Opportunistic Wideband Spectrum Sensing for Cognitive Radios with Genetic Optimization

In cognitive radios (CR) secondary users (SU) transmit alongside primary users (PU). In order to avoid interference SU perform spectrum sensing and adaptive transmission. Ultra-wideband sensing is of primary importance to sense and access opportunistically several bands at a time. The state of the art for opportunistic wideband sensing is given by Multiband Joint Detection (MJD). Constraints on interference and utilization have to be imposed to solve the multiband optimization with convex maximization. In this paper we propose an alternative optimization technique based on the genetic algorithms. The genetic programming performs direct search of the optimal solution one step before the aforementioned mathematical limitations. By operating directly on the objective of the optimization we show that the wideband sensing problem can be solved consistently with the problem formulation and without limitations.