数字电路故障模块诊断的神经网络方法研究

对数字电路系统的测试，提出一种基于神经网络的电路故障模块诊断方法，对方法的诊断系统模型、故障特征的获取、网络结构的确定与调整、诊断的流程等进行了详细论述。由于传统的电路故障诊断方法是通过对故障字典的搜索来确定发生的故障，这对庞大的故障字典来说既复杂又费时，因此文中的方法可显著提高传统故障字典法的效率。

A Neural Network Approach to Block Level Circuits Fault Diagnosis

A technique for fault diagnosis of digital circuits and systems at block level is proposed. The Fourier transform methods are employed to obtain the features of logic blocks, the structure of neural networks which accord with the circuit under test is presented. The neural network gradually acquires and refines the knowledge of test samples via genetic evolution of the topology and weight distribution of network. The fault features can then be used as inputs to the trained networks, and the trained network is used to yield diagnosis results. The proposed technique enhances the efficiency of the conventional fault dictionary methods.