Using X-Canceling to improve diagnosis for linear space compactors

This paper addresses the problem of performing diagnosis using production test results in a test compression environment where a linear combinational space compactor is used to compact the output response from multiple scan chains. A major issue for compacting output response is handling unknown (X) values. In a linear space compactor, whenever an X-value is XORed together with non-X values, it blocks observation of the non-X values. One elegant and low-cost approach for providing X-tolerance is to use an X-compact [Mitra 04] network which is a linear space compactor that is constructed in a way that guarantees detection of errors from any one or two scan chains in the presence of an X in any other scan chain in the same scan-out cycle. A major challenge is performing diagnosis from the output of a linear space compactor where many outputs may be corrupted by X’s. The key idea in this work is to use symbolic canceling to extract more precise diagnostic information from response compacted by a linear space compactor to identify which scan cells captured errors. The proposed approach does not require any additional hardware or extra data to be collected. It uses off-line software-based processing (symbolic simulation combined with Gaussian elimination) to extract information from the compacted test response to deduce error locations even when there are a large number of errors. Experimental results demonstrate the improved diagnostic accuracy that can be obtained with the proposed techniques.