Simple True Random Number Generator Using Capacitive Oscillators for FPGA Implementation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents a simple true random number generator (TRNG) based on capacitive oscillators implemented on an FPGA platform. The results indicate that the proposed TRNG can achieve a throughput of 12 Mbit/s with minimal hardware overhead. The design approach is straightforward and well described. However, the methodology lacks novelty, particularly since a similar approach has already been published by the same author [28].

Besides, I hope the following comments will be helpful to the author.

• The implementation details of the basic capacitive oscillators are not clearly explained; it is recommended to include a dedicated section elaborating on their design.
• Why is the proposed TRNG designed as depicted in Fig.1, such as connecting Pin L12:I1 to F12:Q, many design details should be presented more clearly.
• There is no Fig.5 in this paper, which has been cited in the text.
• Why are the experiments conducted on two different FPGA boards?
• How to control the FPGA chip temperature in the experiment? It must be illustrated clearly.
• Compared to [18], all the parameters in this paper are worse than those of the cited paper.

Author Response

The response is given in the PDF file.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Very interesting solution with a well-documented state-of-the-art. The research presents the new implementation of the random number generator in Series-7 Xilinx devices. The author presents a novel approach reducing the resource utilisation compared to the previous approaches. The author presents an extensive elaboration on the state of the art in the paper's domain. The topic is not new, but the paper presents a significant improvement to existing techniques.

Because external pins are used, in secure applications, the signal generation can be affected by applying external signals to these pins. I would like to see some Author's comment about it in the paper.

Author Response

The response is given in the PDF file.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The provided manuscript is the really interesting unique work, that includes the proposed implementation of TRNG based entropy from external FPGA pin and the practical experiments witn passing of NIST tests. This approach meets the requirements of novelty, but the manuscript if too short and requires extending.
Please consider the following comments.

1. The abstract section is too short. It should be 3 times longer. Abstract begins with the proposed results, despite the fact that the first half of this section should briefly describe the essence of the problem and the motivation that makes it difficult to find a solution.
The same applies to the introduction section. Almost the entire introduction consists of isolated sentences on the superficial essence of the thematic publications, without any substantive discussion (in the text) of the complexity of the task of constructing TRNG. After improvement of this section, please add the last paragraph with the structure of the entire work by sections.

2. Materials of previous individual single author MDPI article of the author [28] compose near a third part of materials in current manuscript (Field-Programmable Gate Array-Based True Random Number Generator Using Capacitive Oscillators, Electronics 2024, 13(23), 4819).
The presented list of references is good enough, except match of over the third part of materials in this manuscript with this already published Article of the author. But it is required for current study and is not a plagiarism. But refrasing should be done.

3. For section 2 please remove 'The' from title and replace it by 'Analysis of' of 'Proposed' depending on your role in the obtaining the represebted TRNG circuit (it is better to rephrase the sentence about standard schematic library components).
For the entire work it is reasonable for the proposed TRNG implementation to clarify following.
- Dependency on temperature (because ring oscillators are critically sensitive to temperature drift).
- Parameners of compiler (because anu PUF or TRNG require modification of parameters, because it is not typical function implementation).
- warnings in report window (because forsing extra connections or metastability will be indicated during compilation as warnings).
- Problem of implementation of metastability in FPGA.

4. The provided figures and tables are informative, but the manuscript can be enriched by adding more grafical materials, that definitely can be provided based on the described design.
- Please extend text comments for all figures and tables to multiline form with clear explanation of material and the dot after.
- Please increase the size of figure 2.
- All tables in the manuscript should meet MDPI template (vertical lines are not allowed).
- Second Table 2 should be renamed to Table 3 (in the main text it is mentioned correctly as Table 3).
- Please add few more figures with ring oscillators or NIST tests and probability distributions.

5. Please consider following recommendations.
- Please remove redundant space in line 10.
- Please add 'ring oscillator', 'TRNG' and 'metastability in FPGA' to the list of Keywords.
- Please avoid multiple definitions of the same abbreviations. It should be done once for the first use only (TRNG, FPGA).
- Please add Discussion section before the Conclusions. Here the mention problems of implementation in FPGA can be highlighted, as well as the distributions and NIST test results.

Author Response

The response is given in the PDF file.

 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

In this paper, a TRNG using capacitive oscillators for FPGAs is presented. The paper is interesting, brief and describes exactly its purpose without being verbose. Although the paper is interesting, my opinion is that there is not a high volume of work or knowledge that is presented. It would be more suitable for a conference, not a journal.

Since such works can be published in journals, I recommend that the author should evolve the proposed architecture to pass the only NIST test that fails. That could add more research and a more complete paper.

Although the design passes most of the test cases and is considered as a good TRNG, it seems that systematically fails to the non overlapping matching test.

I would suggest the author to improve the design to pass all tests in the revised version.

Let the author include doi in all references.

Author Response

The response is given in the PDF file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Most of my previous concerns have been addressed. 

Despite the fact that the novelty of this article is somewhat restricted and it does not exhibit superiority over other works, the author has presented a viable design approach for True Random Number Generators and carried out adequate experimental validation. In my opinion, this constitutes a satisfactory technical report that is worthy of publication. 

Author Response

I would like to thank the reviewer for his positive opinion pertaining to my article and valuable comments. Minor changes introduced to the article text are highlighted in blue.

Reviewer 4 Report

Comments and Suggestions for Authors

My understanding was that the p-value should be greater than 0.01 to pass the test.
Perhaps, I miss something.
In order for the paper to be self contained, let the authors add a small section to describe briefly each test, success values etc.
This seems necessary to avoid reader misunderstandings.

Author Response

I would like to thank once again the reviewer for his comments and suggestions.

Comment:
n order for the paper to be self contained, let the authors add a small section to describe briefly each test, success values etc.

This seems necessary to avoid reader misunderstandings..

Response:
Thank you for this remark. A dedicated paragraph has been added. The new text is highlighted in blue.