基于二进制蜉蝣算法的频谱资源分配

Spectrum Allocation Based on Binary Mayfly Algorithm

针对目前基于图论的认知无线电频谱分配算法存在收敛速度较慢、寻优精度不高等问题,提出了二进制蜉蝣算法,通过引入汉明距离对蜉蝣速度更新公式进行了重新定义,通过Sigmoid函数将蜉蝣位置更新进行了二值化处理,通过有向双点交叉和按位变异对蜉蝣群体的交配和变异行为进行了重新诠释,并将其应用于频谱资源分配,与经典的二进制粒子群算法、离散人工蜂群算法和二进制蜻蜓算法进行比较,实验结果表明,二进制蜉蝣算法收敛速度快、寻优能力强,能够有效地提高频谱资源利用率和频谱分配公平性。

For the slow convergence and barely satisfactory optimization performance of existing spectrum allocation algorithms based on graph theory model, a binary mayfly algorithm(BMA) is presented. In BMA,the mayfly velocity updating formulas are redefined using Hamming distance, and the mayfly location binarization is realized by using the Sigmoid function, while the mating and mutate movements of the mayfly are reinterpreted by oriented two-point crossover and bitwise variation. Then BMA is applied to solve the spectrum allocation problem and compared with the classical binary particle swarm optimization algorithm, discrete artificial bee colony algorithm and binary dragonfly algorithm. Experimental results show that BMA has fast convergence and good optimization ability, which can effectively improve the utilization of spectrum resources and the fairness of spectrum allocation.