| Abstract [eng] |
The technological development is enabling production of increasingly complex electronic systems. All those systems must be verified and tested to guarantee correct behavior. As the complexity grows, testing is becoming one of the most significant factors that contribute to the final product cost. The established low-level methods for hardware testing are not any more sufficient and more work has to be done at abstraction levels higher than the classical gate and register-transfer levels. Realized algorithm, which random generates inputs, computes reaction based on software prototype model and deforming values of inputs one by one determines possible terms of logical functions. Analyzing other inputs determined terms of logical functions are corrected by eliminating partial terms. After random generating and analyzing a lot of inputs terminal terms of logical functions are derived. Algorithm doesn’t guarantee that all and exact terms of logical functions are obtained but those terms could be used when generating test vectors. Derived terms of logical functions’ outputs are recorded with input that formed them and following terms can be used as inspecting tests. Collected results can be used for further researches: schemes testing, defect detection, comparing elements of logical function, improving algorithm. Main aspects of design are introduced. Experimental accurateness of results and factors (initial number of random generated test vectors, improvement coefficient, maximum terms number of one output, etc.) that influences algorithm computation progress and quality of results are examined. For improving algorithm a novel hierarchical test generation approach is used: task of functional model is divided into several smaller tasks and their results are combined. Experimental progresses and results of performed analysis’ (hierarchical and common test generation approaches) are compared. In conclusion opinion about important questions is pithily stated: accepted design solutions, point of experiments and results. |
