Optimization of Cooperative Spectrum Sensing in Multihop Heterogeneous Cognitive Radio Networks

This article investigates the performance of cooperative spectrum sensing (CSS) in a heterogeneous multihop cognitive radio (CR) system in which N secondary users (SUs) with nonidentical receiver operating characteristic (ROC) curves relay their binary decisions via multihop transmission to the combining node (CN). The CN employs the M-out-of-N rule to determine the availability of the primary user's (PU's) spectrum. First, we derive the generalized ROC expressions for the M-out-of-N rule, which plays a pivotal role in understanding the system's performance. Then, we formulate three general optimization problems under the Bayesian criterion: 1) we optimize the M value of the M-out-of-N rule; 2) we optimize the SUs' ROC points together with the M value; and 3) we jointly optimize the SUs' ROC points with the N value of the M-out-of-N rule subject to a Bayes risk constraint. Solutions are provided for the formulated problems. The expressions derived in this article are general and novel, as they can be applied to any choice of the detector employed in the system. This novel approach generalizes numerous existing results on single-hop and multihop CR systems. Finally, we present numerical results by choosing the conventional energy detector (CED) as an example. These results demonstrate the superiority of our proposed scheme over existing ones while ensuring the required quality-of-service at the CN.