Research Of Dynamic Spectrum Allocation In Cognitive Radio

With the dramatic increase in the wireless communication traffic, the phenomenon of wireless spectrum relative deprivation appeared. Some band transmission overload, but some other band have no business to transfer. In order to solve the contradiction between them, an intelligent wireless communication was come up; it is called cognitive radio technology. The cognitive radio technology in the ability to sense changes in the wireless network environment, using artificial intelligence technology to learn from the environment through real time changes in transmission parameters, such as the band, transmission power, or modulation, so as to achieve reliable protection anytime, anywhere communication and full use of spectrum resources.The basic idea of cognitive radio is to occupy the spectrum opportunistically without affect the authorized user in the case of normal communication. So our studies focus on the critical technology of dynamic allocation of spectrum resources. In this paper first introduces the system model of cognitive radio technology, work processes; after reading and analyzing lots of relevant literature, the work started by two technical routes, namely the application of evolutionary computation and the economics of game theory to solve the problem of spectrum allocation; finally we propose and improve two dynamic spectrum allocation models, including the use of evolutionary computation theory which is mainly used in centralized network structure and the use of potential game theory which is applied to distributed network structure.In the centralized network structure, we mainly used evolutionary computation theory to solve the dynamic spectrum allocation problem. Specifically, we analyzed the background of the problem, proposed the use of two variables estimation of distribution algorithms to model the problem, design and implementation of the algorithm, and improve the problem of convergence too early, the last to prove the theoretical correctness of the algorithm.In the distributed network structure, the current major is to use game theory to solve the spectrum allocation problem. We give a brief introduction of game theory and related models of game theory, and improve one of the potential game models. In the course of the game, the benefits function is the only basis to select. It is enough to see the design and calculation of benefit function is the key to solving the game. Our innovative application of DS evidence theory is proposed to calculate the benefit function, not only superior to the traditional weighting operation, but also facilitate the function of benefit expressed in expanding the physical meaning.Cognitive radio technology is one of the revolutionary technologies of wireless communication in recent years. The whole system of cognitive radio technology in the various modules, including spectrum sensing, analysis, and decision makings make the scholars to do a lot of hard exploration. Dynamic spectrum allocation problem is one of the key technologies of cognitive radio system, I hope this study of this paper can play a facilitating role.