Research On Wideband Compressed Spectrum Sensing In Cognitive Radio Networks

In modern society, The need for communication resource and spectrum resource growsdramatically day by day. Limited spectrum resources is becoming a major bottleneck restricting thedevelopment of wireless communication. In order to solve the problem of spectrum scarcity, toachieve dynamic access of spectrum and improve the spectrum utilization, cogonitive radio isproposed. By the perception of the wireless communications environment and automaticallylearning, cognitive radio technology can adaptively change the transmission parameters to achievedynamic spectrum access. Cognitive radio is considered to be a key technology of the nextgeneration of wireless communication network. Spectrum sensing is the first task in cognitive radiotechnology. In one spectrum sensing model, the channel is divided into a plurality of sub-bands anddetect each of the sub-channel, but this technique is too inefficient. Wideband spectrum sensingtechnology can increase the detection efficiency but with too much complexity to sample the signalat Nyquist rate, and sometimes such a complexity and cost is too high to afford in reality.Compressed sensing theory break through the limitations of the Nyquist sampling theorem byproposing a new method of information sampling. Distributed spectrum sensing model caneffectively enhance the performance of spectrum sensing and the spectrum sensing performancedirectly affect the efficiency of cognitive wireless networks, so in this paper we also consider thesecurity issues in spectrum sensing. This paper studies the following three questions:First, based on the inherent sparsity of frequency domain signals in cognitive radio network,this paper proposed distributed wideband compressed spectrum sensing model combiningcompressed sensing technology and spectrum sensing technology as well as the weighted averageconsensus algorithm.Second, spectrum sensing security issues is analysed in distributed wideband compressedspectrum sensing model. In the data fusing stage，on the purpose of attacking the process of spectrumsensing, malicious node sends misleading data to its neighbor CR nodes. To defend against such typeof attack, a distributed wideband compressed spectrum sensing model is proposed in this paper.Third, A new wideband SSDF attacking model is proposed. In this type of attack, a minority ofsub-bands of the wideband spectrum range are attacked randomly by malicious node on the purposeof attacking the process of spectrum sensing at the very sub-band range without being noticed. Todefend against such type of wideband SSDF attacks, a distributed wideband compressed spectrum sensing model is proposed. This spectrum sensing model can defend against wideband SSDF attacks and at the same time achievebetter spectrum sensing utilizing the useful data sent out by malicious nodes.