Testing of analog integrated circuits based on power-supply current monitoring and discrimination analysis

Abstract

A new method for the testing and fault detection of analog integrated circuits is presented. The power-supply current is monitored to detect possible faults in an analog circuit. The spectrum of the power-supply current is used to construct the statistical signature of the fault-free and faulty circuits. The decision of a circuit being fault-free or faulty is taken based on the Bayes decision rule fully taking into account the tolerances on the circuit parameters. Examples are given to show the efficiency and effectiveness of the algorithm.<>