Distributed Data-Driven Learning-Based Optimal Dynamic Resource Allocation for Multi-RIS-Assisted Multi-User Ad-Hoc Network
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
This paper addresses the decentralized dynamic resource allocation optimization problem and introduces an innovative two-loop online distributed Actor-Critical Reinforcement Learning algorithm designed to optimize mobile ad-hoc networks (MANETs) with support for multiple RIS (Reconfigurable Intelligent Surfaces) in finite time, even under uncertain and time-varying wireless channel conditions.
In addition to all these aspects, the rigorously formulated problem constraints include an interference analysis, which guarantees a realistic simulation of the proposed algorithms.
Some parts could be added or clarified
· Page 2, l. 66-67: “enhance energy efficiency“ - Can it be quantified or at least estimated?
· The ranges of the indices $i$, $m$, $j$ should be given in the constraints (6).
· “The optimization problem (P) is characterized as a mixed-integer programming problem“ - Which variables are restricted to the integer domain?
· It would be appropriate to add the time complexity of the studied problem and the proposed algorithms.
· The question is whether the values of 10 transmitters and 10 receivers considered in the simulation are sufficient for general conclusions.
· The symbol $P_{max}$ in Eq. (3) is not characterized.
· Although the abbreviation MIMO is well known, it should be noted that it stands for Multiple-Input- Multiple-Output (page 3, line 93)
Format:
Equation (15) does not satisfy the format of mathematical equations (the same in the free text t=1, t=T above it). Similarly, "each player i" - $i$. on p. 9, line 336, and in several cases in Algorithm 1.
Author Response
Dear reviewer,
Thanks a lot for your constructive comments and suggestions. Please see the attachment for our responses. Thanks
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors
1). The Introduction Section is too large. The authors reduce the introduction section while inserting one more section called “Related Studies and the Current Contribution”, where they can describe the related works and the contribution of the current paper in relation to those works.
2). Section 2.2 is a bit confusing. For example, while eq. (4) is complex, its description is not adequate. Is the quantity calculated by this equation considered as a weighting factor? If the answer is yes, I would expect to see the corresponding standard deviation in the numerator. In addition, I would expect to see a more detailed analysis in that section.
3). In Section 3.1.2. the sets M and J must be rigorously defined so that the reader can easily realize the resulting Cartesian Product.
4). The paper lacks a rigorous statistical analysis of the results, where inference statistics (using the resulting p-values) should verify the robustness of the proposed method.
Comments on the Quality of English Language
Moderate English corrections.
Author Response
Dear reviewer,
Thanks a lot for your constructive comments and suggestions. Please see the attachment for our responses. Thanks
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for Authors
The authors answered my comments.
Comments on the Quality of English Language
See above