Leveraging Machine-Learning for D2D Communications in 5G/Beyond 5G Networks
, Basem M. ElHalawany 3,4,†, Kohei Hatano 1,5,†, Kaishun Wu 3,†and Ehab Mahmoud Mohamed 6,7,†Round 1
Reviewer 1 Report
In this paper, the authors discussed the recent research directions for D2D networks. Then, they summarized the most commonly used ML-based techniques, followed by a discussion on various applications of ML techniques in D2D networks. Moreover, some open issues and technical challenges for future research directions are provided. Finally, a case study is presented to show the efficiency of ML-based methods over conventional methods in future D2D communications. This work is valuable and worthy keeping in view the current trends and demand of ML algorithms in D2D communication. However, the following concerns need to be addressed prior to publication.
- In the abstract line#3 seems incomplete e.g. “Although D2D provides several benefits”, which can be replaced with “Although D2D communication provides several benefits”.
- Please write an abbreviation for “ISM” i.e., in Line#33
- Use proper sentences e.g., the sentence in line#157 “Unsupervised-learning can be sub-categorized into three tasks” the tasks can be replaced with categories/classes.
- The authors need to cite more papers in section 4. This section is very important for the readers. Hence, the researchers will focus on this. Consequently, there should be more recent literature and ML techniques cited in each section for example in NDS i.e., section 4.2, there are only two papers cited. Similarly, in the following subsections. e.g., ML for Power Control, there is only one paper referred in this subsection (4.3).
- It would be better if the pros and cons or at least the shortcomings of each approach in Table 1 are discussed. This will give further insight for readers to carry on their research.
- Please mention the simulation parameters in a Tabular form in section 6.
- Adjust Figure 7 before the conclusion section.
- Please mention the units on the x-axis and y-axis for Figure 7. i.e., for Time and Average Data rate.
Author Response
We would like to sincerely thanks the respected reviewer for the time and efforts he/she spent in reviewing our paper. Also, thanks too much for appreciating our humble work presented in this paper.
Kindly see the attached file
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper mainly discusses how we can apply ML techniques in 5G and 6G network to solve different challenges. There are no technical contributions, instead, it's like a review. Overall, this paper is well written and well structured. Since it's a review in nature, I don't have many comments. See my comments below.
The quality of the figures are not high. For example, in figure 1, the left part of the Ellipse is missing. In figure 5-7, the ratio of the texts are distorted. In terms of color, Figure 7 is much clearer than Fig5 and Fig 6. Their size can also be reduced.
Author Response
We would like to sincerely thanks the respected reviewer for the time and efforts he/she spent in reviewing our paper. Also, we highly appreciate the valuable comments given by the respected reviewer, which we do believe it enhances the clarity and readability of the revised manuscript.
Kindly see the attached file
Author Response File: Author Response.pdf
Reviewer 3 Report
1- All abbreviations have to specify on the first use.
2- Abbreviations on page 13 are well-known technical terms. This part should remove. If the authors want to add Abbreviations table, it should be given after the "Introduction" part, not the end of the manuscript.
3- Abstract is not clear. The points of the paper (problem definition, proposed scheme, contributions..) can be introduced better.
4 Organisation of the paper should be a separate paragraph.
5- The contributions of the paper should be listed for better understanding.
6- Literature review is not sufficient. The author should review similar papers, and introduce the differences between the proposed system model from similar models?
7-The author should clearly state the contribution of the paper instead of the expression "shed light on the recent research directions for D2D networks".
8-The relationship between Figure 2 and Figure 3 should be considered.
9- The Case Study part should be arranged in a more understandable way for communication experts.
10- The literature has to be strongly updated with some relevant and recent papers focused on the fields dealt with the manuscript [1], [2].
[1] Multi-user shared access in massive machine-type communication systems via superimposed waveforms. Physical Communication, 37, 100896 (2019).
[2] "An efficient transceiver design for superimposed waveforms with orthogonal polynomials," 2017 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom), Istanbul, 2017.
Author Response
We would like to sincerely thanks the respected reviewer for the time and efforts he/she spent in reviewing our paper. Also, we highly appreciate the valuable comments given by the respected reviewer
Kindly see the attached file
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
All of my concerns have been addressed. I accept the paper in its present form.
Reviewer 2 Report
The authors have addressed all my comments.
Reviewer 3 Report
Authors addressed all my previous comments, the manuscript is acceptable.
