Two-Level Dynamic Programming-Enabled Non-Metric Data Aggregation Technique for the Internet of Things

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article proposes a complex two-layer data aggregation method for IoT networks to minimize the expected ratio of repeated data values in the refined set of IoT networks and minimize information loss rates as much as possible. I have carefully reviewed your research work and provided the following review comments, hoping to help improve your paper:

 

1.When introducing your two-layer data aggregation method in the first section of the article, you should draw a model diagram of your proposed method or describe the implementation steps of your method in detail, as the article lacks a corresponding flowchart.

 

2. When evaluating the experiment, the article uses images to demonstrate the effect, and does not mention key numbers to illustrate the improvement effect. The evaluation method lacks formulas and theoretical proof.

 

3. Summary of the article and future work. It summarizes the problems solved by the methods proposed in the article, but does not mention the problems in future research, which is inconsistent with the theme of this section.

Comments on the Quality of English Language

Minor editing of English language required

Author Response

File attached.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. In the abstract, the following text is repeated twice:

“To address this, the paper presents a sophisticated two-tier data aggregation approach for IoT networks. First, at the device level, a local data aggregation process filters out outliers and duplicate data before transmission. Second, at the server level, a dynamic programming-based non-metric method identifies the longest common subsequence (LCS) among neighboring device data, which is then shared with the edge module.”

 Please, remove the repeated text.

2. In section 4, the authors should explain how they obtained the values of performance evaluation metrics that resulted from applying approaches other than the one they proposed for comparison. If they had applied these approaches themselves using the same simulator, this comparison would not be fair enough to rely on in preference to the method proposed by the authors.

3. To ensure no bias when comparing performance with other competitive approaches for data aggregation, it is better, from my point of view, to choose similar cases of data aggregation with published performance metrics that are resulting from applying these competitive approaches and then compare with those published results.

4. Please, mention the development environment you employed to  develop the listed algorithms 1-3.

 5. There is no need for the legend in Figures 1 and 8.

 

Author Response

 

Response to reviewer 2

Thank you for the opportunity to revise our manuscript titled "Sophisticated Data Aggregation Technique for the Internet of Things: A Dynamic Programming-enabled Methodology." We are grateful for the time and effort the editor(s) and reviewer(s) invested in reviewing our paper and providing detailed feedback. Your insightful comments and suggestions have been instrumental in enhancing the quality and clarity of our manuscript.

Below, we have provided concise and specific responses to each of the comments raised by the reviewers. Additionally, all changes in the revised manuscript have been highlighted in blue text colour to facilitate easy identification of the modifications. We hope that these amendments adequately address the concerns and further the manuscript's contribution to the field.

1. In the abstract, the following text is repeated twice:

“To address this, the paper presents a sophisticated two-tier data aggregation approach for IoT networks. First, at the device level, a local data aggregation process filters out outliers and duplicate data before transmission. Second, at the server level, a dynamic programming-based non-metric method identifies the longest common subsequence (LCS) among neighboring device data, which is then shared with the edge module.”

 Please, remove the repeated text.

Response:

As recommended by the reviewer, we have removed repeated text from the revised manuscript. These changes are indicated in blue on page [1]. The revised text is also copied below for your reference:

 

2. In section 4, the authors should explain how they obtained the values of performance evaluation metrics that resulted from applying approaches other than the one they proposed for comparison. If they had applied these approaches themselves using the same simulator, this comparison would not be fair enough to rely on in preference to the method proposed by the authors.

Response

Thank you for your comments regarding the performance evaluation metrics in Section 4. To address your concerns, we clarify that all the simulation results presented are derived from experiments conducted using the OMNeT++ simulator, a renowned open-source tool tailored for networking infrastructures. The simulation setup, as detailed in the manuscript, involved various performance metrics including data fusion efficiency, duplicate data ratio, system longevity, packet transmission, and congestion control.

We ensured uniform conditions across all tested algorithms by implementing them in a controlled environment where several devices were randomly deployed within the effective range of multiple server modules, each linked directly to a centralized edge unit for decision-making. This setup mimics realistic IoT deployment scenarios, utilizing the Wasp-Mote Board from Libelium with a standard communication range of approximately 450 meters.

The simulation parameters, including the number and deployment of devices, server modules’ capacity, and the battery power specifications of the devices (ranging from 1,150 mAh to 13,000 mAh), are meticulously outlined in Table 1 in our manuscript. These parameters were consistently applied across both the proposed and existing schemes to ensure a fair comparison.

