Title :
Optimal on-line built-in self-test structure for system-reliability improvement
Author :
Papa, Gregor ; Garbolino, Tomasz
Author_Institution :
Comput. Syst. Dept., Jozef Stefan Inst., Ljubljana, Slovenia
Abstract :
This paper presents a bio-inspired technique for the generation of a deterministic test pattern generator. In contrast to conventional methods, the proposed evolutionary-based approach reduces the gate count of a built-in self-test structure, which is used for the automatic fault detection. The reduced-gate count structure is needed to achieve the test structure with a smaller hardware area overhead, while still satisfying the reliability constraints. The presented optimization approach searches concurrently for the optimal combination of the register cells structure, the test patterns order in the generated test sequence, and the bit order of the test patterns. A comparison of the results with similar studies shows the efficiency of the proposed evolutionary approach, which is therefore very useful in the design of robust and fault-tolerant systems, while maintaining the minimum size of the hardware overhead.
Keywords :
automatic test pattern generation; built-in self test; evolutionary computation; fault tolerant computing; logic testing; automatic fault detection; bio-inspired technique; deterministic test pattern generator; evolutionary-based approach; fault-tolerant systems; optimal online built-in self-test structure; reduced-gate count structure; system-reliability improvement; Biological cells; Built-in self-test; Genetic algorithms; Inverters; Logic gates; Optimization; Registers;
Conference_Titel :
Evolutionary Computation (CEC), 2011 IEEE Congress on
Conference_Location :
New Orleans, LA
Print_ISBN :
978-1-4244-7834-7
DOI :
10.1109/CEC.2011.5949622