Fault simulation under the multiple observation time approach using backward implications

We present an improved procedure for fault simulation under themultiple observation time approach based on state expansion.Under state expansion, an incompletely specified state reachedduring fault simulation is replaced by 2{k} states, each one assigninga different combination to k unspecified present state variables.For each expanded state, additional output values are thenimplied. As a result, a fault that cannot be identified as detectedusing conventional simulation may now be identified as detected.The procedure proposed here enhances state expansion by backwardimplications to take advantage of every present state variablevalue specified under state expansion. As a result of usingbackward implications, fewer states need to be considered afterstate expansion, fewer state expansions are potentially neededfor every fault, and the number of faults that can be efficientlyconsidered is increased. Experimental results are presented tosupport these claims.