本文介绍了应用两种神经网络的方法对肌电信号(Myoelectric Signal)进行分析。作者探索一种更可靠的方法来控制人工假肢多度自由活动,首先采电了一种离散的霍普菲尔德(Hopfield)网来计算一种肌电信号(MES)模型的平均移动时序参数值,文中还证明了采用这种Hopfield网络得出的时序参数值与用最小二乘法Sequential Least Sduars所算出的值是一致的。此外,该网(?)被(?)展应用到那些具备大容量数据和(?)高阶时序模型中, This article describes the application of two neural networks to analyze myoelectric signals. The authors explored a more reliable method of controlling the degree of freedom of artificial prostheses by first harvesting a discrete Hopfield network to calculate mean moving sequence parameter values ​​for a model of electromyographic signal (MES) The paper also proves that the timing parameters obtained by using this Hopfield network are consistent with the values ​​calculated by the least squares method Sequential Least Sduars. In addition, the network (?) Is applied to those with high-volume data and high-order sequential models,