Multi-stage energy detection Spectrum Sensing using group intelligence

Spectrum Sensing (SS) plays a very critical role in Cognitive Radio (CR) systems. Among the techniques used for SS is energy detection, which is very popular because of its simplicity and the fact that it does not require information about the primary signal. However, due to its poor performance over fading channels, cooperation between multiple cognitive devices has been proposed to overcome this disadvantage. In this paper, a simple multi-stage cooperative sensing algorithm is presented. The proposed algorithm combines the local decision at each CR along with a group decision received from a fusion center through a simple rule to produce a collective decision. Performance under non-ideal channel conditions, from CR devices to the fusion center and vice-versa, is studied analytically and via simulations. Results show that the proposed algorithm outperforms the conventional hard decisions-based technique and exhibits a comparable performance to the soft decisions-based one at a much lower complexity. In particular, we show that the proposed algorithm improves the detection probability while keeping the false alarm probability almost the same. In addition, we show that it enjoys more robustness against feedback channel errors than the conventional hard decisions-based algorithm.