 

We believe the consistency in simulation parameters across all compared algorithms justifies our claim that the proposed scheme outperforms the existing ones. However, we remain open to further discussion and would welcome any specific suggestions from the reviewer on alternative approaches or mechanisms for comparison to enhance the robustness of our findings. The highlighted text is copied here as well for your reference.

 

3. To ensure no bias when comparing performance with other competitive approaches for data aggregation, it is better, from my point of view, to choose similar cases of data aggregation with published performance metrics that are resulting from applying these competitive approaches and then compare with those published results.

Response

Thank you for your suggestion on ensuring unbiased comparisons in our data aggregation performance evaluations. In our study, we have indeed considered similar cases of data aggregation by utilizing benchmark datasets with published performance metrics, which are readily accessible online for verification purposes. Furthermore, both the proposed and existing algorithms were rigorously evaluated on real-time datasets collected from our deployed IoT infrastructure.

This approach allows us to maintain a high level of transparency and reproducibility in our comparisons, ensuring that our findings are both reliable and applicable across various real-world scenarios. We believe this methodology supports the robustness of our comparative analysis and the conclusions drawn from it.

4. Please, mention the development environment you employed to  develop the listed algorithms 1-3.

Response

We have mentioned in the simulation section that these results are obtained through the implementation of these schemes in OMNETT++, which is an open-source simulation software particularly developed for networking infrastructures. The highlighted text is copied here as well for your reference.

 

5. There is no need for the legend in Figures 1 and 8.

Response

Thank you for your feedback regarding the legends in Figures 1 and 8. We have carefully considered your suggestion. While we believe that the legends contribute significantly to the readability and comprehension of these figures, we respect your perspective and are open to removing them if you deem it essential for the clarity of the paper. We are committed to adhering to the reviewers' recommendations to ensure the manuscript meets the highest standards of presentation.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In this paper, a two-tier data aggregation methodology is proposed that aimed at enhancing the quality of refining captured data values while minimizing information loss in IoT networks. The comments are as follows.

1. What are the primary causes of duplicate data values in IoT networks, and why is mitigating this issue important for the efficiency and effectiveness of IoT systems?

2. Can you elaborate on the existing methodologies reported in the literature for addressing duplicate data values in IoT networks, and what are their limitations or challenges?

3. The overall presentation needs improvement. 

4. The introduction needs strengthening, with explanations and current research status provided for all keywords. It is recommended to break down longer paragraphs into smaller ones.

5. The grammar section needs attention. For example, "[22] have identified" should be "[22] has identified".

6. What are the potential practical applications or scenarios where this sophisticated data aggregation approach could be deployed in IoT networks, and what are the expected benefits in real-world implementations?

7. Are there any potential drawbacks or limitations to consider when implementing this two-tier data aggregation approach in IoT systems, such as scalability issues or additional computational overhead?

8. How does the proposed approach compare with existing state-of-the-art solutions in terms of performance, complexity, and scalability, and what are the key advantages that make it stand out?

9. Can you discuss any potential future research directions or extensions based on the findings of this study, such as exploring alternative algorithms or considering different network architectures for further improvement?

10. Do not insert images into the conclusion.

Comments on the Quality of English Language

Extensive editing of English language required.

Author Response

Response to reviewer 3

Thank you for the opportunity to revise our manuscript titled "Sophisticated Data Aggregation Technique for the Internet of Things: A Dynamic Programming-enabled Methodology." We are grateful for the time and effort the editor(s) and reviewer(s) invested in reviewing our paper and providing detailed feedback. Your insightful comments and suggestions have been instrumental in enhancing the quality and clarity of our manuscript.

Below, we have provided concise and specific responses to each of the comments raised by the reviewers. Additionally, all changes in the revised manuscript have been highlighted in blue text colour to facilitate easy identification of the modifications. We hope that these amendments adequately address the concerns and further the manuscript's contribution to the field.

In this paper, a two-tier data aggregation methodology is proposed that aimed at enhancing the quality of refining captured data values while minimizing information loss in IoT networks. The comments are as follows.

1. What are the primary causes of duplicate data values in IoT networks, and why is mitigating this issue important for the efficiency and effectiveness of IoT systems?

Response

Thank you for raising this important point regarding the primary causes of duplicate data values in IoT networks and the significance of mitigating this issue. We have added the following text in the paper to address this issue (on page 2 in blue text).

 

2. Can you elaborate on the existing methodologies reported in the literature for addressing duplicate data values in IoT networks, and what are their limitations or challenges?

