Optimization of Drone Base Station Location for the Next-Generation Internet-of-Things Using a Pre-Trained Deep Learning Algorithm and NOMA
, Mahmoud Othman 4, Ishfaq Yaseen 5,*, Amani A. Alneil 5, Azza Elneil Osman 5 and Mohamed Ibrahim Alsaid 5
Round 1
Reviewer 1 Report
This paper is not written clearly and was very difficult to follow along.
1. I did not get what were some of the issues of existing research that you are trying to mitigate? Please clarify that and your contributions/novelty.
2. Go through the manuscript in details and fix incorrect english, grammatical errors.
3. did you conduct any ablation studies?
4. what are some signals that you are using to optimize the VGG network?
Author Response
Comments and Suggestions for Authors
This paper is not written clearly and was very difficult to follow.
- I did not get what were some of the issues of existing research that you are trying to mitigate. Please clarify that and your contributions/novelty.
Response: The existing problem that this abstract aims to mitigate is the challenge of meeting the requirements of next-generation Internet-of-Things (NG-IoT) applications in sixth-generation (6G) mobile networks, which involve large bandwidth, increased network capabilities, and low latency. The abstract notes that placing base stations (BSs) to meet these requirements is economically challenging.
- Go through the manuscript in detail and fix incorrect English, and grammatical errors.
Response: English corrections in manuscripts are corrected and grammar errors are checked.
- did you conduct any ablation studies?
Response: in the result section, we can see PSO, GWO, and CS is one optimizer compared with two-way optimizers -noma
- what are some signals that you are using to optimize the VGG network?
Response: Q-learning and particle swarm optimization to optimize the placement of deep brain stimulation (DBS) electrodes based on the VGG-19 network.
Reviewer 2 Report
In this paper, the authors use a two-layer optimizer based on a pre-trained VGG-19 26 model to identify a DBS, and run non-parametric statistical tests. However, the following comments need to be well addressed before considering acceptance.
Figure 2 is not clear and takes up too much space.
The headings in Figures 4 and 5 are irregular.
The corresponding simulation analysis should be added in the simulation experiment, instead of only displaying the simulation figures.
The author explains the contribution in the introduction, but the comparative literature is not added.
The format of references is not uniform in the current version. For example, there is obvious error with a formatting error in reference [16] and [26]. Please check the whole references carefully for similar problems and correct them.
Comments for author File: 
Comments.docx
Author Response
Comments and Suggestions for Authors In this paper, the authors use a two-layer optimizer based on a pre-trained VGG-19 26 model to identify a DBS and run non-parametric statistical tests. However, the following comments need to be well addressed before considering acceptance.
Figure 2 is not clear and takes up too much space.
Response: figure 2 is revised based on the given suggestion
The headings in Figures 4 and 5 are irregular.
Response: Headings of the figures are corrected
The corresponding simulation analysis should be added to the simulation experiment, instead of only displaying the simulation figures.
Response: we conducted Monte Carlo simulations to evaluate the performance of the methods discussed in the paper. They performed 1000 trials to generate the final results. Four propagation settings were considered, namely suburban, urban, dense urban, and high-rise urban environments, and the propagation parameters for each environment were provided in Table 2 . The purpose of the Monte Carlo simulations was likely to test the robustness and effectiveness of the proposed methods in different environments and scenarios.
The author explains the contribution in the introduction, but the comparative literature is not added.
Response: comparative literature is discussed in article section 2.
The format of references is not uniform in the current version. For example, there is an obvious error with formatting error in references [16] and [26]. Please check the whole references carefully for similar problems and correct them.
Response: References are corrected based on journal format
Reviewer 3 Report
A commendation to the authors on this exciting topic. However, the following limitations were noted:
1. The title is inappropriate based on the content and scope of work done and the need to reflect the need to submit to the journal "Mathematics".
The reviewer will suggest the following:
"Optimization of Drone Base Station Location for the Next Generation Internet of Things using Pre-trained Deep Learning Algorithm and NOMA"
2. Delete all abbreviations in the Abstract.
3. Line 50 " delete the period "." after the "disasters". [3]. should be disasters [3].
4. line 63: Kindly provide the full meaning of CNN and VGG since they appeared for the first time.
5. line 83: Kindly give the full meaning of NOMA and SWIFT here since they appeared for the first time
6. Line 132 please delete the "drone base station. simply use DBS
7. Line 134-135: kindly delete the "non-orthogonal multiple access" and use "NOMA"
8. Line 168: Kindly clarify the "condition" in the flow chart. Be more specific.
9. Lin 214: kindly delete "simultaneous wireless information and power transfer" and simply use "SWIPT"
10. Line 219: (PDMA) technology
11. Equation 11 is not correct "POn" or "PON"? see line 250 for "where PON" is the ........
12. Since the paper is on Next Generation IoT, how do Tables 3 and 4 values relate to the next generation IoT?
13. Line 370-372: Please note that conclusions must be based on the findings of a paper. Could authors show the section(s) of the paper that explored the effectiveness of swarm intelligence approaches?
