论文部分内容阅读
本文主要研究了用Ｈｏｐｆｉｅｌｄ神经网络实时实现最大似然多信号源空间参数估计方法。在分析了传统最大似然信号参数估计的理论和Ｈｏｐｆｉｅｌｄ连续状态神经网络特性后，给出了一种实现最大似然空间参数估计模型。该神经网络模型的模拟实验结果表明，用Ｈｏｐｆｉｅｌｄ网络实时实现的信号参数估计方法具有与传统的最大似然参数估计法拥有同样的理论分析结果。
In this paper, the method of real-time estimation of spatial parameters of maximum likelihood multi-source using Hopfield neural network is studied. After analyzing the theory of traditional maximum likelihood signal parameter estimation and Hopfield continuous state neural network, a model of maximum likelihood space parameter estimation is given. The simulation results of the neural network model show that the real-time signal parameter estimation method based on Hopfield network has the same theoretical analysis results as the traditional maximum likelihood parameter estimation method.