Minimizing memory requirements for deterministic test data in embedded testing
Independent thesis Basic level (university diploma), 10 credits / 15 HE creditsStudent thesis
Embedded and automated tests reduce maintenance costs for embedded systems installed in remote locations. Testing multiple components of an embedded system, connected on a scan chain, using deterministic test patterns stored in a system provide high fault coverage but require large system memory. This thesis presents an approach to reduce test data memory requirements by the use of a test controller program, utilizing the observation of that there are multiple components of the same type in a system. The program use deterministic test patterns specific to every component type, which is stored in system memory, to create fully defined test patterns when needed. By storing deterministic test patterns specific to every component type, the program can use the test patterns for multiple tests and several times within the same test. The program also has the ability to test parts of a system without affecting the normal functional operation of the rest of the components in the system and without an increase of test data memory requirements. Two experiments were conducted to determine how much test data memory requirements are reduced using the approach presented in this thesis. The results for the experiments show up to 26.4% reduction of test data memory requirements for ITC´02 SOC test benchmarks and in average 60% reduction of test data memory requirements for designs generated to gain statistical data.
Place, publisher, year, edition, pages
2010. , 43 p.
deterministic test patterns, deterministic test data, embedded testing, minimizing memory, boundary scan, multiple components, components of same type, duplication of test data, jtag, ieee 1149, concatenator, concatenation of test patterns
IdentifiersURN: urn:nbn:se:liu:diva-54655ISRN: LIU-IDA/LITH-EX-G–10/006–SEOAI: oai:DiVA.org:liu-54655DiVA: diva2:306536
2010-03-15, Alan Turing, Linköpings Universitet, Linköping, 10:00 (English)
Majeed, Mudassar, M.Sc.Ingelsson, Urban, Ph.D.
Larsson, Erik, Professor