Study Of Spectrum Sensing On Cognitive Radio

This dissertation mainly studies the compressive sensing technology and its application in spectrum sensing for wideband cognitive radio. In the beginning, the basic concepts of cognitive radio and its recent developments are introduced. Followed by which there is a study of the traditional spectrum sensing algorithms for cognitive radio. Based on the previous introduction, a spectrum sensing method based on compressive sensing for wideband cognitive radio is deeply studied. Then a modified method as well as the corresponding simulation and analysis are presented in detail. Furthermore, the MTM-SVD algorithm combined with the compressive sensing which is used for the signal spectrum estimation and spectrum hole judgment is discussed. The distortion between the estimated spectrum which obtained from the proposed algorithm and the MTM-SVD algorithm is analyzed through the simulation, then the algorithm is tested by using actual measured GSM signal. The test results show that the proposed algorithm can achieve high probability of correct judgment with the advantages of compressive sensing. In addition, some relevant problems about the spectrum sensing mechanism are studied with the emphasis on the adjustment and optimization of detection period. A method using decay factor to adjust the sensing period dynamically is emphasized. It converts the multi-objective optimization problem into the single-objective one by using cost control factors. For the different cost control factors, the corresponding optimized initial sensing period and decay factor can be obtained by resolving the optimization problem.