Abstract:
Signal classification plays an important role in spectrum sensing for cognitive radios to identify and avoid interference from other wireless devices. In this paper, we s...Show MoreMetadata
Abstract:
Signal classification plays an important role in spectrum sensing for cognitive radios to identify and avoid interference from other wireless devices. In this paper, we study a network of cognitive radios that jointly perform signal classification via cooperation. We propose a simple but effective linear cooperation scheme to fuse pre-processed measurements collected from spatially distributed cognitive radios. Our objective is to maximize the probability of successful classification subject to some constraints on the probabilities of misclassification. By applying a divide-and-conquer strategy and new constraint relaxation methods, we are able to derive the closed-form expressions for the optimal weight coefficient for each contributing cognitive radio. The design of such a cooperative signal classification system is further studied through numerical simulation.
Date of Conference: 22-27 May 2016
Date Added to IEEE Xplore: 14 July 2016
ISBN Information:
Electronic ISSN: 1938-1883