Test des Circuits Analogiques à Base de la Logique Floue par Voie de Simulation

Abstract

With the fast progress of microelectronics and semiconductor technology, integrated analog electronic systems are growing in more sophisticated and complex functions and become very needy of a high reliability . Therefore, the automatic methods requirement for their tests and diagnostic is a crucial issue for researchers in this field. The development of effective approaches and techniques for detection and classification of defects in analogue circuits is the cornerstone of any test design research project aimed at improving the performance of integrated circuits. This thesis introduces the fuzzy logic techniques use for hard faults detection and location in analog electronic circuits built with bipolar transistors. For this purpose, primary, simulations of the circuit under test (CUT) are performed before the test stage by investigating the response of the circuit under test in faulty and fault-free conditions. Afterward , two signatures parameters— the output voltage and the supply current—are observed in DC domain simulation and used to build a fault dictionary. This is used effectively to produce a fuzzy inference system (FIS) which is able to reorganizing structured data digitally and uses them to improve fault location and identification more accurately Finally, the results of different experiments are presented to demonstrate the applicability of the proposed method by increasing its efficiency.