Optimization of Linear Collaborative Spectrum Sensing with Genetic Algorithms

In cognitive radios (CR) secondary users (SU) transmit alongside primary users (PU). In order to avoid interference SU perform spectrum sensing and adaptive transmission. Reliable detection of existing transmissions in wide geographical regions need to perform collaborative sensing. The state of the art for efficient collaborative sensing is given by linear statistics combination. Complex mathematical reformulations and limitations have been suggested to solve the combination optimization, due to its non convex nature. In this paper we propose an alternative optimization technique based on genetic algorithms. The genetic programming performs direct search of the optimal solution one step before the reformulations needed previously. We demonstrate that, by operating directly on the objective of the optimization, the collaborative sensing solution is found consistently with the problem formulation.