Cooperative Spectrum Sensing In Cognitive Radio Technology

This paper mainly studied the spectrum sensing in cognitive radiotechnology. In practice, the key issue of cognitive radio networks is howto find spectrum holes and identify authorized users. Therefore,spectrum sensing as a key technology to solve this problem plays animportant role in the realization of cognitive radio.In order to make more image in cognitive radio technology, thispaper made a simple experiment to illustrate the application ofcognitive radio technology. Then introduced some existing single-nodespectrum sensing algorithm, and the advantages and disadvantagesof various algorithms are also described and compared.In this paper, firstly improving traditional cooperative sensing withtwo methods. One is based on D_S evidence theory approach,One isexponent weighted based on SNR. the complexity of the traditionalweighted collaboration method can obviously reduced by fusingweighted coefficient with D_S evidence theory.the affects of low-SNRusers can be reduced by the method of exponent weighted based onSNR, the role of high-SNR users also can be given full play,so as toimprove the detection performance and noise immunity of the system.In order to reduce the adverse effect of fading channel todetection result,and improve detection performance, cluster-basedcooperative spectrum sensing with double adaptive energy thresholdsand optimal data fusion is proposed in this paper. In this method,theC-V data fusion criterion is firstly introduced,and then calculates themethod of determining the threshold in the adaptive dual-thresholdmethod, describes the using process of optimal data fusion incooperative sensing.The C-V data fusion is an optimal data fusion,moreaccurate results can be get with this data fusion.The detection threshold can be changed according to ambient in the adaptivedual-threshold method,enhancing the adaptability of system.At thesame time,the uncertain value between the two thresholds can beremoved, on one hand the adverse effect of uncertain values can bereduce, on the other hand the processed data can be reduced. Thescheme also introduces the clustering mechanism, selecting thecognitive users with the best channel quality as cluster heads,the datain the same cluster first made in the cluster head, then the results of thecluster head will be spread to the fusion center,the errors in thetransmission and the amount of data can be reduce by this method,obtaining good diversity gain. Simulation results show that,in the -26dBcase, the detection probability can be increased 10%,when usingadaptive dual-threshold method than using single threshold method;the detection probability can be increased 20%,when using C-Vcriterion than using half criterion.