Research On Channel Capacity Optimization Method In Cognitive Sensing Network

In recent years,with the rapid growth of the number of access devices in cognitive sensor networks,the traditional wireless communication technology has been difficult to meet people's demand for high-bandwidth and high-rate communication services,the channel capacity optimization method of cognitive sensor networks has become a research hot-spot.The traditional optimization method has more restrictions on the objective function,high complexity,and difficult to determine the convergence.In order to further improve the speed of optimization,optimization algorithms such as particle swarm optimization(PSO)with distributed parallel computing can be used to solve the channel capacity optimization problem.However,the standard particle swarm optimization is easy to fall into the local optimal region during the optimization process.Therefore,the thesis focuses on how to solve the premature convergence problem of PSO-based channel capacity optimization and two methods of channel capacity optimization based on improved particle swarm optimization algorithm are emphatically studied from the perspective of algorithm parameters and algorithm fusion.Firstly,the cognitive sensor network system model,signal model,the selection of the receive weight vector and some related parameters are researched.The channel capacity optimization problem is analyzed and some commonly used channel capacity optimization methods are introduced.The basic principles of channel capacity optimization methods based on gradient search and particle swarm optimization are respectively discussed and then the two optimization methods are compared and analyzed.Secondly,the method of fixed parameter of standard particle swarm optimization is changed to the way of dynamic adjustment from the perspective of balancing the global search and local search.The evaluation indexes of the convergence degree of several species are discussed concretely and the influence of each parameter of the algorithm on the search process is analyzed.Based on the current population diversity,the adjustment strategy of inertia weight and acceleration factor is given respectively,and a method of channel capacity optimization based on dynamic adjustment of inertia weight and acceleration factor is proposed.The simulation results show that using the channel capacity optimization method based on dynamically adjusting the inertia weight and acceleration factor,the optimization speed is faster,the system capacity is improved,the better symbol error performance is achieved,and a certain ability to get rid of the local optimum can be obtained.Finally,in order to further enhance the capability of getting rid of local optimum in the late stage of channel capacity optimization and improve the accuracy of search,the genetic algorithm and chaotic search method are researched and the improved strategy of introducing crossover,mutation and chaos in PSO search process is given.On the basis,a method of channel capacity optimization based on chaos genetic is proposed.From the aspects of optimization speed,optimization accuracy,system capacity and error performance,the chaos genetic-based channel capacity optimization method is simulated and compared with the method of channel capacity optimization based on PSO and the method of channel capacity optimization based on dynamically adjusting the inertia weight and acceleration factor.The simulation results show that the chaos genetic-based channel capacity optimization method not only has faster convergence speed,but also gets stronger ability to get rid of local optimal and higher accuracy of final local search,and has the best overall performance.