Voting Model Based Diagnosis of Bridging Faults in Combinational Circuits

A novel algorithm for diagnosing all two-line single bridging faults in combinational circuits is presented. It uses the accurate voting model to model bridging faults. SOPS, a compact data structure is used to represent the set of possible faults. This makes the algorithm space efficient. A set of rules is used to reduce the fault list. These rules are based on stuck-at fault simulation, making the algorithm time efficient. The diagnosis algorithm also guarantees that the fault which causes the failure will remain in the final fault list. Experimental results are presented. 1 Introduction In a circuit, when two or more distinct lines are unintentionly connected, due to a defect, we have a bridging fault. In this work we assume that the fault is such that only two lines are shorted, and the resistance between the two shorted points, compared to the pull-up and pull-down resistances, is negligible. Also, the circuit contains a single bridging fault only. These faults are known as Two-..