The Study Of Switched Current Circuit Test And Fault Diagnosis Methods

Switching Current (Switched Current SI) technology is a simulation sampling data signal processing technology completely adopt digital CMOS process technology which has been bring forward by the late1980s.The switched current technique, based on current mode, is a continuous time analog signal processing technique using discrete time sampled-data system that aims to replace switched-capacitors (SC). It also has the advantages of low-voltage, low-power, high-speed, small size, broadband, good high-frequency characteristics and large dynamic range. Switched Current technology does not require linear floating capacitance and high-performance operational amplifiers and compatible with standard digital CMOS process completely, which is an attractive feature due to the tendency for the integration of large analog/digital systems in a single chip. In the field of discrete-time analog circuits, the SI technique becomes more and more recognized as an interesting alternative to the classical switched capacitor (SC) technique.The rapid progress in modern electronic and computer technology promotes the advent of system-on-chip and mixed-signal integrated circuits. On the other hand, the structure of electronic equipment become more complex and the scale become huger. Higher and newly demands on circuit test are put in order to improving the reliability of electronic equipment. Intensive study on fault diagnosis theories and methods is urgent task. With many years’development, the analog circuit tests acquire some research progress. However, concerning the testing aspect, the test and fault diagnosis techniques proposed for switched current circuit are still almost empty, which greatly limits the development of switched current technology. Moreover, because of non-ideal characteristics, non-zero output conductance, limited bandwidth and switch charge injection of MOS transistor in the switched current circuit, these factors determine that the switched-current circuit testing and fault diagnosis is a very difficult tast, and the systematic and breakthrough progress has not made. Based on this, in the paper, the testing and fault diagnosis methods for switched current circuit are deeply researched systematically. The hard fault test method of switched current circuit based on fault model is preliminary studied. On this basis, several new switching current circuit test and fault diagnosis methods are proposed, and circuit examples are used to verify. The main contributions of this dissertation are presented as follows:1. The development and advantages of switched-current technology are summaried compared with SC technique. The basic functions and features of switching current ASIZ simulation software are introduced. Then the hard fault and soft fault model of switched current circuit are analysised systematically. The hard fault test method of switched current circuit based on fault model is preliminary studied. The two examples of switched current circuit are tested.2. Switched current circuits test using Pseudo Random method is proposed. Pseudo-random excitation signal is introduced to the SI circuits test, and the application of pseudo-random testing techniques is discussed. By checking the constructed signatures (impulse response samples) against the derived tolerance ranges, we can infer the correctness of the device under test (DUT) without explicitly measuring the original performance parameters. The generation of pseudo-random excitation signal. We also describe a technique of mapping the tolerance ranges in the process board space, the device space, performance space and the signature space. A fifth order Butterworth low-pass filter and a sixth order Elliptic band-pass filter have been used as test benches to assess the effectiveness of the proposed technique. Test results demonstrate that high fault coverage can be achieved with low cost test equipments.3. A novel method based on a fault dictionary that uses entropy as a preprocessor to diagnose faulty behavior in switched-current (SI) circuit is presented. The proposed method uses a data acquisition board to extract the original signal form the output terminals of the circuit-under-test (CUTs). These original data are fed to the preprocessors for feature extraction and finds out the entropies of the signals which are a quantitative measure of the information contained in the signals. The proposed method has the capability to detect and identify faulty transistors in SI circuit by analyzing its output signals with high accuracy. Using entropy of signals to preprocess the circuit response drastically reduces the size of fault dictionary, minimizing fault detect time and simplifying fault dictionary architecture. The method can classify not only parametric faults but also catastrophic faults. It is applicable to analog circuits as well as switched-current ones. A low-pass and a band-pass SI filter and a Clock feedthrough cancellation circuit (CKFT) have been used as test beached to verify the effectiveness of the proposed method. The result reveals that our method requiring one feature parameter reduces the computation and fault diagnosis time. 4. A switched-current (SI) circuit fault diagnosis system based an backpropagation (BP) neural networks is proposed. Our system uses a data acquisition board to extract the original training data for the neural network from the output terminals of the circuit-under-test (CUT). The original data are preprocessed by feature extraction to find out the entropy and kurtosis of signals. As a result, using entropy and kurtosis of signals to preprocess the impulse responses drastically reduces the number of inputs to the neural network classifier, simplifying its architecture, minimizing its training and processing time and improving the performance of the network. Our fault diagnostic system, trained and tested using our proposed preprocessing techniques, achieves98%accuracy in classifying faulty transistors. Our work also shows that ASIZ simulations can be used to extract appropriate features for training the neural network. Moreover, the neural network based preprocessing and fault dictionary technique can locate and identify not only hard faults but also soft fault because neural networks are capable of robust classification even in noisy environment. When there are a large number transistors occurring failure simultaneously, the method can achieve high rates of fault classification.