A logic hazard detection and elimination method

Abstract

This paper is concerned with the detection and elimination of static logic hazards due to single and multiple input changes. The method involves only simple manipulation of the Boolean expression corresponding to the logic network. It not only detects the existence of hazards and determines the variable sets with respect to which the hazards exist, but also, for each of these sets, pin-points the subcubes within which each input transition involving exactly the changes of all variables in the set produces the hazardous output. Further, it can determine the terms the absence of which from the expression cause the hazards. Thus, the hazard can be eliminated by including these missing terms in the expression. The theoretical foundation of the method is given first followed by an example illustrating the procedure and its effectiveness. The procedure is formulated into two different algorithms, one for saving time and the other for saving memory. They have been implemented in FORTRAN. A related result on prime implicant generation is also included.