The Research On Spectrum Sensing Technology Based On Particle Filter

Nowadays,spectrum sensing algorithms are used recursively for convergence forgetting to probe further into the computational complexity.Based on the Hidden Markov model,the paper proposes the algorithm which is made of Particle Filtering and Bayesian approximation.This modified algorithm mixes the advantages of these two algorithms.The new Mixed Filter can make a great balance between calculation complexity and time complexity.During the process of the algorithm,this paper calculates the running time and the accuracy of the results.In addition,the later part discusses the different environment of application.One the one hand,only in Hidden Markov model,the algorithm simulates the state of sub-band in mathematical model.On the other hand,the algorithm will be used in the Hidden Markov model supported with Autoregressive model to simulate the fading channel.The results show that,comparing with Particle Filtering,the Mixed Filtering gets the same accuracy and can reduce the running time.