A pattern skipping method for weighted random pattern testing
B. Konemann · 2002
Conventional stored pattern testing of very large scan design networks is increasingly burdened by the immense volume of pattern data that must be explicitly stored and manipulated for testing. Weighted random pattern testing uses an effective test data coding scheme to reduce the required explicit storage at the expense of increased test time. A novel method is introduced for efficiently skipping over ineffective WRP tests without adding any substantial storage overhead. Examples indicate that the method in a very practical way combines the short test times of stored pattern methods with the storage efficiency of WRP.>