Research On Spectrum Sensing Algorithms In Cognitive Radio

With the innovation of communication technology, wireless communicationtechnology got rapid development. At the same time the number of users usedwireless communications services are also surged, and thus the demand for wirelessspectrum resource becomes more and more strong. Due to the unreasonable of theexisting spectrum allocation policy, the spectrum utilization ratio is low. In order tosolve the contradiction between the low frequency spectrum efficiency and thegrowing demand for spectrum, cognitive radio technology as a new technology wasproposed. The accurate and effective sensing for the target spectrum is the foundationof cognitive radio technology.Firstly, the chapter one in this paper introduces the concept of cognitive radioand the spectrum sensing technology in cognitive radio. Then the signal detectiontechnology which included the binary signal hypothesis testing, performanceparameters of the signal detection and judgment criteria was introduced in the chaptertwo. Then some detection algorithms in this article are introduced, such as ED, CAV,MME and EME.Secondly, because energy detection algorithm and other classical detectionalgorithm didn’t use the correlation among the signals, the chapter3of this paper putsforward a kind of spectrum sensing algorithm based on statistical noise covariancematrix， and the algorithm makes full use of the received signal correlation. Themathematical basis of the proposed algorithm is the difference between the receivedsignal sample covariance matrix and the noise covariance matrix. For the existingdetection algorithms will be affected by noise uncertainty, the chapter has modifiedthe model which proposed in this chapter to eliminate the influence of the noiseuncertainty. Last, the theoretical decision threshold of algorithm is derived usingmultivariate statistical theory. Simulation results verify the effectiveness of theproposed algorithm.Finally, because these detections based on energy detection algorithm aresusceptible to the influence of the noise uncertainty and low signal-to-noise ratio,furthermore, the chaotic system have immunity to noise and sensitive to the initialvalue. The chapter4puts forward a weak primary user signal detection algorithmbased on chaos theory. The chapter simply introduces the detection model and theprinciple of Duffing system, and then introduces how to utilize the maximumLyapunov exponent to judge the existence of the weak primary user. From thesimulation results show that the detection performance of the proposed algorithm inthis paper is better than energy detection algorithm, and the impact of noiseuncertainty and low signal-to-noise ratio in this paper are smaller than the energydetection.