AoI-Aware Resource Scheduling for Industrial IoT with Deep Reinforcement Learning
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper "AoI-aware Resource Scheduling for Industrial IoT with Deep Reinforcement Learning" addresses a crucial aspect of the industrial Internet of Things (IIoT) – resource scheduling in time-sensitive scenarios. The focus is on the Age of Information (AoI), a key metric that measures the freshness of data in IIoT environments. The paper presents a novel approach to optimize resource scheduling, balancing the need to minimize AoI and maximize network throughput. feedback to improve the paper:
• Discuss the scalability of the proposed algorithm in larger and more complex IIoT networks.
• Address how the algorithm can be implemented in real-world IIoT settings and potential challenges.
• Evaluate the energy efficiency of the proposed approach, given the resource constraints in IIoT devices.
• Examine the algorithm's robustness in dynamic network conditions, including variable packet arrival rates and channel conditions.
• Discuss the requirements for training and accessibility for IIoT operators to effectively utilize the proposed system.
• Explore the integration of the proposed algorithm with existing IIoT systems and infrastructures.
• Compare the deep reinforcement learning approach with other machine learning techniques in resource scheduling.
• Analyze the economic impact, including cost-benefit analysis, of implementing the proposed scheduling method.
• Suggest future enhancements and potential research directions for further improving the algorithm.
• Provide case studies or results from pilot programs to demonstrate the practical application and effectiveness of the proposed solution.
• Author can read the following papers to increase the technical strength of the paper:
Future of Internet of Things: Enhancing Cloud-Based IoT Using Artificial Intelligence
Adaptive Ontology-Based IoT Resource Provisioning in Computing Systems
Minor editing of English language required
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe submission topic is scientifically relevant.
Paper structure must be improved and the following remarks should be fully addressed: (i) Introduction needs to be completely reviewed with more context information, a better explanation about the relevance of this research, and should end with the problem statement (in brief) and a final paragraph presenting the document's organization; (ii) section 1.2 must be section 2 and more systematic review of related works and state-of-the art must be provided, with a discussion at the end that shows the valisity of the research problem; (iii) sections 2 (the current section 2, of course) and 3.1 should be put all togheter in same section; (iv) Then a new section 4 named 'Proposal' should be created, starting with the requirements for you proposed algorithm; (v) current section 4 will be renumbered to 5 and will have the name changed to 'Evaluation', ending with a discussion that states clearly the results and contributions, since I missed that; (vi) Conclusions section must include a sum upof the work done, including limitations, which are absent.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for Authorsaccept
Author Response
Dear Reviewer：
On behalf of my co-authors, we appreciate for your positive and constructive comments and suggestions on our manuscript entitled “AoI-aware Resource Scheduling for Industrial IoT with Deep Reinforcement Learning” (electronics-2857301).
Best wishes！
Reviewer 2 Report
Comments and Suggestions for AuthorsThe topic is interesting. Some remarks:
- In the abstratct should not use acronyms. Please review.
-English need to be revised by a native speaker
- Introduction section should provide more context for the research. By the way, the research problem must be stated as it is: a problem, not a solution proposal. Please review.
- Research background needs more detail about current state-of-the-art, proving that the research problem is really a problem.
- The proposal is clearly presented.
- In Evaluation and Simulation I would like to see more detail concernig the used dataset for the purpose. Also, some research on the dataset structure and organziation should be included, since it has the ability to change your parameters. As it is, the section 5 is more a discussion but withouth showing what, when, how you did the simulation. And we would like the authors to include a way to compare the results with a dataset not using your proposal, so we can really understand the evaluation.
- In conclusion the authors do not refer any limitation. Is really possible?
Comments on the Quality of English LanguageToo many problems with English language. I strongly suggest to use a native speaker or a professional proof-reader.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
