一种实现最佳用户检测的非线性优化神经网络

本文提出并讨论了实现码分多址（CDMA）系统上最佳多用户检测（MUD）的一种神经网络方法。该方法通过将最佳多用户检测视为非线性优化组合问题，利用神经网络能有效求解非线性优化问题的优势，导出了一种非线性优化神经网络来实现最佳多用户检测，理论分析和计算机模拟表明，所提出的神经网络具有可实时应用的动态性能和较传统方法优越得多的误码率性能和抗多址干扰的性能。

Nonlinear Optimum Combination Neural Network for Implementation of Optimum Multiuser Detectors

A neural network approach for implementation of optimum multiuser detector(MUD) in Code-Division Multiple Access Channels is proposed and discussed. By viewing the MUD problem as the nonlinear programming problem, which could be solved by neural network efficiently, a nonlinear optimum combination neural network (NOCNN) for the implementation of optimum MUD is derived. By theoretical analyses and comuter simulation, it is shown that the proposed NN has the dynamic ability of real-time application and the abilty of anti-interference of multiple access, and has the less error bit rate than the traditional methods.