Assessing combinatorial interaction strategy for reverse engineering of combinational circuits

Abstract

T-way test data generators play an immensely important role for both hardware and software configuration testing. Earlier work concludes that t-way test data generator can achieve 100% coverage without having to regard for more than 6 way interactions. In this paper, we investigate whether or not such a conclusion can be applicable for reverse engineering of combinational circuits. In this case, we reverse engineer a faulty commercial eight segment display controller using our t-way test data generator in order to redesign the replacement unit. We believe that our application of t-way generators for circuit identification is novel. The results demonstrate the need of more than 6 parameter interactions as well as suggest the effectiveness of cumulative test data for reverse engineering applications.