Channel Allocation Algorithm Based on Swarm Intelligence for a Wireless Monitoring Network
Round 1
Reviewer 1 Report
The authors propose a channel allocation algorithm based on the classic bacterial foraging optimization algorithm. Overall, the work is fine. The following comments need to be addressed.
1) In the abstract, clarify the novelty of the proposed algorithm compare to the existing ones.
2) In abstract, state main numerical findings.
3) In introduction, write your main contributions, advantages and disadvantages of the proposed algorithms in bullets.
4) In introduction, write a paragraph to describe the contents of the paper.
5) Define parameters, for example what is QoM in line 249
6) Fix typos, for example "Where" in line 252 should be by small "w"
7) Figures 12, 13, explain in details why the performance remains constant after iterations 100 in figure 12, while it remains constant after iteration 80 as shown in figure 13.
8) Discuss the complexity of the proposed algorithm.
9）The authors need to clarify the novelty compared to existing works. This should be clearly mentioned in the abstract and introduction sections. The main contributions should be written in bullet points in the introductions.
10）The current manuscript does not explain the proposed algorithm well. Many symbols are not well defined. This negatively affects readability.
Author Response
Point 1: In the abstract, clarify the novelty of the proposed algorithm compare to the existing ones.
Response 1: Thanks for the reviewers' comments! In this revised version, we have added novelty and innovation points of the algorithm in the abstract and reflected the performance improvement of the algorithm by comparing it with other algorithms.
Point 2: In abstract, state main numerical findings.
Response 2: Thanks! Referring to your comments, in this revised version, numerical comparisons are added to the abstract to highlight the advantages of the proposed algorithm in this paper by comparing its performance and speed with other algorithms.
Point 3: In introduction, write your main contributions, advantages and disadvantages of the proposed algorithms in bullets.
Response 3: Thank you for your comments. The contributions of this paper are supplemented in the introduction, and the advantages of the algorithm are discussed in three points. (1) a multiradio wireless monitoring network is proposed based on the existing single radio node network model; (2) a discrete BFO channel assignment algorithm for multiradio wireless monitoring nodes is designed for the multiradio wireless monitoring network; (3) the effectiveness of the algorithm is demonstrated by simulation and practical experiments. To highlight the description of the advantages of the algorithm is discussed in three paragraphs in the introduction.
Point 4: In introduction, write a paragraph to describe the contents of the paper.
Response 4: Based on your comments, an overview of the overall content of the paper is given at the end of the introduction, which introduces the structure of the article. And add the following paragraph to describe the content of the paper. In this paper, a form of swarm intelligence, DBFO, is proposed to achieve the optimal channel allocation for multiradio, multichannel wireless monitoring networks. We support this view based on theoretical argument and experiments.
Point 5: Define parameters, for example what is QoM in line 249
Response 5: QoM is short for ‘quality of monitoring’. QoM appears for the first time in the manuscript on line 57. Based on your comments, the full name has been added to the definition of line 249.
Point 6: Fix typos, for example "Where" in line 252 should be by small "w"
Response 6: Based on your valuable comments, I checked the entire paper and corrected a total of five typos.
Point 7: Figures 12, 13, explain in details why the performance remains constant after iterations 100 in figure 12, while it remains constant after iteration 80 as shown in figure 13.
Response 7: In case 1 the node radio is allowed to multiplex channels, there will be multiple radio stations assigned to the same channel, resulting in wasted resources, and as the number of radio stations sharing the channel increases, the amount of information will grow exponentially decreasing, so more iterations are needed to reach the optimal solution; in case 2 the node prohibits different radio stations from multiplexing channels and adds constraints to the radio channel assignment, so the speed is relatively fast.
Point 8: Discuss the complexity of the proposed algorithm.
Response 8: Based on your comments, we have analyzed the complexity of the algorithm in the pseudocode section of Section 4.5. The pseudocode shows that the complexity of the algorithm is related to the rated number of convergence, replication, and migration Nc, Nre, and Ned. The complexity of the algorithm is o(n)=Nc*Nre*Ned. In the 5.1 simulation experiment, Nc=50, Nre=4, and Ned=2, the algorithm needs to perform at most o(n)=400 steps to reach the optimal solution.
Point 9: The authors need to clarify the novelty compared to existing works. This should be clearly mentioned in the abstract and introduction sections. The main contributions should be written in bullet points in the introductions.
Response 9: Based on your comments, the abstract section and the introduction have been added accordingly.
Point 10: The current manuscript does not explain the proposed algorithm well. Many symbols are not well defined. This negatively affects readability.
Response 10: Based on your comments, a new table has been added to the paper to introduce the 21 terms and symbols in the algorithm and explain their meanings before introducing the algorithm in Section 4. The character parsing following each formula is also checked, and the missing character parsing is added to make the algorithm more readable.
All the modified contents are marked with red color in the revision. Please check them, thanks!
