A new method of test generation for sequential circuits

The constant development of integrated circuits makes the automatic test pattern generation problem of sequential circuits a challenging area. Test generation procedures based on genetic algorithm (GA) have been showed to be effective in achieving high fault coverage. This paper presents a new approach to the test generation,for synchronous sequential circuits. In this work, we address the problem of the automation test sequences generation based on particle swarm optimization (PSO) and fault simulation for sequential circuits. We emphasize three aspects: initialization, test sequence generation and test set compaction. Experimental results illustrate the effectiveness of the approach. PSO-based STPG is generally superior to GA-based STPG in terms of achieved fault coverage and required CPU time.