Cooperative Computing Offloading Scheme via Artificial Neural Networks for Underwater Sensor Networks

Round 1

Reviewer 1 Report

Cooperative Computing Offloading Scheme via Artificial Neural Networks for Underwater Sensor Networks

 

The present work establishes on well explained context.

 

 It is interesting and relevant.

 

There are some “error! Reference source not found”

 

The algorithm on page 14 could be summarize

 

Reference format should be equal

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have presented the application of ANN in UWSN for a cooperative computing offloading scheme. The topic is interesting, but there are several aspects to be improved in this manuscript. Following, I include a series of comments aimed at enhancing the quality of the paper:

1.       The introduction of the analysed problem in the abstract is too long. Please reduce the provided information to just one or two sentences.

2.       Consider avoiding the use of acronyms in the abstract, especially for those acronyms used once or twice. If the acronym is necessary, define it the first time it is used.

3.       In the abstract, the authors have to highlight their results, including numerical values of the performance of their proposal.

4.       The paper must be more focused on the sensors in order to ensure that it falls within the scope of the journal. Thus, more information and references to sensors and UWSN must be included in the introduction.

5.       To describe the contribution of the paper, it is recommended to use bullet points instead of different paragraphs.

6.       The aim of the paper must be extended.

7.       The section should be named Related Work, without the final s. Check this mistake with the word work (not works) along the paper.

8.       At the beginning of the related work, the authors must add a short paragraph introducing the content.

9.     At the end of each subsection of the related work section, authors must identify the gap in the current proposals and detail how their proposal will fill this gap.

10.  Sections 5 and 6 have the same title. Check and correct it, please.

11.  Part of the content of section 5 is not properly allocated as performance evaluation results. The authors have to create a new section in which the details of the simulation are provided, including the simulation parameters and their values, the metrics used to evaluate the performance, and the selected software for simulations. In this section, the authors also have to identify the selected algorithms used for the comparison and include their references.

12.  The results must be divided into subsections, providing more detailed comments and analysis.

13.  The authors should include a conclusion section where they summarise the contributions and findings of their work and include their future work.

 

General comments

1.       All equations must be cited in the text. Please check the equations and ensure that all of them are mentioned in the main text.

2.       When an acronym is defined, the first letter of each term should be capitalised.

 

3.       There are multiple errors in citing references. Check: Error! Reference source not found.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper discusses the application of artificial neural networks (ANNs) in cooperative computing offloading for underwater sensor networks (USNs), introducing edge servers (base stations) and utilizing coalition game theory for task processing. Simulations show the proposed approach outperforms existing methods. However, some details and explanations are lacking, making it difficult to fully assess the validity of the arguments.

Comments

1. The authors state that ANNs can help minimize the overall cost of computational delay and increase service performance, but they fail to provide concrete details on exactly how they achieve this. It would have been advisable to provide clear examples or data on the workings of ANNs and why they are an effective tool in this regard.

2. The authors do not offer a rationale for their choice of ANN model architecture. They should have justified why they opted for a specific number of layers and neurons, and how these choices contribute to the scheme's performance.

3. While the dynamic environment of USNs is considered, the authors fall short in explaining how the scheme handles base station movement. They must provide specifics to validate their claims.

4. Though the paper claims the proposed approach outperforms existing methods through simulations, the authors give no details on the methodology or metrics used for evaluation. Specifics are needed here to substantiate these claims.

The paper discusses the application of artificial neural networks (ANNs) in cooperative computing offloading for underwater sensor networks (USNs), introducing edge servers (base stations) and utilizing coalition game theory for task processing. Simulations show the proposed approach outperforms existing methods. However, some details and explanations are lacking, making it difficult to fully assess the validity of the arguments.

Comments

1. The authors state that ANNs can help minimize the overall cost of computational delay and increase service performance, but they fail to provide concrete details on exactly how they achieve this. It would have been advisable to provide clear examples or data on the workings of ANNs and why they are an effective tool in this regard.

2. The authors do not offer a rationale for their choice of ANN model architecture. They should have justified why they opted for a specific number of layers and neurons, and how these choices contribute to the scheme's performance.

3. While the dynamic environment of USNs is considered, the authors fall short in explaining how the scheme handles base station movement. They must provide specifics to validate their claims.

4. Though the paper claims the proposed approach outperforms existing methods through simulations, the authors give no details on the methodology or metrics used for evaluation. Specifics are needed here to substantiate these claims.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The topic of the article is interesting and have an adequate level of novelty since the Authors proposed a new method to define a scheme to process the computational tasks based on coalition game theory. The formal analysis and the mathematics are well-written. Unfortunately, there are a lot of English and formal errors. Also, the discussion of the results can be enlarged.

Please, find below a list if the issued to be resolved:

 

Line 9-10: the first sentence does not sounds good. Please reword

Line 21: the symbol “tanh” written in this way does not work. The reader can be misled and may not understand that is the hyperbolic tangent. Please find a way to clarify what it is (in the abstract).

Lines 87-200: Space doubled.

From Line 104: You reported the state-of-the-art of both Task Offloading,  Cooperative Computing Offloading Scheme, and Dynamic Characteristics in USNet. The part of the related works of ANN is missing. Please add something about ANN state-of-the-art also considering different applications like, for example, the use of ANN with satellite images.

Figure 1: Can be explained better in the corresponding caption. Is it a unique figure? Can be it splitted in panels?

Equation from 3 to 9 have the numbering on next line. Please correct.

Figure 2: The caption must be enlarged.

Lines 486-495-497-507-516: Some citations links aren’t working. Please correct.

Line 498: What ?(1) does it mean. It is not explained.

Figure 4 What does the colour of the boxes represent? Are they codified? If so please explain. If not, please use a unique colour.

Line 532 The point is missing.

Line 600 The point is doubled.

There are some english formal errors to be corrected. Also, check that you did not use the same words one sentence after another.

For example the adverb "However" to start two consecutive sentences.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I congratulate the authors for their efforts in the review process. Most of the comments are correctly addressed. There are only two small issues to be modified:

I suggest that 5.3. Simulation Results should be a completely new section 6. Simulation Results.

 

In conclusion, future work must be included in a new paragraph.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

accept the paper

Author Response

Thank you for your recognition and affirmation. We really appreciate your hard work in reviewing our manuscript. 

Reviewer 4 Report

Dear Authors,

I recognize your efforts to address all the points raised in my previous review. In my opinion, the revised version of the manuscript can be published in the journal, after minor revision (mostly regarding English editing).

English language can be further improved.

Author Response

Thank you for your comment. Following your suggestion, we have further polished the language, and during this period, I invited my friend Muhammad Waqas (he is Senior Lecturer in School of Computing and Mathematical Sciences, Faculty of Engineering and Science, University of Greenwich, UK, London) to proofread the English language of our manuscript. The text revised is highlighted in YELLOW in the revised manuscript.

We would like to thank you again for your hard work in reviewing our manuscript.