摘要： 为使人工鱼群算法在最短的时间内取得多用户检测问题的最优解，在MC-CDMA系统基础上设计一种神经网络人工鱼群算法。人工鱼的3种神经网络行为使神经网络人工鱼群算法在解决多用户检测该类组合优化问题时，减少搜索的随机性和任意性，加快原鱼群算法的收敛速度。仿真结果证明，该算法能够快速收敛，且其抗多址干扰能力和抗远近效应能力优于已有应用智能算法的多用户检测器。
关键词:
多载波码分多址,
多用户检测,
人工鱼群算法,
Hopfield神经网络
Abstract: In order to obtain the optimal values of multiuser detection problem in the shortest time, a neural network fish swarm algorithm is proposed in MC-CDMA systems. Three kinds of neural network behaviors of artificial fish may solve the multiuser detection problem with low randomness and arbitrariness of search, and speed up the convergence speed of original fish swarm algorithm. Simulation results show that the algorithm can achieve the global optimal value in fast convergence rate, and it is superior to the multiuser detectors based on some intelligence algorithms in terms of the multiple access interference and the mitigation of near-far effect.
Key words:
MC-CDMA,
multiuser detection,
artificial fish swarm algorithm,
Hopfield neural network
中图分类号:
高洪元, 刁鸣. MC-CDMA系统的神经鱼群算法检测器[J]. 计算机工程, 2010, 36(24): 180-182.
GAO Hong-Yuan, DIAO Ming. Detector of Neural Network Fish Swarm Algorithm in MC-CDMA Systems[J]. Computer Engineering, 2010, 36(24): 180-182.