A Test Point Selection Method Based on Fault Response Matrices Framework

Abstract

This paper presents a novel method for test point selection based on fault response matrices framework. The main purpose of this method is to minimize the set of test points by constructing a fault response matrices framework, to resolve fault effects vanishing and fault detection degradation, and to achieve fewer test point set and higher fault coverage. According to the analysis of the fault response matrices and their derived matrices, the efficiency of the response of each node on the fault is ranked and a new propagation cone for undetectable faults is defined to narrow candidate test point set. We propose a custom testability measure to evaluate the impact of candidate test points on design, and obtain an optimal configuration of the most efficient test points accurately and rapidly. This methodology fully considers the correlation among test points, faults, and test vectors across internal signals resolving timing constraints to increase faults coverage. It can narrow the hard-to-test areas of random test patterns, and reduce the test vector set volume of deterministic test patterns. Our experimental results demonstrate the superiority of the proposed method: higher test coverage, minimum test time, fewer deterministic test patterns, and smaller hardware overhead.