改进ABC算法求解认知无线网络频谱分配问题

Spectrum Allocation Mechanism Based on Improved ABC Algorithm for Cognitive Wireless Networks

针对认知无线网络中频谱分配问题,提出了一种基于改进人工蜂群算法的多目标组合优化算法。首先将频谱分配问题转换成多目标优化问题,然后利用人工蜂群算法的寻优能力来实现频谱最优的分配方案。其中,在雇佣蜂搜索阶段采用新型杂交算子加快收敛速度;跟随蜂搜索阶段引入一种新的概率选择方式保证种群多样性;侦察蜂搜索阶段利用混沌算子来提高全局搜索能力。最后,通过频谱分配仿真对这里提出的算法进行了验证,结果表明:与其他算法相比,这里算法能够较好地跳出局部最优的束缚,具有优化效果佳、稳定性好、鲁棒性强的优点,可以在满足多个优化目标的前提下获得更合理的频谱分配方案。

Aiming at the problem of spectrum allocation in cognitive wireless network, a novel multi-objective optimization algorithm, i.e., improved artificial bee colony algorithm is proposed. Firstly, the problem of spectrum allocation has been transformed into multi-objective optimization problem. Then, the artificial bee colony algorithm has been used to search the optimal scheme with the strong search performance. In the stage of employed bee search, a new cross operator is adopted to increase the speed of convergence. In the stage of following bee search, a new probability of selection strategy is introduced to keep the diversity of the population. And in the stage of scout bee search, the chaotic search operator is used to improve the ability of global search. Lastly, the proposed algorithm had been verified through spectrum allocation simulation. All the results showed that our proposed algorithm had the ability to jump out of local optimum and fast find the global optimum and it has good optimization results, nice stability and strong robustness. And we can obtain a more reasonable spectrum allocation scheme in the case of satisfying multiple optimization objectives.