Research On Spectrum Sensing For Cognitive Radio System

In recent years, with the rapid development of wireless communication, Spectrumavailable for allocation is becoming a desperate shortage of resources; however, theutilization of the allocated spectrum is very low. In this background, this new concept ofcognitive radio is proposed and attracted a lot of researchers to study on it. Cognitive radiois considered to be the most important scheme to solve this problem. Spectrum sensing isone of the key technologies to ensure the normal application of cognitive radio. Since thenoise uncertainty and the primary signal unknown to the secondary user. The traditionalspectrum sensing algorithm is difficult to satisfy the need of this new communicationtechnology. Detection of spectrum holes under little knowledge of the signal informationbecomes a hotspot in this field.Currently, Researchers has proposed corresponding spectrum detection algorithmswhich focus on the three directions as followed: The primary transmitter based spectrumsensing algorithm; the secondary users cooperation based spectrum sensing algorithm andinterference temperature based spectrum sensing algorithm. The primary transmitter basedspectrum sensing algorithm and the secondary users cooperation based spectrum sensingalgorithm become the mainstream direction of the current spectrum sensing methods. Itgets more attention and lots of research achievements have made by researchers. Forinstance, the primary user transmitter based spectrum sensing algorithm is mainly aboutenergy detection, matched filter detection, cyclostationary detection, wavelet detection andeigenvalue detection. Although these algorithms can solve a certain problem of spectrumsensing, but it usually depends on the environmental noise, signal characteristics or otherinformation. To solve this problem, this paper proposes lots of algorithms in differentenvironment for spectrum sensing.The research significance and background of spectrum sensing are presented in thefirst two sections of the paper. Then the current domestic and international issues ofspectrum sensing and analyzed the major problem on spectrum sensing algorithm areintroduced. In the second chapter, a brief introduction is made on the definition and theclassification of spectrum sensing. Finally, more detailed description is presented on thecurrent classification of spectrum sensing.In order to find a free frequency bands available to use, the secondary user need todetect the spectrum in a broad bandwidth, however, it must to face of many different typeof modulated primary signal, the secondary users impossibly known all the information ofprimary signal, so the detection algorithm should detect the primary signal in the scenarioof unknown the information is need to the secondary user. This paper proposed a generalized likelihood ratio test algorithm for the detection of primary with unknownsignal parameters; Neyman-Pearson criterion based test statistic is given for a restrictedprobability of false alarm, which maximized the probability of detection. In the computersimulation results, it can be seen that the proposed algorithm efficiently improves theperformance both in the scenario of the unknown signal information and the environmentof the low SNR.For the secondary user who received the primary user signal through a fading channel,a compressed sensing theory based cooperative detection algorithm is proposed in thispaper, the broadband multi-user detection system model and the channel parameterestimation algorithm are given, as well as the primary user signal estimation is given basedon the minimum Euclidean distance algorithm with the channel parameter estimated. Thenthe signal to noise ratio gain improved based on the compressed sensing algorithm toenhance the secondary user’s detection performance. Finally, the average consensusalgorithm is given to combine the detection results. Computer simulation analyze thedetection performance of proposed algorithm under different compression ratio, and thenthe curves of the signal to noise ratio gain versus compression ratio and SNR is given, andit verified the effectiveness of the proposed algorithm.With the rapid development of wireless communication technology, the demand forhigh-speed data transmission is growing, in this paper, the multi-antenna OFDM signaltransmission system is researched, according to the characteristics of the signal, the cyclicprefix correlation coefficient based detection algorithm is proposed, in this algorithm, it isgiven the expression the maximum and minimum eigenvalue of covariance matrix. The teststatistic is given, which based on the ratio the maximum and minimum eigenvalue, and thecorresponding detection probability and the false alarm probability is also given. Finally,the detection performance of proposed algorithm is verified through the computersimulation results.Hidden terminal is a urgent problem that the cognitive radio systems must be solving,since the hidden terminal problem always induced by the serious channel fading ofwireless communication environment or the shadowing of the buildings, which bring thedifficulties of the spectrum detection. To solve this problem, this paper proposed anon-fixed cluster based cooperative spectrum sensing algorithm. The non-fixed cluster iscomposed by the sensing nodes around the secondary user, and all the sensing nodescollaborate to complete the spectrum sensing with the secondary user, due to the non-fixedclusters is utilized, the detection performance has been improved and the detectionefficiency has been enhanced, finally, the computer simulation results verified the detection performance of the proposed algorithm.Finally, chapter7concludes the whole paper and look forward to the next stepresearch work.