Response

Thank you for your inquiry regarding the existing methodologies for managing duplicate data values in IoT networks. We have extensively reviewed and discussed these methodologies in the literature review section of our manuscript on pages 4-5. Each method is thoroughly analyzed, and their respective limitations and challenges are delineated. These details have been highlighted in blue in the revised manuscript for easy reference and to facilitate a deeper understanding of the current challenges in this field.

We aim to provide a comprehensive overview of the state-of-the-art techniques and to highlight where further research and development is needed to overcome these existing limitations. Should you require more specific details or additional references beyond what is presented, we would be pleased to provide them. Part of the revised text is also copied below for your reference.

 

3. The overall presentation needs improvement.

Response

Thank you for your feedback on the presentation of our paper. We have taken your comments into consideration and have made comprehensive improvements throughout the manuscript. These changes are highlighted in blue in the revised document to facilitate easy identification and review. We believe that these modifications enhance the clarity, flow, and overall readability of the paper. We appreciate your guidance and are open to further suggestions to improve our work.

 

4. The introduction needs strengthening, with explanations and current research status provided for all keywords. It is recommended to break down longer paragraphs into smaller ones.

Response

Thank you for your constructive feedback on the introduction of our manuscript. We have carefully revised this section to provide a clearer explanation of the key concepts and the current status of research in this field. We have also aimed to delineate the problem statements more distinctly. These revisions have been highlighted in blue in the updated document to assist in easy identification.

 

Additionally, we have broken down longer paragraphs into smaller, more digestible ones to improve readability and ensure that each concept is adequately emphasized and accessible. We trust that these changes address your concerns effectively, and we welcome any further suggestions to enhance our presentation. The changes are in blue text in the introduction section of the revised paper.

5. The grammar section needs attention. For example, "[22] have identified" should be "[22] has identified".

Response

As suggested by the reviewer, grammar is improved.

 

6. What are the potential practical applications or scenarios where this sophisticated data aggregation approach could be deployed in IoT networks, and what are the expected benefits in real-world implementations?

Response

The proposed scheme is feasible for both traditional as well as resource-constrained networks such as the Internet of Things, wireless sensor networks, and ad-hoc networks. The main benefit of this scheme is that the accuracy and precision ratio of the underlying system are improved. Additionally, its effectiveness becomes high as it has to carry out operations on a relatively smaller and duplicate-free dataset. We have added the following text in the revised manuscript on page 3.

 

7. Are there any potential drawbacks or limitations to consider when implementing this two-tier data aggregation approach in IoT systems, such as scalability issues or additional computational overhead?

Response

Thank you for your inquiry about the potential limitations of our proposed two-tier data aggregation approach when implemented in IoT systems. We recognize that additional computational overhead and processing time are required at both the device and server levels. Moreover, there is an increase in storage requirements as the algorithms need to be deployed across these modules.

We have carefully analyzed and compared the processing time of our proposed approach with existing techniques, as detailed in Section 4.1.5, Computational OR Processing Time Metric. The results, illustrated in Figure 8, demonstrate that while our approach does introduce some additional overhead, it remains efficient and outperforms current methods by completing the aggregation process within the minimum possible time.

Regarding scalability, our approach is designed to be adaptable to various network sizes and complexities. However, as with any system, there may be trade-offs between scalability and performance at extreme scales, which could be an avenue for future research. We appreciate your attention to these details and are committed to continuous improvement of our methodologies. Section 4.1.5 is copied here for your reference.

 

8. How does the proposed approach compare with existing state-of-the-art solutions in terms of performance, complexity, and scalability, and what are the key advantages that make it stand out?

Response

Thank you for your question concerning the comparative analysis of our proposed approach against existing state-of-the-art solutions. In the simulation section of our revised manuscript, which is highlighted in blue for ease of reference, we provide an in-depth discussion of the performance, complexity, and scalability comparisons.

Our proposed approach demonstrates significant advantages in performance metrics, exhibiting superior efficiency in data aggregation and processing times. In terms of complexity, while our method introduces novel elements that add to the computational layers, it remains competitively manageable when compared to existing solutions. Regarding scalability, the two-tier structure of our approach is designed to adapt effectively to varying network sizes and densities, ensuring robust performance even as the system expands.

The key advantages that distinguish our method include its enhanced accuracy in outlier detection and its innovative use of dynamic programming to optimize data aggregation. These strengths contribute to its standout performance in both small-scale and large-scale IoT network scenarios. We believe that these enhancements, alongside the method's adaptability and efficient processing capabilities, solidify its position as a noteworthy advancement in IoT data management.

