Research On Spectrum Allocation Based On Particle Swarm Optimization Algorithm And Genetic Algorithm

With the wide application of wireless communication technology in various offields, the demand of spectrum resource of devices with the frequency is increasing, sothe shortage of spectrum resources has become a bottleneck in the development ofwireless communication technology. Wireless spectrum is a scarce and nonrenewableresource, the rational use of it is very important. As a technology to improve theutilization of spectrum resources, cognitive radio is put forward in such environment.Its starting point is in the premise of not affecting the normal communication ofauthorized users, making cognitive users access to the idle spectrum and taking use ofit dynamically; Its core is to discover and use the idle spectrum resources, as well asimprove the overall utilization rate of spectrum resources; The problem is due to thewaste of resources caused by the fixed spectrum allocation. How to efficiently allocatethe idle spectrum has been found, which is an important research direction. So, aroundthis hot topic, we begin our study in this paper.First of all, this paper reviews and summarizes the electromagnetic spectrumallocation of cognitive radio technology and introduces several common distributionmodels. In the distribution of technology:(1) network is divided into the centralizedand the distributed according to the different communication network architecture;(2)it is divided into the cooperative and non-cooperative based on the internal networkusers who have exchange and feedback information or not.(3)this is divided into staticand dynamic allocation based on the real-time adjustment of distribution. In thedistribution aspect:(1)The one is divided into different Overlay sharing model andUnderlay model based on the access mode;(2)The one is divided into interferencetemperature model and graph model based on the interference of the attitude;(3) Theone is divided into a model of game theory and auction model based on the fairness ofthe cognitive users in the network. Secondly, in fact, spectrum allocation problem is the optimization of resourceallocation problem, this paper based on the graph theory model, use the swarmintelligence optimization algorithm namely particle swarm algorithm(PSO)and geneticalgorithm(GA)to solve the spectrum allocation problem, make the network efficiencyand fairness as the measurement reference performance index. According to theprinciple of particle swarm algorithm and genetic algorithm, and the planning of thealgorithm process, as well as algorithm realized in the spectrum allocation problem,combined with the simulation results, this paper teases out the advantages and thedisadvantages of the particle swarm optimization (PSO) and genetic algorithm(GA) onthis issue and make some analysis.Finally, this paper constructs a fusion algorithm (PSO-GA)based on particleswarm optimization algorithm and genetic algorithm, In view of the problems that theprematurity and the low convergence speed of traditional particle swarm algorithmand genetic algorithm in the practical application of the electromagnetic spectrumresource allocation, This paper embeds the evolution thought of the crossover andmutation in genetic algorithm operation into particle swarm algorithm, and introducesthe linear inertia weight function over the weight parameters into the fusion algorithm,so as to improve the limitation of these two algorithms in the frequency spectrumresource allocation, and contrast the performances of the three algorithms throughsimulation experiments. Ultimately this paper has superiority in the overall networkperformance and users’ fairness.