Cognitive Radio Engine Model Utilizing Soft Fusion Based Genetic Algorithm for Cooperative Spectrum Optimization

Cognitive radio (CR) is to detect the presence of primary users (PUs) reliably in order to reduce theinterference to licensed communications. Genetic algorithms (GAs) are well suited for CR optimizationproblems to increase efficiency of bandwidth utilization by manipulating its unused portions of theapparent spectrum. In this paper, a binary genetic algorithm (BGA)-based soft fusion (SF) scheme forcooperative spectrum sensing in cognitive radio network is proposed to improve detection performance andbandwidth utilization. The BGA-based optimization method is implemented at the fusion centre of a linearSF scheme to optimize the weighting coefficients vector to maximize global probability of detectionperformance. Simulation results and analyses confirm that the proposed scheme meets real timerequirements of cognitive radio spectrum sensing and it outperforms conventional natural deflectioncoefficient- (NDC-), modified deflection coefficient- (MDC-), maximal ratio combining- (MRC-) and equalgain combining- (EGC-) based SDF schemes as well as the OR-rule based hard decision fusion (HDF). Thepropose BGA scheme also converges fast and achieves the optimum performance, which means that BGAbasedmethod is efficient and quite stable also.