Spectrum Sensing with Largest Eigenvalue in Cooperative Cognitive Vehicular Networks

Abstract

Cognitive Radio is a novel technique to solve spectrum scarcity in vehicular networks, and spectrum sensing is the initial step to realize Cognitive Radio. Cooperative spectrum sensing, which obtains the final sensing result from the detection decisions of multiple vehicles, has been commonly discussed to improve the sensing reliability in Cognitive Vehicular Networks, where energy detection method has been widely used due to the simplicity. However, eigenvalue based detectors, such as largest eigenvalue detector, can also improve the sensing performance by taking advantage of multi-antenna gain. This paper proposes a cooperative sensing architecture combining the largest eigenvalue detector and hard decision fusion to guarantee the sensing accuracy, and analyzes the detection performance. We give the Gaussian approximation and simulation results for the detection performance.