On statistical behavior of branch coverage in testing behavioral VHDL models

Hajjar, Abd El Salam Al; Chen, T.; Mayrhauser, A. von

doi:10.1109/hldvt.2000.889565

Cited by 10 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model was developed based on a more accurate statistical description of branch coverage [10], where the branch coverage process is decomposed into two random variables: interruptions ¡ (one or more new branches are covered at time¨), and sizes of the interruptions, ¢ , given that there is an interruption at time¨. The distribution function describing £ is found to be a non-homogeneous Poisson process, and the probability of having an interruption during the testing process is estimated as…”

Section: ¦mentioning

confidence: 99%

“…These models assume certain statistical behavior with regard to coverage, which include Binomial, Poisson, and Logarithmic Series distributions. However, it has been shown, theoretically and experimentally [10], that the underlining assumptions about the statistical behavior of coverage are invalid for behavioral models. We have presented a more accurate statistical behavior based on verification results of 14 VHDL models applying one million test patterns to each model.…”

Section: Introductionmentioning

confidence: 98%

“…However, the parameters in the Bayesian models proposed in [18] are determined based on the user specification and are fixed throughout the process of determining stopping points. This paper presents new results, which further improve the efficiency of our previously proposed models [10,18]. The contributions of this paper include: ¢ The parameters in the Bayesian model are dynamically adjusted using the Response Surface Method (RSM) based on the coverage history at a given time.…”

Section: Introductionmentioning

confidence: 98%

“…It is shown that the resulting statistical behavior is different from the assumptions made in the existing statistical models derived from software engineering. The statistical model proposed previously in [10,18] takes into account many issues in VHDL model verification, such as timing and sequential phenomenon and correlation of coverage items or clumping, i.e. dependency of covering one branch on covering other branches.…”

Section: Introductionmentioning

confidence: 99%

“…, presented in [10], we derive the statistical model for the branch coverage increase for a given history of testing simulation. Let …”

mentioning

confidence: 99%

See 4 more Smart Citations

Improving the efficiency and quality of simulation-based behavioral model verification using dynamic Bayesian criteria

Hajjar

Chen

Proceedings International Symposium on Quality Electronic Design

Self Cite

View full text Add to dashboard Cite

In order to improve the effectiveness of simulation-based behavioral verification, it is important to determine when to stop the current test strategy and to switch to an expectantly more rewarding test strategy. The location of a stopping point is dependent on the statistical model one chooses to describe the coverage behavior during verification. In this paper, we present a dynamic Bayesian (DB) and a confidence-based dynamic Bayesian (CDB) stopping rules for behavioral VHDL model verification. The statistical assumptions of the proposed stopping rules are based on experimental evaluation of probability distribution functions and correlation functions. Fourteen behavioral VHDL models were experimented to determine the high efficiency of the proposed stopping rules over the existing ones. Results show that the DB and the CDB stopping rules outperform all the existing stopping rules with an average improvement of at least 69% in coverage per testing patterns used.

show abstract

Section: ¦mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%