Adaptive Genetic Algorithm Verification Technology Driven by Functional Coverage

Abstract:

With the increasing of Chip integration and operation velocity, the gap between validation and design is becoming larger and larger. Now the focus is on generating efficient test vector. Different weights are appropriately set based on the relationships between the code logical depth of function test points and the test points covered by a test vector. A test vector generation method based on adaptive genetic algorithm is proposed. The experimental results show that the method can reduce the time of writing constraint files, auto-generate a stimulation, which is more efficient and targeted, and improve the efficiency and reliability of the chip functional verification.

• Periodical: Applied Mechanics and Materials (Volumes 644-650)
• Pages: 2264-2268
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.644-650.2264
• Online since: September 2014
• Authors: Jing Kai Shi*, Jian Lang Wu, Yi Bin Wang
• Keywords: Automatic Test, Code Logical Depth, Functional Coverage, Genetic Algorithm (GA), Test-Point Weight

References

[1] Bergeron J: Writing Test Benches Using System Verilog [M] . New York, Springer, America, (2006).

[2] International Technology Roadmap for Semiconductors 2011 reports design on http: /www. itrs. net/Links/2011ITRS/ 2011Chapters/2011Design. pdf.

[3] Holland J: Adaptation In Natural And Artificial Systems [M] . University of Michigan Press, America , (1975).

[4] . AiMing Liu: The Fundamentals Of Digital IC Testing, Janpan , (2003).

[5] Joseph I. Stataman: Deep Space Network Revitalization Operationsf or the​21st Century, America, (2007).

[6] Verma S, Harris IG, Ramineni K. Automatic: generation of functional coverage models from behavioral Verilog descriptions. Franch , (2007).

DOI: 10.1109/date.2007.364407