Research On Spectrum Allocation Of Cognitive Radio Networks Based On Intelligent Optimization Algorithms

With the development of technology,the required spectrum resources are growing rapidly,but the traditional spectrum allocation method can not overcome the contradiction between the shortage of frequency resources and the increasing requirements for frequency resources.The traditional spectrum allocation model has some problems,such as low spectrum utilization and can not ensure the fairness of spectrum use among users.The graph coloring spectrum allocation model not only improves the efficiency of spectrum allocation,but also ensures the fairness of spectrum use by users.This thesis focuses on the spectrum allocation model of graph coloring theory based on intelligent optimization algorithm.The main contents are as follows:(1)The seagull optimization algorithm is analyzed,studied and implemented.Aiming at the problem that the seagull optimization algorithm has strong optimization ability,but the convergence speed is not ideal in the later stage and is easy to fall into local optimization in some cases,a graph coloring model spectrum allocation algorithm based on improved seagull optimization algorithm is proposed.Firstly,according to the location distribution of primary and secondary users and the occupation of the channel by the primary user,the available spectrum matrix,spectrum benefit matrix and interference constraint matrix are calculated through the graph colored spectrum allocation model.Secondly,the dimension l of individual population is calculated according to the number of 1 in the available spectrum matrix,and the initial population is randomly generated.Then the seagull algorithm is used to solve the population.In the iterative process,the interference constraint matrix is used to carry out interference free constraints on the solving individual,calculate the interference free allocation matrix and fitness value,and clone,mutate and select the solution of the seagull optimization algorithm to avoid the premature convergence of the algorithm to the local optimum,so as to obtain the optimal result.(2)Aiming at the problem that jellyfish optimization algorithm can well balance exploration and utilization,and can find the optimal value in less time and faster convergence speed,but may fall into local optimal solution,a tabu jellyfish optimization algorithm is proposed to solve the spectrum allocation model of graph coloring theory.In this algorithm,the population is randomly initialized according to the graph coloring spectrum allocation model,and then solved by the jellyfish optimization algorithm.In the iterative process,the interference constraint matrix is used to make interference free constraints on the solving individual,the interference free allocation matrix is calculated,and the individual fitness is solved.When the optimal value is not changing,the solution result is taken as the input,and the tabu search algorithm is used for global solution,Thus,the circuitous search is reduced,so as to maintain the exploration of different effective search paths and obtain the optimal results.The analysis and simulation results show that the two improved intelligent optimization algorithms provided in this thesis are better than genetic computing and particle swarm optimization algorithm in solving the spectrum allocation model of graph coloring theory,and obtain better results in dealing with the spectrum allocation problem in cognitive radio networks.