An Intrusion Detection System for RPL-Based IoT Networks
and Vassilios G. Vassilakis *Round 1
Reviewer 1 Report
1. Table 1 is not clear. Why is the author using it?
2. Conclusion write precisely
3. Table 2 and figure 12 are giving the same information.
4. What is the computational complexity of the proposed approach? - Discuss the limitations and future scope of this work.
5. Authors need to clearly discuss their proposed architecture, and proposed methodology, and justify their approach, If possible, they can summarize their proposed method in an algorithmic method. Additionally, authors can justify their analysis or results that show the achieved parameters with respect to the use cases and be able to verify their results benchmarked against related existing literature.
6. Elaborate more on the Dataset, what are the highlights and what is the significance of the dataset choosen for this research paper.
7. The authors should add other exciting works as there are many research papers in this similar research area. Like:
1. Lakshmanna, K., Kaluri, R., Gundluru, N., Alzamil, Z. S., Rajput, D. S., Khan, A. A., ... & Alhussen, A. (2022). A Review on Deep Learning Techniques for IoT Data. Electronics, 11(10), 1604.
2. Vashishtha, M., Chouksey, P., Rajput, D. S., Reddy, S. R., Reddy, M. P. K., Reddy, G. T., & Patel, H. (2021). Security and detection mechanism in IoT-based cloud computing using hybrid approach. International Journal of Internet Technology and Secured Transactions, 11(5-6), 436-451.
In a conclusion, the technical content is good. Therefore, the contribution of this article is also satisfactory. I am accepting an article with minor revision for publication in this journal.
Author Response
Reviewer #1:
- Table 1 is not clear. Why is the author using it?
Authors’ response: We agree that Table 1 does not add much value to the manuscript, as all this information has already been discussed in Section 3. Therefore, we excluded Table 1 from our revised manuscript.
- Conclusion write precisely
Authors’ response: We have improved Conclusion and provided more specific information about the overhead introduced by the proposed solution.
- Table 2 and figure 12 are giving the same information.
Authors’ response: Thank you for this comment. In the revised version, we retained Table 2 and deleted Figure 12 so as not to repeat the same information.
- What is the computational complexity of the proposed approach? - Discuss the limitations and future scope of this work.
Authors’ response: The computational complexity of the proposed approach is very low. This is evidenced by low CPU usage overhead (no more than 2%) as shown in Section 5. We have highlighted this information in the Evaluation and Conclusion sections.
- Authors need to clearly discuss their proposed architecture, and proposed methodology, and justify their approach, If possible, they can summarize their proposed method in an algorithmic method. Additionally, authors can justify their analysis or results that show the achieved parameters with respect to the use cases and be able to verify their results benchmarked against related existing literature.
Authors’ response: Thank you for this comment. In Section 5, our results have been benchmarked against the LHP protocol of [37] and show a negligible overhead of 1.4% in terms of TX/RX power consumption. On the other hand, our proposed solution is able to detect a large class of greyhole attacks, which is not possible by the original LHP protocol.
- Elaborate more on the Dataset, what are the highlights and what is the significance of the dataset choosen for this research paper.
Authors’ response: In this study, we don’t use datasets for training and testing purposes, as our detection methods do not rely on machine learning or deep learning, as e.g., in [13] or [16]. Instead, we generate network topologies and attack traffic using the Cooja simulator as explained in Section 4. This methodology is similar to other works, such as [39-41],
- The authors should add other exciting works as there are many research papers in this similar research area. Like:
- Lakshmanna, K., Kaluri, R., Gundluru, N., Alzamil, Z. S., Rajput, D. S., Khan, A. A., ... & Alhussen, A. (2022). A Review on Deep Learning Techniques for IoT Data. Electronics, 11(10), 1604.
- Vashishtha, M., Chouksey, P., Rajput, D. S., Reddy, S. R., Reddy, M. P. K., Reddy, G. T., & Patel, H. (2021). Security and detection mechanism in IoT-based cloud computing using hybrid approach. International Journal of Internet Technology and Secured Transactions, 11(5-6), 436-451.
Authors’ response: Thank you for this suggestion. We have cited the suggested papers and also included some other recent works that focus on similar research problems. In particular the following papers have been added in Introduction and Section 3: [13-17]
In a conclusion, the technical content is good. Therefore, the contribution of this article is also satisfactory. I am accepting an article with minor revision for publication in this journal.
Authors’ response: Thank you for your feedback.
Reviewer 2 Report
In this paper, the author proposes a novel intrusion detection system (IDS) has been designed and implemented for the RPL-based IIoT.
