基于改进混合遗传算法的信道分配研究

用遗传与模拟退火相结合的混合算法对信道分配问题进行研究,并通过加入"寻优式爬山"与大规模基因突变两种优化方法对混合算法进行改进,克服了一般遗传算法收敛速度慢以及易于陷入局部最优解的缺点。给出了算法的实现流程,并针对几个典型信道分配问题对一般遗传算法、遗传与退火混合算法、改进后的混合算法进行仿真。仿真结果证明改进算法较其他2种算法至少节省80%的时间,并具有更好的稳定性,是解决信道分配问题的一种很好的算法。

Study on Channel Assignment Problem with Hybrid Genetic Algorithm

In this paper,an improved hybrid genetic algorithm based on genetic and simulated annealing is presented.Two strategies,optimizing hill-climbing and large-scale gene mutation,are used to overcome the disadvantages of the primary hybrid algorithm,which easily converg to local optima.We give the details of the algorithm and simulate several benchmark channel-assignment problems using the three algorithms.The result shows that the new approach saves at least 80% of the time and is more stable.It is a good method for solving the channel-assignment problem.