基于Hopfield神经网络多用户检测技术的研究

Research on Multi-user Detection Technology based on Hopfield Neural Network

多用户检测中的MMSE算法随着用户数增加而受到限制,MMSE的优化问题可以转化为Hopfield神经网络的能量函数的最小化问题。针对基于Hopfield神经网络多用户检测算法进行仿真和分析,仿真表明Hopfield神经网络多用户检测算法在抗多址干扰和远近效应上有所提高。

MMSE algorithm will be restricted with the increasing number of users. Optimization of the MMSE can be transformed into minimization of energy for the Hopfield neural network function. This paper simulates and analyses performance of the multi-user detection algorithm based on the Hopfield neural network. Simulation shows that the algorithm of Hopfield neural network makes an improvement on the restricting multiple access interference and near-far effect.