Reviewer 2 Report
Although the paper presents an interesting solution to dynamic channel allocation there is not a clear scientific contribution or novelty.
Formally the structure and description of the methods require to be improved. The use of seldom terms and seldom definition of symbols makes the paper very difficult to analyse. For instance, several symbols are used without a formal definition or beforehand description. For instance, in the algorithm, the general parameters of swarm optimization are mixed with the specifically targeted variables which makes the algorithm difficult to analyse. I would recommend use a custom representation of symbols adapted to the specific research problem of channel allocation.
The wireless scenario considered for simulations and validation of the algorithm is not clearly described. This issue makes the results difficult to evaluate and compare to other methods and state-of-the-art.
Author Response
Point 1: Formally the structure and description of the methods require to be improved. The use of seldom terms and seldom definition of symbols makes the paper very difficult to analyse. For instance, several symbols are used without a formal definition or beforehand description. For instance, in the algorithm, the general parameters of swarm optimization are mixed with the specifically targeted variables which makes the algorithm difficult to analyse. I would recommend use a custom representation of symbols adapted to the specific research problem of channel allocation.
Response 1: Thanks for the reviewers' comments. In this revised version, in order to make it easier to read and understand the population intelligence-based "discrete bacterial foraging optimization" channel allocation algorithm proposed in the paper, we have added a new table to introduce the 21 terms and symbols in the algorithm and explain their meanings before introducing the algorithm in Section 4. The parsing of the characters following each formula is also checked, and the missing ones are added to make the algorithm more readable.
Point 2: The wireless scenario considered for simulations and validation of the algorithm is not clearly described. This issue makes the results difficult to evaluate and compare to other methods and state-of-the-art.
Response 2: According to your comments, we have supplemented the hardware environment of the experiment in the simulation experiment session, providing data on CPU, GPU, video memory, random access memory, computer system, and the experimental platform OPNET. The model established by OPNET is described at the network level, node level, and process level, and the interface cards and transmission rates of the nodes are introduced.
All the modified contents are marked with red color in the revision. Please check them, thanks!
Reviewer 3 Report
1. The duplication of plagiarism in Turnitin is 35% which is a little higher; the authors need to paraphrase the manuscript to reduce this rate is less than 20%.
2. Discuss the limitations of existing works that motivated the current research, which may add to the abstract with the quantitative experimental results.
3. The technical contribution of this work should be added to the end of the introduction section. Maybe discuss in more depth the current work's main contributions, the research's significance, and the potential outcomes.
4. Section 5 discussion of the Experimental results should be further elaborated to show how they could use for real applications. Recommend authors compare the results of the proposed approach with more algorithms in terms of, e.g., executed complexity computation, measures of accuracy, and performance, with various previous works in the literature.
Author Response
Point 1: The duplication of plagiarism in Turnitin is 35% which is a little higher; the authors need to paraphrase the manuscript to reduce this rate is less than 20%.
Response 1: Thanks for the reviewers' comments. In this revised version, we have compared the check report, paraphrased the manuscript, and checked the full text. The parts that were paraphrased in the revised version are marked with red color.
Point 2: Discuss the limitations of existing works that motivated the current research, which may add to the abstract with the quantitative experimental results.
Response 2: According to your comments, the comparison between the proposed algorithm in this paper and existing works is added to the abstract to reflect the novelty of the proposed algorithm in this paper, and the enhancement of the proposed algorithm in this paper is also reflected through the comparison with other algorithms.
Point 3: The technical contribution of this work should be added to the end of the introduction section. Maybe discuss in more depth the current work's main contributions, the research's significance, and the potential outcomes.
Response 3: According to your comments, the technical contributions of this work are presented in the introduction section in three points (1) a multiradio wireless monitoring network is proposed based on the existing singleradio node network model; (2) a discrete BFO channel assignment algorithm for multiradio wireless monitoring nodes is designed for the multiradio wireless monitoring network; (3) the effectiveness of the algorithm is demonstrated by simulation and practical experiments. In order to highlight the description of the advantages of the algorithm, it is discussed in three paragraphs in the introduction.
Point 4: Section 5 discussion of the Experimental results should be further elaborated to show how they could use for real applications. Recommend authors compare the results of the proposed approach with more algorithms in terms of, e.g., executed complexity computation, measures of accuracy, and performance, with various previous works in the literature.
Response 4: In Section 5, in comparison with other algorithms, the novelty of the DBFO proposed in this paper and the other three algorithms LP, MAB, and GT are further elaborated, and it is proved by experimental results that the DBFO proposed in this paper is better than them in terms of monitoring quality and convergence speed, with faster computation speed as well as higher accuracy. And through the longitudinal analysis, it is proved that the DBFO proposed in this paper has higher robustness.
All the modified contents are marked with red color in the revision. Please check them, thanks!
Round 2
Reviewer 1 Report
I accept the work for publication in its current form. I do not have any further comments.
Reviewer 3 Report
Figures 2 and 3 may lower resolution.
They should be higher resolution.
