Title :
COGENT: COmpressing and compacting GEnetic algorithms and Neural networks based automatic Test generator
Author :
Shara, Shekhar Agrawal
Author_Institution :
Bradley Dept. of Electr. & Comput. Eng., Virginia Polytech. & State Univ., Blacksburg, VA, USA
Abstract :
Genetic algorithms are one of the most powerful tools used in the field of testing to deliver high-quality automatic test pattern generation (ATPG) for both sequential and combinational circuits essentially because ATPG is a search and optimization problem. Artificial neural networks, using the learning methodology can be used to solve any class of linearly non-separable problems, which includes classification and recognition. In this paper, these two tools have been combined to deliver a high performance ATPG, COGENT. COGENT also compacts and compresses/decompresses data. The compression algorithm used is a variation of the Coulomb´s scheme. The results for all the different cases examined have been tabulated and discussed.
Keywords :
automatic test pattern generation; combinational circuits; data compression; genetic algorithms; learning (artificial intelligence); linear programming; neural nets; search problems; sequential circuits; COGENT; Coulomb scheme variation; artificial neural network; automatic test pattern generator; classification; combinational circuit; compression algorithm; data compression; data decompression; deterministic method; genetic algorithm; high performance ATPG; learning methodology; linearly nonseparable problem; optimization problem; recognition; search problem; sequential circuit; Artificial neural networks; Automatic test pattern generation; Automatic testing; Circuit faults; Circuit testing; Fault detection; Genetic algorithms; Neural networks; Sequential analysis; Test pattern generators;
Conference_Titel :
Soft Computing in Industrial Applications, 2003. SMCia/03. Proceedings of the 2003 IEEE International Workshop on
Print_ISBN :
0-7803-7855-5
DOI :
10.1109/SMCIA.2003.1231352
