Cognitive radio decision engine using hybrid binary particle swarm optimization

The adaptation of the radio parameters is one of the basic capabilities of cognitive radio decision engine. Many researches focus on using genetic algorithm (GA) to select the optimal parameters. However, the convergence speed of GA is low. Recently, particle swarm optimization (PSO) has been used a lot in the cognitive radio system. In this paper, we proposed a hill-climbing binary particle swarm optimization (BPSO) which optimize optimal individual after one iterative operation by hill-climbing algorithm. The proposed method would enhance the local search capability at the later stage of each generation of BPSO. We designed a multi-carrier system for performance analysis. Through different weighting scenarios multiple objective fitness functions, the simulation results illustrate the trade-off between the fitness function and the transmission parameters configuration. And the results show that the hill-climbing BPSO is better than pure BPSO in stability and average fitness value.