Fibonacci Group Consensus Algorithm Based on Node Evaluation Mechanisms

Round 1

Reviewer 1 Report

The topics covered in the manuscript are of great interest and impact for a large audience, and this is certainly a good thing. However, the manuscript contains a number of problems, some of which are highlighted below:

- Mathematical formulas must be completely rewritten and even rethought. It is also recommended to use a LaTeX editor. (A striking example: the use of juxtaposed symbols such as >= is, to say the least, out of place in a scientific article.)

- The references and bibliography are not very up to date and/or international. You use mathematical tools like Fibonacci numbers and you should also look into recent mathematical innovations on blockchain protocols. Many recent important works are not mentioned to the reader. An example: in doi 10.3390/math10173040 mathematical tools such as elliptic curve based lotteries, neural networks, etc. are used and they seem to constitute a rather innovative framework. Search among the most relevant and recent papers on this example.

- A comparison of Tx/sec is made when the consensus is needed to generate a block and not to understand how "fast" a network is. Where were the 100 used nodes? Where they in the same machine?

- The idea of halving the number of nodes participating in the consensus to increase performance can have important disadvantages and, in any case, should be discussed in more depth.

- Some statements are very strong, they should be moderated.

- It is not clear why the Fibonacci sequence is preferable to so many other possible functions or sequences. Why exactly Fibonacci? Then, from the point of view of security, why prefer Fibonacci?

- The figures are often of poor quality and, like the tables, not well proportioned or well placed.

- The graphs that measure the "performances" lack information. For example, which PBFTs were used?

- What does the "delay" refer to? At the time of communication in the network of nodes? Why not compare this algorithm with a DPoS?

- Is the consensus based on a defined function (Fibonacci) and therefore is it possible to know a priori which nodes will participate in the consensus?

- Manuscript editing needs improvement in many ways. There are also extra and missing spaces in the text.

- Etc.

 

In conclusion, the manuscript requires a very major revision. Only if the authors carefully follow the comments and observations above, the manuscript could continue the review process.

The English language used in the paper does not contain serious problems, but can be improved in some points. Instead, as already mentioned above, the editing of the paper is very poor.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript shows some methods for removing mistakes and optimising the parallel, independent decision. It does not fit direct into the journal's scope. It should be better to send it to a journal with computer science in scope as the main subject. 

 

Shortcomings: 

1. There are many mistakes in language. It is very hard to read it. The authors should use a pro spellchecker or native speaker correction. 

2. More than half of reference items are out of date. Ref items older than three years have ancient data in the manuscript's subject. The ancient referenced data courses that the article's novelty cannot be assessed. After correcting this point, authors should write a new version of the abstract, background research and conclusion sections. 

3. There is too small background research. Background research is based partly on ancient data. 

4. All abbreviations must be explained at the first occurrences in the text. 

5. After the section (subsection), the reader should find some introduction for the deeper subsection. 

 There are many mistakes in language. It is very hard to read it. The authors should use a pro spellchecker or native speaker correction. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear authors,

Thank you for sharing your research on  SP-PBFT consensus algorithms that take a speculative approach depending on the network environment of the system and apply different steps to reach consensus for different network environments.

You may find suggestions/remarks on your proposed published paper in the uploaded file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Thank you for your submission. Very interesting. 

The topic is important, focusing on improving the PBFT consensus algorithm, which is highly relevant and interesting. The presentation of results seems clear and detailed, providing a good understanding of the performance of the proposed algorithm compared to the traditional PBFT and its improved versions. The paper appears well-structured, with clear sections for introduction, methods, results, and conclusions, which aids in readability and understanding.

The introduction provides a good overview. The scientific soundness of the work could be improved by providing more details about the research design and methodology, ensuring that the approach is robust and reproducible. It may help to provide more details on the node reputation calculations and Fibonacci grouping implementation.

I would like to see a discussion about how the most reputable nodes are protected from overworking and clarification of how the Fibonacci function would prevent accumulation of reputation value. The work distribution appears to lean heavily on a small subset of nodes.

Overall, excellent work.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors did a good work of editing. The answers to the questions raised are very detailed indeed. I think the manuscript is now ready or nearly ready for publication. Small editing details can be defined with the editorial office. In the meantime, to improve further and save time, I point out some things that I could see in reading:

 

- I would suggest improving the quality of some figures like Fig. 2.

- Consult the editorial office for optimal sizing of tables and figures.

- What can be said about the latest research involving Fibonacci numbers in mathematics and especially algebra and geometry? These are very important lines of research that go back centuries, but which have also led to interesting developments in the last few years.

- The mathematical formulas would need some adjustment. For example, words like "Credit" should be written in Roman in both formulas and text, while mathematical variables should be written in mathematical characters (similar to italics). In the formula in line 286 the opposite is done almost exactly... Check and correct all the formulas.

- Check all the scattered formulas also in the text. For example, in lines 641 and 642, the big O notation should be written in mathematical characters (similar to italics).

 

 

The English language is good, except for minor details which can be checked by the editorial office. 

Reviewer 2 Report

After the corrections, the manuscript reads much better. It would also be worth replacing the abbreviation in the title of the second section. Using abbreviations in titles is not a good habit, making it difficult for the reader to understand. After replacing it, the manuscript will be able to the next step of the publication process. 

English is better, but I still see small mistakes.