Design of Optimal Frameworks for Wideband/Multichannel Spectrum Sensing in Cognitive Radio Networks

Abstract

Several optimal detection frameworks for wideband/multichannel spectrum sensing in cognitive radio networks are proposed. All frameworks search for multiple secondary transmission opportunities over a number of narrowband channels, enhancing the secondary network performance while respecting the primary network integrity and keeping the interference limited. Considering a periodic sensing scheme with either uniform or non-uniform channel sensing durations, the detection problems are formulated as joint optimization of the sensing duration(s) and individual detector parameters to maximize the aggregate achievable secondary throughput capacity given some bounds/limits on the overall interference imposed on the primary network. It is demonstrated that all the formulated optimization problems can be solved using “convex” optimization if certain practical constraints are applied. Simulation results attest that the proposed frameworks achieve superior performance compared to contemporary frameworks. To realize efficient implementation, an iterative low-complexity algorithm which solves one of the optimization problems with much lower complexity compared to other numerical methods is presented. It is established that the iteration-complexity and the complexity-per-iteration of the proposed algorithm increases linearly with the number of optimization variables (i.e. the number of narrowband channels).