An Automation Program for March Algorithm Fault Detection Analysis

The efficiency of a Memory BIST for embedded memory testing depends on the fault coverage of the implemented test algorithm. A fault simulator is necessary to analyze. The fault detection performance analysis on a March algorithm is indispensable to determine its fault coverage. This manual analysis process is very tedious and time-consuming, and thus, a fault simulator is necessary to perform fault simulation on a memory model. This paper presents the development of an automation program to analyze a March algorithm fault detection performance autonomously, without performing fault simulation. It works by identifying the pair of sensitizer and detector of each fault model, the memory cells values trend, and finally producing a detailed fault detection analysis result and the fault coverage. The program test on March SR and March CL algorithms produced accurate fault detection analysis and the expected fault coverages, which were completed in less than 3 ms.