Genetic Algorithm with Multi-Parent Crossover for cooperative spectrum sensing

Cognitive Radio (CR) technology aims to reuse the underutilized frequency spectrum by creating new radio access opportunities. Cooperative spectrum sensing (CSS) is proposed to allow multiple secondary users (SUs) to scan the spectrum and verify the existence of primary user (PU). The sensing information of PU is sent by the collaborated SUs to a common fusion center to combine all the data and make a final decision on PU availability. Data aggregation is made by two different schemes, namely soft detection fusion (SDF) and OR-logic hard detection fusion (OR-HDF). In this paper, a new Genetic Algorithm with Multi-Parent Crossover-based SDF scheme (GA-MPC) is proposed and its performance is compared with that of Standard Genetic Algorithm based-SDF scheme (SGA). Computer results show that the GA-MPC outperforms SGA in terms of convergence performance and stability. Furthermore, GA-MPC outperforms SGA, other SDF schemes and the OR-HDF scheme as it can achieve greater detection probability given a probability of false alarm.