N-detect测试的优化方案  

Optimization scheme of N-detect test

作　　者：王欣[1] 李锦明[1] WANG Xin;LI Jinming(School of Semiconductors and Physics,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学半导体与物理学院,山西太原030051

出　　处：《微电子学与计算机》2025年第4期131-137,共7页Microelectronics & Computer

摘　　要：未建模故障引发的芯片失效问题日益严重,创新性的N-detect(多重检测)故障模型应运而生,其显著提升了未建模故障的覆盖率。然而,N-detect的测试向量繁多,成本高昂。为克服这一挑战,引入嵌入式多重检测(Embedded-Multi-Detect,EMD)算法进行优化,该算法利用测试向量中的无关位进行额外检测,扩大了测试范围。优化结果表明,这一创新在不增加测试向量的前提下提高了覆盖率,实现了测试向量的高效利用,提高了测试向量效率。The problem of chip failure caused by unmodeled faults is becoming more and more serious,and the innovative N-detect(multiple detection)fault model is proposed,which significantly improves the coverage of unmodeled faults.However,the test vectors of N-detect are numerous and the cost is high.To overcome this challenge,this paper introduces the EMD(Embedded-Multi-Detect)algorithm for optimization,which cleverly uses the irrelevant bits in the test vector for additional detection and expands the test range.The optimization results show that the innovation not only improves the coverage without increasing the test vector,but also realizes the efficient use of the test vector,and achieves the goal of improving the efficiency of the test vector.

关 键 词：多重检测 嵌入式多重检测 可测试性设计 

分 类 号：TN47[电子电信—微电子学与固体电子学]