基于量子果蝇优化的认知无线网络频谱分配*

将频谱分配的二进制编码转化为量子序列编码,提出一种基于量子果蝇优化的认知无线网络频谱分配方法。首先,将果蝇优化算法(FOA)转化为量子果蝇优化算法(QFOA)算法,拓展FOA算法的应用范围；然后,采用选择、交叉、变异操作改进QFOA算法,提高算法收敛速度,增加样本种群多样性,避免算法陷入局部最优；接下来,利用改进QFOA算法对频谱分配的量子序列进行寻优,寻求最优的网络效益或者用户公平性,得到网络整体性能最优的频谱分配策略。仿真结果表明,改进的QFOA算法收敛速度快且跳出局部最优能力强,应用到认知无线网络频谱分配中,增加了网络资源利用率,提高了网络的整体性能。

Spectrum Allocation Based On Quantum Fruit Fly Optimization Algorithm In Cognitive Radio Network

The binary codes of spectrum allocation are transformed into quantum sequences, so a spectrum allocation based on quantum fruit fly optimization is proposed. First, the fruit fly optimization algorithm is transformed into the quantum fruit fly optimization algorithm, and the range of application of the FOA is expanded. And then, the QFOA is improved by operations which are selection, crossover and mutation. The convergence rate is improved, the sample population diversity is increased, the algorithm falling into local optimum is avoided, and so on. Next, the quantum sequences of spectrum allocation are searched by the improved QFOA, to seek the network utility or user fairness and obtain the spectrum allocation strategy which gains the best network performance. The simulation shows that the improved QFOA has fast convergence rate and strong ability of avoiding local optimum. And the utilization of network resources is increased, the network performance is improved, after using the algorithm for spectrum allocation in cognitive radio network.