An Optimized, Dynamic, and Efficient Load-Balancing Framework for Resource Management in the Internet of Things (IoT) Environment

Reviewer 1

Reviewer Comment

Response to Reviewer's Comments

1. The abstract is not well written, as the background is too long, which should be briefly introduced or moved to Introduction section, and the detailed work and contribution needs to be further clearly presented and emphasized.

The abstract has been reorganized and as per review suggestions, some basic contents have been moved to introduction section.

2. The introduction should be much improved. For instance, 1) The first sentence of the first three paragraph simultaneously talked about the influence of IoT to people’s 11 daily life; 2) The reviewer has the no idea of the advantages of fog computing, which is not well introduced; 3) “Despite these limitations” is not clear.

The introduction section has been re-organized based on the comments and further the details about IoT, cloud computing and Fog computing has been introduced for better clarity.

A brief discussion about fog computing basics and its advantages in IoT systems have been added in paragraph 3 of introduction.

3. Both the motivation and contributions need to be revised. The motivation is not clear and demonstrate why the authors investigated this work. The logic of contributions should also be improved. For example, “this paper discusses and proposes a solution for three massive problems: load balancing, energy use, and computing cost. Therefore, load balancing, reducing operational costs, and power consumption are essential topics and issues in the IoT.”

The motivation and contributions have been revised to clearly demonstrate it.

4. What is the difference and advantages of cloud computing, fog computing, and mobile edge computing? The above three terms are used in this work, and not clearly distinguished.

The basics of Cloud, edge and Fog computing has been introduced in the section 1.1 (Background) with brief introduction to all these pardigms and these introductions have been appropriately referenced.

5. It is suggested to introduce the following recent works in energy efficiency  [R1]-[R2] and IoT [R4] fields to highlight the state-of-the-art of this paper.

[R1] “Refracting RIS aided hybrid satellite-terrestrial relay networks: Joint beamforming design and optimization,” IEEE Transactions on Aerospace and Electronic Systems, vol. 58, no. 4, pp. 26 3717-3724, Aug. 2022.

[R2] “SLNR-based secure energy efficient beamforming in multibeam satellite systems,” IEEE 28 Transactions on Aerospace and Electronic Systems, early access, Jul. 2022, doi: 10.1109/TAES.2022.3190238. 30

[R3] “Joint beamforming design for secure RIS-assisted IoT networks,” IEEE Internet of Things 31 Journal, vol. 10, no. 2, pp. 1628-1641, Jan. 2023.

[R4] “Supporting IoT with rate-splitting multiple access in satellite and aerial-integrated 33 networks,” IEEE Internet of Things Journal, vol. 8, no. 14, pp. 11123-11134, Jul. 2021.

Thank you for your suggestions. These suggested references have been included in the study.

6. The format of formulas is not acceptable in this paper, the formulas (12)-(15) should be combined into one optimization problem.

The formulas have been merged as a single formula as per suggestion.

7. Where are the detailed steps to solve the problem (12)-(15)?

The formulas (12)-(15) have been merged as per reviewer comments that is now (12). The solution part is the evaluation based on these formulas. The solution / evaluation has been discussed in detail in the section 4 under result and discussion section.

8. The analysis of computational complexity of the proposed method and benchmark schemes should be add.

The solution / evaluation has been discussed in detail in the section 4 under result and discussion section with CloudSim. CloudSim was used to help model and create fog and cloud computing environments. It evaluates the effectiveness of cloud resource management rules for various application and service types in varied load, energy performance, and system size scenarios.

Further the benchmarking / comparison of the output has been done with the similar models in section 4 namely EEFO (Energy-Efficient Opportunistic), DEERA (Dynamic Energy-Efficient Resource Allocation), ELBS (Efficient Load Balancing Security), and DEBTS (Delay Energy Balanced Task Scheduling).

Reviewer 2

Reviewer Comment

Response to Reviewer's Comments

In terms of novel contents this paper does not bring any particular contribution, the only point that merits some attention could be the DEELB method, but there is a lot of confusion on IoT related terms (IoT paradigm, IoT network, IoT applications, etc.) that don’t are in line with the extensive available bibliography. In this respect a more detailed state-of-the-art description of the Cloud-Fog-Edge-IoT continuum, with appropriate references, should be considered.

The basics of Cloud, edge and Fog computing has been introduced in the section 1.1 (Background) with brief introduction to all these pardigms and these introductions have been appropriately referenced.

Another issue is the wrong use of the fog layer in the proposed architecture, as depicted in Figure 1 and later in the solution: fog is an intermediate layer (fog nodes don't use batteries), in proximity with the IoT device. IoT devices should connect to the fog layer, taking advantage of reduced network latency and better throughput.

Figure 1 has been updated accordingly.

The proposed modelling does not fit with the mentioned main points, so the paper should be modified accordingly, including relevant bibliography.

There is a huge description of basics of load balancing, however there is no mention on how this has been applied to the proposed solution.

Latest references have been included throughout the article based on the reviewer suggestions.

The results section describes a comparison of results of different algorithms, but the modelling description is not detailed enough to explain how different parameters such as Packet Loss Ratio, Bandwidth Utilization, Throughput, reliability, Packet Delivery Ratio could be measured.

Given that, the modelling description is not satisfactory, it is difficult to evaluate results. In summary there is only a benchmarking of DEELB algorithm with other similar ones with CloudSim.

The proposed model has been thoroughly explained the section 3. As it as not been practically implmented but a proposed framework only, the simulated results have been evaluated and compared with CloudSim.

The paper is written in fluent English, but the context of the Cloud-Fog-IoT environment is poorly described, with not so relevant contents on load balancing.

There are a lot of acronyms that are used without an explanation (e.g. AIoT), as a general rule, all non-standard abbreviations/acronyms should be written out in full on first use.

The number of references should be improved.

The basics of Cloud, edge and Fog computing has been introduced in the section 1.1 (Background) with brief introduction to all these pardigms and these introductions have been appropriately referenced.

Further, the acronyms have been defined at the first use throughout the manuscript. Further more latest references have been added.