一种新的次优多用户检测器

针对码分多址接入（CDMA）系统中最优多用户检测器的指数计算复杂度问题，从组合优化的角度，提出了一种基于遗传算法和局部最小算法的次优多用户检测器，这种多用户检测器，通过遗传算法进行初步选择，再用局部最小算法进行局部优化，可避免单纯使用遗传算法时收敛速度慢，易于出现早熟收敛等缺点，且计算复杂度小，计算机仿真结果表明，该检测器的比特误码性能接近最优多用户检测器。

A New Suboptimum Multiuser Detector

With respect to the exponential computation complexity of the optimal multiuser detection in CDMA systems, a new sub optimal multiuser detector which employs a genetic algorithm and a local minimization algorithm was proposed from a combinatorial optimization viewpoint. Using this approach, the genetic algorithm provides a good initial point, and the local minimization algorithm optimizes it locally. The slow convergence rate and premature convergence of the pure genetic algorithm are avoided when the proposed algorithm is used. The computer simulation results show that the performance approaches the optimal multiuser detector with reduced computation complexity.