CONSIDERING TESTABILITY AT BEHAVIORAL LEVEL - USE OF TRANSFORMATIONS FOR PARTIAL SCAN COST MINIMIZATION UNDER TIMING AND AREA CONSTRAINTS

We address the problem of transforming a behavioral specification so t hat synthesis of a testable implementation from the new specification requires significantly less area and partial scan cost than synthesis from the original specification. The proposed approach has three compo nents: a library of relevant transformation mechanisms, an objective f unction, and an optimization algorithm, The most effective transformat ions for testability optimization are identified by analyzing the fund amental relationship between transformational mechanisms and topologic al and functional properties of the computations that affect testabili ty, A dynamic, two-stage objective function that estimates the area an d testability of the final implementation, and also captures enabling and disabling effects of the transformations, is developed. Optimizati on is done using a new randomized branch and bound steepest descent al gorithm, Application of the transformation algorithm on several benchm ark examples demonstrates significant simultaneous improvement in both area and testability of the final implementations.