, and the actual frequency utilization is not high ever. Therefore, it is necessary to study the more efficient dynamic spectrum management technology to achieve the maximizing of spectrum utilization.This thesis presents a dynamic spectrum management scheme based game theory, which mainly studies around spectrum sharing strategy, aims to improve the frequency utilization and the maximize the spectrum revenue. The main content of this thesis include:(1) creating a dynamic spectrum sharing model based on utility function in game theory and proposing a method of compensation for spectrum using based on game theory; which to be aimed at core technical issues of dynamic spectrum management for the core technical issues:spectrum sharing strategy between users,(2) from the perspective of the utility function, comparing the different utility functions respectively based on the analysis of maximizing spectrum efficiency, maximizing system throughput, minimizing system interference level and ensuring the fairness of users。(3) To solve the limitations of the existing sharing models based on Cournot game, an improved cyclical borrower competing model is proposed, which studies the optimal pricing method to share spectrum between the spectrum providers and loan parties under the network environment.(4) An optimized spectrum sharing algorithm is designed, through introducing eager coefficient, it can execute quantitative assessment for the extent of competition of the network spectrum resources. So as to take a targeted spectrum sharing strategy according to the extent of competition for the network spectrum, which can achieve the maximum spectrum utilization. This paper focused on the strategy of spectrum leasing user who request bandwidth, which is based on the static and dynamic game conditions respectively, and technical methods of achieving the maximum spectrum gain. Finally, we do some simulation and analysis on the static game of the algorithm and the impact of various parameters on the best revenue functions in MATLAB environment.The results show that the spectrum sharing strategy based game theory is relatively fair for the majority of the network spectrum leasing users. Which suit for the overall interest of spectrum utilization in communication system also can effectively reduce the interference level. At the same time, through the loan game model in the spectrum transaction, simplifies the complexity of the network spectrum sharing mechanism between users. Also, it can achieve dynamic spectrum sharing between multiple users. The spectrum loan and transactions between the spectrum providers and loan parties is constrained by maximizing the utility function of the spectrum proceeds. This lease the disorderly conduct between the spectrum loan parties, and maximize the maintenance of the network environment and the fairness of spectrum sharing.