247_pdfsam_VLSI TEST PRINCIPLES & ARCHITECTURES

247_pdfsam_VLSI TEST PRINCIPLES & ARCHITECTURES -...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
216 VLSI Test Principles and Architectures Genetic algorithms based on [Holland 1975] were used in CRIS [Saab 1992] in which two individuals were evolved in each generation, which replaced the least fit individuals with some probability. The fitness measure was based on the fault-free activities in the internal nodes in the circuit. This allowed the fitness evaluation to be simple, as only logic simulation is required, thus significantly reducing the computation costs. The circuit is divided into various partitions based on each primary output, and the fitness function favors those individuals that produce similar levels of activity in each partition. It has been presumed that vectors that induce high levels of activity are expected to result in higher fault coverage. As the fitness metric is an estimate of fault coverage, the resulting test sets are longer and may have lower fault coverage compared to deterministic test generators. GATTO [Prinetto 1994] targeted sequential circuits and was based on GAs in
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/16/2011 for the course ENGINEERIN mp108 taught by Professor Elbarki during the Spring '08 term at Alexandria University.

Ask a homework question - tutors are online