Development of automated neighborhood pattern sensitive faults syndrome generator for static random access memory

Testing is one of the main key in advanced semiconductor memory technologies. In the past, memory testing only focuses on fault detection. With the increasing complexity of memory devices, fault diagnosis is becoming very important to locate and identify type of fault. One of the memory faults is Neighborhood Pattern Sensitive Faults (NPSF). NPSF is one of the faults that are hard to test due to higher number of cells to be tested at one time. Moreover, most of the memory test algorithm does not have the capability to detect and diagnose NPSF. Therefore, the purpose of this thesis is to develop NPSF detection and diagnose software for Static Random Access Memories (SRAM). The development of this Automated NPSF Syndrome Generator (ANPSFSG) is to improve the process of analyzing NPSF detection and to generate the fault syndrome for NPSF diagnosis. This automated generator will facilitate NPSF analysis as manual fault analysis is no longer practical due to increasing memory size. The algorithms used in this generator are based on March algorithm. Three types of March algorithms which are March 17N, March 12N and MarchPS 23N are selected to validate the tool in term of their compatibility for NPSF detection and diagnosis.Suitable data background is identified and a test procedure is developed for each algorithm. All test procedures are integrated into comprehensive database which is developed using Microsoft Access software. The ANPSFSG is able to list detected diagnosed faults as well as to calculate and display fault diagnostic resolution. A user-friendly Graphical User Interface (GUI) is developed using Microsoft Visual Basic software to load and display the algorithm under test and display the result. The results produced by the tools are then validated with other research finding. This tool can be used to ease the process of developing a new March test algorithm for NPSF.

Preview PDF FK 2011 114R.pdf Download (896kB) | Preview

Actions (login required)

View Item