Spatial diversity aware data fusion for cooperative spectrum sensing

Studies have shown that when data fusion schemes are used in cooperative spectrum sensing, there is a significant gap between the available resources and the ones perceived by the network. In this paper a cluster based adaptive counting rule is proposed, where the local detectors that experience similar signal conditions are grouped by the fusion center in clusters and where the data fusion is then done separately at each cluster. The proposed algorithm uses the correlation between the binary decisions of the local detectors over an observation window to select the cluster where each local detector should go. It was observed that in the case where there is only one signal source, that the proposed algorithm is able to achieve the same level of performance when compared to the perfect clustering algorithm where full information about the signal conditions at each local detector is available.