基于模拟退火的并行进化规划多用户检测算法

Multi-user detection algorithm based on simulated annealing parallel evolutionary programming

模拟退火和多种群并行进化规划是2种较好的改进进化算法性能的方法。将这2种思想有机地结合起来,提出了一种基于模拟退火的并行进化规划多用户检测算法。在该算法中,进化在多个不同的子群中并行进行,利用模拟退火算法的爬山性能,避免单种群进化过程中出现的过早收敛现象,提高整个算法的收敛速度。仿真结果表明,这种新的多用户检测算法抗多址干扰和抗远近效应能力都优于单种群的模拟退火进化规划多用户检测算法,并且在多址干扰和远近效应存在的条件下,其收敛速度明显优于基于单种群的模拟退火进化规划检测器。

Simulated annealing and multi-group parallel evolutionary programming are two helpful methods which can improve the performance of evolutionary algorithm. The two ideas are well combined a new multiuser detection, that is the parallel evolutionary programming algorithm based on simulated annealing, is proposed in the paper. In the algorithm, the evolutions of subgroup is performed in parallel with the hill climbing performance of simulated anneal- ing, so this algorithm can avoid the premature convergence of the alone group evolutionary process and improves the convergence speed of the algorithm. The simulation results show that the new algorithm not only is superior to the multi-user algorithm based on the alone group simulated annealing evolutionary programming, but also can faster con- vergence than the detector based on the alone group simulated annealing evolutionary programming in the terms of multiple-access interfere and near-far resistance.