Coverage Fulfillment Automation in Hardware Functional Verification Using Genetic Algorithms

Round 1

Reviewer 1 Report

1. Introduction: The introduction is written very poorly. The writing could be improved. There are several places in the manuscript where the passages are unclear, awkwardly phrased. Research issues, and contributions are not clearly understood. The authors could introduce the background of this research to better highlight the importance of this research topic. The authors could discuss the main differences between existing studies and their research, and the novelty of the article. The authors could add the organization of the paper at the end. The authors are suggested to prepare a separate section for literature review.
   • Line 11 on Page 1: Reference style check; incomplete sentence
   • Use the unified style in the same manuscript, for example, Field Programmable Gate Array - FPGA vs. design under test (DUT).
2. Materials and Methods & Results: These sections are not appropriate for review; the authors should improve these two sections with clear descriptions and proper math expressions. All tables and figures should include all information required for understanding, e.g., units. The author(s) did not use the standardized, unified figure style in the manuscript. Moreover, the authors should interpret each figure in more detail with their insights/findings.
   • Too general explanation/description about methods 
   • Need to adjust the scale of each figure. Use the mathematical expressions, not a screenshot or so.
   • Add more explanation about the parameter, set, or variables in the equations. 
   • When using an acronym, you need to introduce it with full terminology in the first instance, e.g., UVM. 
3. Discussion: Not sure what is the relation between their analysis and discussions. The discussions were not sound and justified by the results. Furthermore, in this section, the authors have to discuss the unique contribution(s) of their study.
4. Conclusions: The conclusion is poor; no special conclusions were obtained for the objectives of this study. I think this full section needs to be re-written after the other changes I’ve mentioned. The overall structure of this manuscript is not organized well.

Author Response

Dear reviewer,

The article was modified in many places. Please see the attachment where we answered at your requests which helped us to increase the value of the article. We send you the article version where all the changes we made are highlighted. In order to see the final version of the document, on a local copy, we followed the following steps in Microsoft Word: Review -> Accept -> Accept all changes and stop tracking

Thank you!

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript titled “Coverage fulfillment automation in hardware functional verification using genetic algorithms” treats several approaches to the verification process based on genetic algorithms. The authors claim that in all situations, the percentage of stimulus sets generated using genetic algorithms with a high coverage value was higher than the values that resulted when random simulations were employed.

Employing genetic algorithms for hardware functional verification is provides a certain degree of guidance or supervision to the overall process of coverage fulfillment assessment. This aspect may be desired and useful. The randomized approach might still be needed for some functionalities, hence combining the approaches should be at least considered.

Suggestions:

1. The reference to the literature need to be broader in scope and to go into some comparative details.

2. Present more clearly their concrete implementation of the genetic algorithms proposed, perhaps accompanied by flow diagrams or pseudo-code.

3. Emphasize the difference not only among the proposed versions but make a comparison with the industry standards.

4. Generally, the manuscript has a value that needs to be put in perspective of the common practices used in the industry in order to project beneficial differences.

Author Response

Dear reviewer,

The article was modified in many places. Please see the attachment where we answered at your requests which helped us to increase the value of the article. We send you the article version where all the changes we made are highlighted. In order to see the final version of the document, on a local copy, we followed the following steps in Microsoft Word: Review -> Accept -> Accept all changes and stop tracking

Thank you!

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors not only addressed the suggestions made in the first round of the reviewing process, but improved significantly the manuscript in terms of the clarity of presentation and the delineation regarding their contribution to the field. Congratulations!