论文部分内容阅读
随着电路越来越走向集成化和模块化,大多数电路都同时包含数字和模拟的混合信号,数模混合电路故障诊断技术也因此成为一个重要的课题。提出了利用小波分析和神经网络对混合集成电路故障进行检测的方法。基本思想是通过小波变换对原始采样信号进行检测,再利用神经网络对小波变换的结果进行分类,最后给出故障的信息。通过MATLAB仿真实验,证明该方法对混合电路的故障检测非常有效。
As circuits become more and more integrated and modular, most circuits contain both mixed digital and analog signals, and digital-analog hybrid circuit fault diagnosis techniques have therefore become an important issue. A method of detecting hybrid integrated circuit fault using wavelet analysis and neural network is proposed. The basic idea is to detect the original sampled signal by wavelet transform, and then use neural network to classify the result of wavelet transform, and finally give the fault information. Through MATLAB simulation experiments, it is proved that this method is very effective for the fault detection of hybrid circuits.