Test-data volume optimization for diagnosis
Proceedings of the 49th Annual Design Automation Conference
Hi-index | 0.00 |
This paper proposes an approach for improving the diagnostic capability of a test-set used in the initial phases of the diagnosis process, when the system is quickly tested with a set of vectors aimed at making the fault observable with the smallest number of vectors. The selection policy identifies the optimal test set with respect to both minimal cardinality and maximum coverage, exploiting an ILP problem formulation. Approach and rationale are supported by analytical methods and criteria, and validated by a set of experimental results.