An efficient method for collaborative compressive spectrum sensing in cognitive radio networks

In Cognitive Radio (CR), spectrum sensing is an important process for implementing the whole CR system. As each single CR node has a limited search range, collaborative compressive sensing can significantly improve the ability of detecting the spectrum usage for whole CR networks. In this model, each CR node takes the linearly measurements from the powers of all channels. Then the measurements are sent to a fusion center (FC), where the occupied channels can be detected through recovery algorithms by exploiting the joint sparsity property. Although a variety of recovery algorithms for collaborative spectrum sensing and joint sparse optimization methods have been proposed, designing some more efficient algorithms is still staying in demand. Thus, we proposed a new recovery method for collaborative spectrum sensing in this paper, which is named as Joint-SAMP (sparsity adaptive matching pursuit) algorithm. Compared with previous collaborative spectrum sensing methods and joint sparse optimization algorithms, our method presents superior performance. Besides, Joint-SAMP also has the feature of sparsity adaptive, which makes it more suitable for the application in spectrum sensing. Furthermore, we also give the theoretical analysis of the recovery behavior of the Joint-SAMP algorithm. Simulation results also confirm the effectiveness of our method.