A Technique to Aggregate Classes of Analog Fault Diagnostic Data Based on Association Rule Mining
Analog circuits are widely used in different fields such as medicine, military, aviation and are critical for the development of reliable electronic systems. Testing and diagnosis are important tasks which detect and localize defects in the circuit under test as well as improve quality of the final product. Output responses of fault-free and faulty behavior of analog circuit can be represented by infinite set of values due to tolerances of internal components. The data mining methods may improve quality of fault diagnosis in the case of big data processing. The technique of aggregation the classes of fault diagnostic responses, based on association rule mining, is proposed. The technique corresponds to the simulation before test concept: a fault dictionary is generated by collecting the coefficients of wavelet transformation for fault-free and faulty conditions as the preprocessing of output signals. Classificator is based on k-nearest neighbors method (k-NN) and association rule mining algorithm. The fault diagnostic technique was trained and tested using data obtained after simulation of fault-free and faulty behavior of the analog filter. In result the accuracy in classifying faulty conditions and fault coverage have consisted of more than 99,09% and more than 99,08% correspondingly. The proposed technique is completely automated and can be extended.
Traditionally, diagnostics of analog circuits are implemented using which here on will be referred as Fault Dictionary (FD), each row of which contains the upper and lower boundaries of the range of possible values for controlled parameters in different test nodes for all considered states of the circuit, i.e. fault-free and faulty states containing different kinds of faults. Fault detection occurs during the output response measurement of the circuit-under-test (CUT) and sequential comparison value is obtained within the boundaries in FD rows. The condition of the CUT is diagnosed when the measured value lays in the boundary range of the corresponding row in FD.
Proposed technique reduces complexity of fault detection due to associative mode of operation as well as decreases the high size of the FD thanks to implementation of the FD as artificial neural network with fixed architecture for different number of considered faults. Algorithms which are used in this technique are parallel and ready to run on the clusters. The proposed technique can be described in the following set of main steps: 1) Fault simulation using Monte-Carlo analysis taking into account the component tolerances. 2) Wavelet-decomposition of CUT’s output responses. 3) Class Aggregation based on k-nearest neighbors algorithm (k-NN) and Association Rule algorithms 4) Building Machine Learning Model
The technique of construction for the classifier for analog fault testing and diagnosis was done by using the extraction of the essential characteristics based on wavelet transformation, Monte-Carlo method, association rules mining algorithms, and machine learning algorithm. The proposed technique helps to produce the high reliable analog and mixed-signals integrated circuits. The experimental verification of the prediction quality was performed on the most widely used filter topologies. The results obtained for the Sallen-Key filter demonstrate the high precision of prediction (> 99, 09%) and fault coverage (> 99, 08%) in the task of fault diagnostics. The proposed technique uses algorithms which were parallel and prepared to handle the big data obtained in result of the exhaustive simulation of analog circuits.
 S. Mosin, “Design-for-testability automation of mixed-signal integrated circuits,” in 2013 IEEE International SOC Conference. IEEE, Sept 2013, pp. 244–249. [Online]. Available: http://ieeexplore.ieee.org/document/6749695/
 ——, “Automated simulation of faults in analog circuits based on parallel paradigm,” in 2017 IEEE East-West Design Test Symposium (EWDTS). IEEE, Sept 2017, pp. 1–6. [Online]. Available: http://ieeexplore.ieee.org/document/8110133/  H. G. Stratigopoulos and S. Sunter, “Efficient Monte Carlo-based analog parametric fault modelling,” in Proceedings of the IEEE VLSI Test Symposium. IEEE, apr 2014, pp. 1–6. [Online]. Available: http://ieeexplore.ieee.org/document/6818741/
 A. Kavithamani, V. Manikandan, and N. Devarajan, “Soft fault classification of analog circuits using network parameters and neural networks,” in Journal of Electronic Testing: Theory and Applications (JETTA), vol. 29, no. 2. Springer US, apr 2013, pp. 237–240. [Online]. Available: http://link.springer.com/10.1007/s10836-013-5370-3
 M. Quispe-Ayala, K. Asalde-Alvarez, and A. Roman-Gonzalez, “Image classification using data compression techniques,” Engineers in Israel, pp. 349–353, nov 2010.
 N. K. Ahmed, A. F. Atiya, N. E. Gayar, and H. El-Shishiny, “An Empirical Comparison of Machine Learning Models for Time Series Forecasting,” Econometric Reviews, vol. 29, no. 5-6, pp. 594–621, aug 2010.
 A. H. Asikainen, J. Ruuskanen, and K. A. Tuppurainen, “Consensus kNN QSAR: A versatile method for predicting the estrogenic activity of organic compounds in silico. A comparative study with five estrogen receptors and a large, diverse set of ligands,” Environmental Science and Technology, vol. 38, no. 24, pp. 6724–6729, nov 2004. [Online]. Available: http://pubs.acs.org/doi/abs/10.1021/es049665h http://ieeexplore.ieee.org/document/5662206/
 R. Agrawal, T. Imielinski, and A. Swami, “Mining association rules between standets of items in large databases,” Proceedings of the 1993 ACM SIGMOD international conference on Management of data, vol. 22, no. May, pp. 207–216, 1993.
 C. Ordonez and E. Omiecinski, “Discovering association rules based on image content,” in Proceedings IEEE Forum on Research and Technology Advances in Digital Libraries. IEEE Comput. Soc, 1999, pp. 38–49. [Online]. Available: http://ieeexplore.ieee.org/document/777689/
 R. Agrawal and R. Srikant, “Fast algorithms for mining association rules,” in 94 Proceedings of the 20th International Conference on Very Large Data Bases, vol. 1215, 1994, pp. 487–499.
 M. J. Zaki and M. J. Zaki, “Scalable Algorithms for Association Mining,” IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, vol. 12, pp. 372—-390, 2000. [Online]. Available: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.28.656
 J. Han, J. Pei, and Y. Yin, “Frequent Pattern Tree : Design and Construction,” Networks, vol. 29, no. 2, pp. 1–12, 2000. [Online]. Available: http://portal.acm.org/citation.cfm?doid=342009.335372