9. Can you discuss any potential future research directions or extensions based on the findings of this study, such as exploring alternative algorithms or considering different network architectures for further improvement?

Response

Thank you for your interest in the potential future research directions stemming from our study. We have outlined these prospects in the revised manuscript, with the specific discussion highlighted in blue on pages 16-17 for your convenience. To directly address your query, our findings open several avenues for further investigation. These include the exploration of alternative algorithms that may offer improved efficiency or reduced computational overhead. Additionally, we see potential in examining different network architectures that could leverage the strengths of our proposed data aggregation method, possibly enhancing its scalability and robustness in diverse IoT environments. The text has also been copied here for your reference.

 

10. Do not insert images into the conclusion.

Response

Thank you for your suggestion regarding the inclusion of images in the conclusion section. We have reviewed the manuscript to ensure compliance with this guideline and confirm that no images have been inserted in the conclusion section. If there are any further adjustments needed, please let us know, and we will address them promptly.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Minor edits required:

1- In Figure 2, remove the legend as there is only one curve in the figure. You can change the y-axis title to be: “Device reading, temperature (oC)”. Also, the x-axis title should be changed to be: “Time (s)”. Add the note “Sampling time interval =4s” to the Figure caption.

2- In Figure 9, remove the legend as there is only one curve in the figure. Also, the x-axis title should be changed to “Number of devices in the IoT”

3- In page 2 of the pdf, line 69, correct the reference citation mark [ijcs].

Comments on the Quality of English Language

1- In all the figure captions, only the first word must begin with a capital letter; the remaining words must begin with lowercase letters.

2-It is not appropriate to start a sentence with a citation mark (as the subject) like the following examples:

“[25] has identified numerous potentials of a well-known procedure, ..”, page 5, line 203.

and

“[26] have proposed and described an effective methodology..” page 5, line 206

Instead, you can rephrase the above sentences as follows:

“The work of [25] has identified numerous potentials of a well-known procedure,..”,  or “Potentials of a well-known procedure have been identified in [25] ....”.

and

“The work of [26] has proposed and described an effective methodology..” , or “An effective methodology has been proposed and described in [26] ...”

Author Response

Reviewer Comments	Authors Response
Minor edits required: 1- In Figure 2, remove the legend as there is only one curve in the figure. You can change the y-axis title to be: “Device reading, temperature (oC)”. Also, the x-axis title should be changed to be: “Time (s)”. Add the note “Sampling time interval =4s” to the Figure caption.	Thank you for the valuable insights provided by the reviewer. We have incorporated the suggested changes, such as removing the caption and adjusting the titles of the x-axis and y-axis as recommended. These modifications have been clearly highlighted in the revised manuscript for the reviewer's convenience.
2- In Figure 9, remove the legend as there is only one curve in the figure. Also, the x-axis title should be changed to “Number of devices in the IoT”	We deeply value the insightful suggestion from the esteemed reviewer. We have now implemented it by removing the legend from Figure 9 and adjusting the titles of the x-axis in accordance with your recommendations. These revisions notably improve the clarity and impact of the presentation.
3- In page 2 of the pdf, line 69, correct the reference citation mark [ijcs].	Thank you for your valuable feedback. We have now accommodated the changes recommended by the reviewer by inserting the correct reference and highlighted them in the revised manuscript.
4- In all the figure captions, only the first word must begin with a capital letter; the remaining words must begin with lowercase letters.	Thank you. We have now implemented the changes suggested and updated the revised manuscript accordingly.
5-It is not appropriate to start a sentence with a citation mark (as the subject) like the following examples: “[25] has identified numerous potentials of a well-known procedure, ..”, page 5, line 203. and “[26] have proposed and described an effective methodology..” page 5, line 206 Instead, you can rephrase the above sentences as follows: “The work of [25] has identified numerous potentials of a well-known procedure,..”,  or “Potentials of a well-known procedure have been identified in [25] ....”. and “The work of [26] has proposed and described an effective methodology..” , or “An effective methodology has been proposed and described in [26] ...”	We are grateful to the reviewer for his insightful comments and kind suggestions. These recommendations will aid us in this and future research endeavours. We have now incorporated the suggested changes and highlighted them in the updated version of the paper.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The revision is acceptable.

Comments on the Quality of English Language

Good.

Author Response

Thank you for the review. We have tried our level best to address all your concerns.