Research On Cooperative Spectrum Sensing In Cognitive Radio Networks

As the rapid development of the wireless communication, the need of spectrumresource is increasingly tense day by day. The conflict between the low utilization of theLFB and the increasing demand has hindered the development of the wirelesscommunication. As a smart spectrum shareing technology, the cognitive radiotechnology has effectively relieved the lack of spectrum resource.Spectrum sensing, which is one key technology of cognitive radio, has beenconsidered as a hot spot. After the Analysizing and summaring the related researches athome and abroad, spectrum sensing was mainly studied in this paper. And then deepresearches were done upon the cooperative spectrum sensing based on energy detectionand data fusion algorithms in cooperative sensing. The main research work in this paperis as follows:①Three typical algorithms including energy detection, matched filter detectionand cyclostationary feature detection were researched. And the difference among themwas analysed through simulation. Besides, a popular algorithm, energy detectionalgorithm, was focused.②Cooperative spectrum sensing based on energy detection was studied. The hardcombination with “or” criterion and “and” criterion were analysed, and the analyzingresults showed this: lower missed detection probability under “or” criterion; lower falsealarm detection under “and” criterion. The effect on spectrum detection performance byRayleigh channel, Rician channel and Nakagami channel was researched and analysed，as well as the variable fading factors. The simulation proved the superiority ofcooperative sensing under fading environments, and some valuable conclusions wereobtained.③Typical data fusion algorithms in cooperative sensing were studied andcompared, then advantages and disadvantages of each algorithm were summaried.Combining the advantages of decision fusion and data fusion, an optimal algorithmagainst DSNR-WC algorithm was proposed by indicating the distance and SNRinformation with5bit. Under the mostly same detection performance, the optimalalgorithm was employed to reduce the bandwidth of the control channel, lower thecomputational complexity and enhance the detection efficiency. The feasibility and effectiveness of proposed algorithm were porved by comparative simulation analysis atlast.