| The development of Intelligent Transportation System(Intelligent Transportation System,ITS)requires high level techniques upon vehicular networks.Various sensors deployed on vehicles need to transport a large number of sensing information in real time,and at the same time all kinds of vehicular services(including the traffic safety services,traffic efficiency services,vehicular information services and on-board entertainment services,etc.)grow explosively.These require vehicular communication to support Gbps data rate.However,wireless spectrum resources assigned to vehicular networks cannot meet this requirement due to the limited bandwidth.There is large bandwidth at millimeter wave(10GHz-300GHz)which can support high data rate and low delay transmission,so mm Wave(millimeter wave)communication becomes a key candidate technology for vehicular communication in the future.Fixed assignment of spectrum resources makes the mmWave spectrum resources used inefficiently.Spectrum sensing of cognitive radio(CR)can acquire the state of spectrum(idle/occupied)to realize dynamic spectrum sharing.Therefore,it is necessary to study spectrum sensing in vehicular networks where vehicles use mmWave spectrum to communicate.This thesis is supported by Chinese National Nature Science Foundation Project "Sensing and utilization of polarization resource in dense heterogeneous wireless environment"(No.61571062).Based on the characteristics of mmWave signals and vehicular mobility in vehicular networks,algorithms of mmWave spectrum sensing in cognitive vehicular networks are studied in this thesis.A correlation based mmWave spectrum sensing algorithm is proposed for local sensing and an optimal cooperating users and D-S fusion based cooperative spectrum sensing algorithm is proposed for cooperative sensing.The main research contents and contributions of this thesis are as follows:As for local mm Wave spectrum sensing in cognitive vehicular networks,exiting algorithms do not build spectrum sensing model according to the characteristics of millimeter wave signal and cannot adapt to mobility of vehicles in vehicular networks.To solve this problem,a correlation based mm Wave spectrum sensing algorithm is proposed in this thesis.According to the beamforming technology of mm Wave and the interference and noise limited regime of mm Wave,the expression of the signal received by the cognitive user is analyzed and the binary hypothesis model of spectrum sensing is established,and the test statistics are derived by likelihood ratio.Then,the new observation variable is constructed according to the correlation as a result of vehicular mobility,and new test statistics are derived to realize the detection based on the test statistics of multiple sensing cycles.The probability of false alarm and the probability of detection which can be realized by this algorithm are deduced theoretically.Finally,the performance of the proposed algorithm is verified by MATLAB simulation,and compared with the performance of existing mm Wave spectrum sensing algorithm based on random process.In the simulation scenario of this thesis,while SNR is 0dB,the probability of detection of the proposed mm Wave spectrum sensing algorithm can be 0.99.As for cooperative mm Wave spectrum sensing in cognitive vehicular networks,it is difficult for exiting algorithms to choose optimal cooperative users and to cope with the uncertainty of local sensing.To solve this problem,a cooperative spectrum sensing algorithm based on optimal cooperative users and D-S fusion is proposed in this thesis.First of all,according to the similarity of sensing results of cognitive users and similarity of the group mobility of cognitive users,users that participate in cooperation are chose.The maximization problem for choosing optimal cooperative users is established by maximizing similarity of sensing results of cognitive users while ensuring the group mobility of cognitive users to satisfy the restricted condition.The optimal cooperative users are obtained through adaptive genetic algorithm.Then,based on the uncertainty of noise and interference,the uncertainty of the local sensing of cooperative users is calculated.Based on this uncertainty,the fusion center fuses sensing results from cooperative users through D-S fusion and then makes a global decision.Finally,the performance of the proposed algorithm is verified by MATLAB simulation.Compared with the existing cooperative perception algorithm,this algorithm has better performance.In the simulation scenario of this thesis,while SNR is-10dB,the probability of detection of the proposed mmWave spectrum sensing algorithm can be 0.90 and while SNR is-5dB,it can be 0.99. |
PDF Full Text Request