User correlation and double threshold based cooperative spectrum sensing in dense cognitive vehicular networks
Siting Zhu, Caili Guo, Chunyan Feng, Xia Liu · 2016
Spectrum scarcity is becoming increasingly serious in vehicular networks due to the deployment of the Intelligent Transportation System. Cognitive Radio (CR) is a promising technology to overcome this problem. However, as the traffic density is always rising, the higher correlation index of the network and the increasing number of vehicles cause new problems for CR. The deterioration of sensing performance caused by correlated users and huge signaling overhead of the vehicles should not be ignored. In this paper, we consider a Dense Cognitive Vehicular Network (DCVN) and propose an user correlation and double threshold based cooperative spectrum sensing algorithm (UCD-CSS) to achieve a trade-off between sensing performance and system overhead. First, we employ an improved double threshold (DTH) method to reduce system overhead and withstand noise uncertainty. Moreover, a correlation based cooperative spectrum sensing (CSS) algorithm is established to avoid the performance deterioration caused by correlated observations in DCVNs. The probability of detection, the probability of false alarm and the average sensing bits are analyzed. In addition, the impacts of traffic density and failed reporting probability on performance are further demonstrated through simulations. The results clearly reveal the benefits of adopting the proposed UCD-CSS algorithm in dense vehicular networks.