However, there still are some problems in this paper:
1. In the Abstract section, the authors do not describe how far their proposed method can achieve.
2. In the introduction section, this paper does not have illustrations of motivation.
3. It is recommended that the author introduce the meaning of each symbol in the algorithm in detail and write it in a table.
4. Authors should add more relevant literature to increase the persuasiveness of the article, find related articles such as: 'Image Super-resolution with An Enhanced Group Convolutional Neural Network ' https://arxiv.org/abs/2205.14548 , 'Multi-stage image denoising with the wavelet transform ' https://www.sciencedirect.com/science/article/abs/pii/S0031320322005301 , 'A heterogenous group CNN for image super-resolution ' https://ieeexplore.ieee.org/abstract/document /9919146.
5. Figure 5 is not complete, please include the complete picture.
6. Authors are advised to cite references in the compared algorithms.
7. A punctuation mark appears on page 9, line 436.
8. Inconsistent '[]' notation in Algorithm 5.
9. Authors are advised to use '*' in tables 2 and 3 to represent the best data.
Author Response
Reviewer #2:
In this paper, the author proposes a novel intrusion detection system (IDS) has been designed and implemented for the RPL-based IIoT.
However, there still are some problems in this paper:
- In the Abstract section, the authors do not describe how far their proposed method can achieve.
Authors’ response: Thank you for your comment. In the revised manuscript, we have updated the Abstract and provided more specific information about the overhead introduced by the proposed solution.
- In the introduction section, this paper does not have illustrations of motivation.
Authors’ response: In the revised version, in the Introduction we included some statistics from recent reports to highlight the size of the problem and at the same time identify the motivation for our research.
- It is recommended that the author introduce the meaning of each symbol in the algorithm in detail and write it in a table.
Authors’ response: We agree that it would be helpful to summarise this information in a table. In Section 4 we added Table 1 which explains the meaning of symbols used in Algorithms 1-5.
- Authors should add more relevant literature to increase the persuasiveness of the article, find related articles such as: 'Image Super-resolution with An Enhanced Group Convolutional Neural Network ' https://arxiv.org/abs/2205.14548 , 'Multi-stage image denoising with the wavelet transform ' https://www.sciencedirect.com/science/article/abs/pii/S0031320322005301 , 'A heterogenous group CNN for image super-resolution ' https://ieeexplore.ieee.org/abstract/document /9919146.
Authors’ response: Thank you for this remark. We have cited the suggested papers and also included some other recent works that focus on similar research problems. In particular the following papers have been added in Introduction, Section 3, and Conclusion: [13-17] and [59-60].
- Figure 5 is not complete, please include the complete picture.
Authors’ response: Thank you for bringing this to attention. In Subsection 5.1.1., we included a more detailed explanation of Figure 5. We believe that now the role of each node in this specific scenario is clearer.
- Authors are advised to cite references in the compared algorithms.
Authors’ response: In the revised manuscript, we have cited all the references related to the compared algorithms. In particular, we cited the following references [37] and [49].
- A punctuation mark appears on page 9, line 436.
Authors’ response: This punctuation mark has been removed.
- Inconsistent '[]' notation in Algorithm 5.
Authors’ response: We have now corrected the [] notation in Algorithm 5, where the i parameter used to appear both in italics and normal fonts. We have now corrected this to appear in italics in the square brackets [].
- Authors are advised to use '*' in tables 2 and 3 to represent the best data.
Authors’ response: Thank you for this comment. We agree that generally it would be helpful to highlight the best data, however in our case these data are not directly comparable to each other. For example in Table 2 TX consumption is always lower than RX consumption and RAM usage is always lower than ROM usage. Also, it is not convenient to directly compare the different node types, for example the detector IDS is expected to have higher overhead than the dummy node given their different roles in the operation of the IDS.
Reviewer 3 Report
In this paper, the authors proposed a novel intrusion detection system (IDS) and implemented for the RPL-based IIoT. The article is technically sound and well-written. The authors should only proofread the contents to eliminate grammatical and punctuation mistakes. The article is ready for publication in the current form from my side.
Author Response
Reviewer #3:
In this paper, the authors proposed a novel intrusion detection system (IDS) and implemented for the RPL-based IIoT. The article is technically sound and well-written. The authors should only proofread the contents to eliminate grammatical and punctuation mistakes. The article is ready for publication in the current form from my side.
Authors’ response: Thank you for your feedback. We have performed a thorough proofreading of the manuscript and we believe that all grammatical and punctuation mistakes have now been eliminated.
Round 2
Reviewer 2 Report
This paper has been improved a lot.
