Research On Blind Detection Of Spectrum Technology In Cognitive Radio Sensor Networks

Wireless sensor networks(WSN)receive much attention because of its widely applications,but the problem of spectrum resources shortage limits its development.Cognitive radio sensor networks(CRSN)combine cognitive radio(CR)technology with wireless sensor networks and provide a solution to this problem.In order to access the authorized spectrum without interrupting the communication of primary user(PU),cognitive nodes of cognitive radio sensor networks need to constantly perceive the surrounding radio environment to detect whether the surrounding spectrum resources are occupied by the primary user or not.Therefore,spectum sensing is not only an important basis for cognitive radio technology,but also a significant function of distinguishing cognitive radio sensor networks from wireless sensor networks.However,there are still some challenges of researching on spectrum sensing detection technology in cognitive radio sensor networks.Firstly,existing algorithms using compressive sensing(CS)theory need to reconstruct the original signal which have high computation complexity.Sencondly,the conventional energy detection(CED)method requires a large number of samples and the detection performance is susceptible to noise fluctuation at low signal-to-noise ratio.In order to solve the above problems,this thesis presents two spectrum detection algorithms:(1)A high-order statistic(HOS)based compressive wideband spectrum blind detection(HOSCWSBD)scheme is proposed.This algorithm utilizes the compressive observation data to reduce the amount of data required for sampling,and then chooses high-order statistic as the decision data of spectrum detection.It is not necessary to recover the original signal,can reduce the computation complexity,and has a good detection performance without any prior knowledge about the primary user.(2)A double-threshold cooperative spectrum sensing in small sample energy detection scheme is proposed.This algorithm does not need to know any prior knowledge about the primary user and reduces the interference of the cognitive user to the primary user at low signal-to-noise ratio by using double thresholds.It overcomes the problem that the conventional energy detection needs a large number of samples to transmit datas which leads to the high energy consumption of cognitive radio networks nodes by making use of the Cube-of-Gaussian(CoG)approximation approach to deal with the detection results.In the fusion center(FC),it uses the “majority-voting” rule to make the final decision and improves the detection performance of the whole system.