Minimizing the detection error of cognitive radio networks using particle swarm optimization

Weighting the coefficients vector is the principal factor influencing the detection performance of cognitive radio networks that uses soft-detection fusion (SDF) based cooperative spectrum sensing. Maximal ratio combining- (MRC-), equal gain combining- (EGC-) and continuous genetic algorithm- (CGA-) based SDF are well suited for optimizing the detection performance and thus ensure safe access of spectrum by CR users. However the mentioned methods suffer from slow convergence and/or sub-optimality. In this paper, the use of particle swarm optimization (PSO) algorithm under MINI-MAX criterion is proposed to optimize the weighting coefficients vector so that the total probability of decision error is minimized. The performance of the PSO-based proposed method is examined and compared with GA-based technique as well as other conventional SDF schemes through computer simulations. Numerical results confirm the effectiveness of the proposed method.