Study On Cooperative Spectrum Sensing Techniques In Cognitive Vehicular Networks

With the rapid development of intelligent transportation,the demand of communication for vehicular network has also increased geometrically,and the spectrum resource is facing a serious shortage.Cognitive radio technology can enable unauthorized users to access the authorized frequency band dynamically without interfering the nonnal communication of authorized users.It can make full use of the spectrum resource and change the situation of spectrum shortage.There is a dense traffic scenario in urban traffic,which makes the received signals of vehicles experience similar shadow fading,and makes the cooperative spectrum sensing get strong correlated results,which may impair the spectrum sensing performance.Therefore,it is necessary to select the weak correlated sensing vehicles for cooperative decision.In this paper,cooperative spectrum sensing technology in a dense vehicular scenario is studied.Two different cooperative spectrum sensing schemes are proposed for dense vehicular networking scenarios with single antenna and multiple antennas.The achievements and contributions of this paper are summarized as follows:(1)For a single antenna cognitive vehicular networking scenario,a cooperative spectrum sensing scheme based on vehicular correlation coefficients and weighted signal-to-noise ratios(SNRs)is proposed under urban dense traffic conditions.Firstly,the correlation coefficient matrix is established by using the velocity distribution model in dense vehicular scenario,the time and space spectrum opportunities are also considered.Secondly,using dual threshold energy detection technology,together with the method of combing the weight of vehicular correlation coefficient and signal-to-noise ratio,vehicular cooperative spectrum sensing is carried out.Finally,simulation is completed under different channel conditions and fusion algorithms.The results of the simulations show that the proposed scheme can achieve better detection performance.(2)For a multi-antenna cognitive vehicular networking scenario,a cooperative spectrum sensing scheme based on vehicular correlation coefficient and covariance absolute value(CAV)of received signals is proposed under the condition of dense urban traffic,which effectively solves the problem that detection performance is greatly affected by noise in traditional spectrum sensing schemes.Firstly,by comparing the performance of four single vehicular multi-antenna spectrum sensing algorithms,such as energy detection,maximum and minimum eigenvalue detection,CAV detection and arithmetic geometric eigenvalue detection,a more suitable CAV detection algorithm for a multi-antenna scenario is established.Secondly,CAV is applied in vehicular local multi-antenna sensing.By comparing the results of energy detection,a more effective multi-antenna vehicular cooperative spectrum sensing scheme is proposed.The results of the simulations show that the proposed cooperative spectrum sensing scheme can reduce the false alarm probability when the detection probability is equal.