Author Response
Comments and Suggestions for Authors
A commendation to the authors on this exciting topic. However, the following limitations were noted:
- The title is inappropriate based on the content and scope of work done and the need to reflect the need to submit to the journal "Mathematics".
The reviewer will suggest the following:
"Optimization of Drone Base Station Location for the Next Generation Internet of Things using Pre-trained Deep Learning Algorithm and NOMA"
Response: the title is modified based on the submission and core concept of research
- Delete all abbreviations in the Abstract.
Response: all the abbreviations are deleted
- Line 50 " delete the period "." after the "disasters". [3]. should be disasters [3].
Response: the dot is removed based on a comment
- line 63: Kindly provide the full meaning of CNN and VGG since they appeared for the first time.
Response: Abbreviations are stated in the line
- line 83: Kindly give the full meaning of NOMA and SWIFT here since they appeared for the first time
Response: meaning is stated in the line
- Line 132 please delete the "drone base station. simply use DBS
Response: deletion is done based on the suggestion
- Line 134-135: kindly delete the "non-orthogonal multiple access" and use "NOMA"
Response: deletion is done based on the suggestion
- Line 168: Kindly clarify the "condition" in the flow chart. Be more specific.
Response: deletion is done based on the suggestion
- Lin 214: kindly delete "simultaneous wireless information and power transfer" and simply use "SWIPT"
Response: deletion is done based on the suggestion
- Line 219: (PDMA) technology
Response: deletion is done based on the suggestion
- Equation 11 is not correct "POn" or "PON"? see line 250 for "where PON" is the ........
Response: Equation is done based on the given suggestion
- Since the paper is on Next Generation IoT, how do Tables 3 and 4 values relate to the next generation IoT?
Response: These results may have implications for the design and optimization of next-generation IoT networks. They suggest that different optimization algorithms may perform differently depending on the specific context and parameters of the network. Therefore, researchers and engineers developing IoT systems may need to carefully evaluate and compare different algorithms to select the most suitable one for their application. They may also need to consider the number of search agents and generations required to achieve optimal performance
- Line 370-372: Please note that conclusions must be based on the findings of a paper. Could authors show the section(s) of the paper that explored the effectiveness of swarm intelligence approaches?
Response: The conclusion is modified based on the given suggestion
Reviewer 4 Report
A Drone Base System for the Next-Generation Internet of Things using Pre-trained Deep Learning Methods and NOMA : The manuscript is about using drones instead of conversational base stations for IoT communications by using machine learning algorithms. The positions of the drones are determined for the lower path loss. The topic is current and well-written, and the manuscript has good grammar.
The authors should consider the following suggestions provided by the reviewer to improve the scientific depth of their manuscript. They should also address the following comments to improve the quality of the presentation of their manuscript.
1- The abstract is explanatory. Adding the numerical results achieved by the proposed method should be better than existing methods.
2- The authors should re-check the manuscript for notation according to a matrix, vector, and constant notations (i.e., bold, capital letter, lowercase letter).
3- It should be better to space-check all manuscript. (There are some double spaces and no space. )
4- The resolution of the Figures could be better. Markers or colours could be better. And the caption of Figure 12 should be checked.
5-The literature must be strongly updated with relevant papers focused on the requirements of the 5G within the manuscript [1].
[1] Adversarial machine learning security problems for 6G: mmWave beam prediction use-case." In 2021 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom), pp. 1-6. IEEE, 2021. DOI: 10.1109/BlackSeaCom52164.2021.9527756
Author Response
Comments and Suggestions for Authors
A Drone Base System for the Next-Generation Internet of Things using Pre-trained Deep Learning Methods and NOMA: The manuscript is about using drones instead of conversational base stations for IoT communications by using machine learning algorithms. The positions of the drones are determined for the lower path loss. The topic is current and well-written, and the manuscript has good grammar.
The authors should consider the following suggestions provided by the reviewer to improve the scientific depth of their manuscript. They should also address the following comments to improve the quality of the presentation of their manuscript.
- The abstract is explanatory. Adding the numerical results achieved by the proposed method should be better than existing methods.
Response: an abstract is checked based on the given suggestion
- The authors should re-check the manuscript for notation according to a matrix, vector, and constant notations (i.e., bold, capital letter, lowercase letter).
Response: all the notations are checked in the equation based on the suggestion
- It should be better to space-check all manuscripts. (There are some double spaces and no space. )
Response: spacing is checked based on the given suggestion
- The resolution of the Figures could be better. Markers or colors could be better. And the caption of Figure 12 should be checked.
Response: figures and resolution are checked based on the given suggestion. captions are checked
5-The literature must be strongly updated with relevant papers focused on the requirements of the 5G within the manuscript [1].
[1] Adversarial machine learning security problems for 6G: mmWave beam prediction use-case." In 2021 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom), pp. 1-6. IEEE, 2021. DOI: 10.1109/BlackSeaCom52164.2021.9527756
Response: the reference is included based on suggestion.
Round 2
Reviewer 1 Report
No further comments
Reviewer 2 Report
No further comments.
Reviewer 3 Report
Concerns have been addressed
Reviewer 4 Report
The author answered all my questions. In my opinion the manuscript is acceptable.
