多特征故障字典及其在模拟电路可测性分析中应用

Multi-feature fault dictionary and its application in testability analysis for analog circuits

提出一种新的模拟电路可测性分析模型——多特征故障字典模型。该模型首先获得输出端的阶跃响应输出波形,提取出波形的多个特征量,将所有特征量形成一特征向量,不同故障类别对应的特征向量组成一特征矩阵。然后由PSPICE软件中的Monte-Carlo分析的结果计算出阈值,在该阈值下将特征矩阵转换成整数编码表,从而计算可测性分析指标-检测率与隔离率。仿真与实测数据表明:与传统模拟电路可测性分析方法相比,提出的方法优点有:1)不需要传递函数,可以适用于复杂电路;2)仅需要一个输出测点,容易实现;3)对于容差模拟电路,同样可以取得较好的检测率与隔离率。

A new testability analysis model for analog circuits i.e.the multi-feature model method is proposed in this paper.Multi-features can be extracted according to the step response waveform at the output point of the cir-cuit.All features form a feature vector, and the all feature vectors of different faults can form a feature matrix .A threshold vector could be obtained by Monte-Carlo analysis in PSPICE .The feature matrix is changed into an inte-ger-encode dictionary by the threshold vector.Then, the dictionary is used to calculate testability analysis index-fault detection rate （FDR） and fault isolation rate （FIR）.The simulation and experiment data show that the ad-vantages of the proposed method in comparison to traditional testability analysis methods for analog circuits are：1 ） It can be used in complex analog circuits because it does not need the transfer function;2） It is easy to implement because it only needs one output test point;3） It can get good FDR and FIR even for some analog circuits with tol-erances.