The Application Of Swarm Intelligence Optimization Algorithm In Spectrum Allocation

With the development and progress of wireless communication and popularization of the wireless terminal,the more and more radio spectrum resources are required.But the number of the radio spectrum resources,which can be used,are certain.To meet the demand of spectrum resources,the method of current spectrum allocation must be changed.Aiming at the actuality of spectrum allocation,the cognitive radio network technology was come forward.The secondary user is sensing the vacant spectrum,to obtain the available spectrum resource.When the secondary user doesn't interfere the primary user,the spectrum allocated for the secondary users by the system.The spectrum allocation of cognitive radio network is to improve the utilization of spectrum resources,and to achieve the best performance of network.In the other world,the spectrum allocation will make for the network utility and the user fairness are best.The spectrum allocation of cognitive radio networks is similar to the combination optimization problem.So the spectrum allocation can be solved by the swarm intelligence optimization algorithms,which are used to solve the NP hard problem.Firstly,the cognitive radio network and the swarm intelligence optimization algorithms are studied.The key technologies and the research situation of the cognitive radio network are learned.The spectrum allocation is further studied,to analyze and compare the different model,such as the game theoretic model,the auction model,the graph coloring model and the interference temperature model.The fruit fly optimization algorithm and the artificial bee colony algorithm are researched intensely.The feasibility of spectrum allocation is analyzed.The optimization goal of swarm intelligence optimization algorithms for spectrumallocation is established.Then,the quantum fruit fly optimization algorithm is used for the spectrum allocation.The quantum concept is introduced into the fruit fly optimization algorithm,to obtain the quantum fruit fly optimization algorithm.The quantum fruit fly optimization algorithm is improved by the operations,which are selection,crossover and mutation.So the convergence rate is improved,the sample population diversity is increased,and the algorithm falling into local optimum is avoided,and so on.The improved quantum fruit fly optimization algorithm was used to search the quantum sequences of spectrum allocation,seek the network utility or the user fairness and obtain the spectrum allocation strategy,which gains the best network performance.The simulation shows the promising results: the improved quantum fruit fly optimization algorithm has fast convergence rate and strong ability of avoiding local optimum.And the utilization of network resources is increased,the network performance is improved,after using the algorithm for spectrum allocation in cognitive radio network.Finally,the quantum artificial bee colony/fruit fly optimization algorithm fusion is applied to spectrum allocation based on the quantum fruit fly optimization algorithm.A new objective function is proposed by combining the network utility with the user fairness.The fruit fly optimization algorithm and the artificial bee colony algorithm are fused,not only is the global optimizing capacity of the artificial bee colony algorithm absorbed,but also the strong local searching ability of the fruit fly optimization algorithm is remained.The adaptive concept is introduced,to control the step size of fruit fly and to improve the diversity of the population.At the same time,the artificial bee colony/fruit fly optimization algorithm fusion is quantized,and it is used for the spectrum allocation.The simulation verifies the new objective function of the spectrum allocation,and proves the superiority of the quantum artificial bee colony/fruit fly optimization algorithm fusion for spectrum allocation.Not only is the search ability enhanced,but also the network performance of the spectrum allocation isimproved.