A study of faulty signature for diagnostics
John C. Chan, Baxter F. Womack · 2002
A study of faulty signatures in computer hardware testing is presented. A faulty signature is caused by one of many possible input sequences that contains errors at different locations. The faulty signature is analyzed, and it is found that the number of errors among the sequences are distributed with binomial characteristics. Based on this information, a scheme for identifying the failing tests is proposed. A fault-isolation process is discussed.>