A Multi-Strategy Improved Sparrow Search Algorithm for Solving the Node Localization Problem in Heterogeneous Wireless Sensor Networks
Round 1
Reviewer 1 Report
Overall, the work has a sufficient merit to be published. However, please consider the following suggestions/amendments:
The introduction is too short and lack of depth. The background studies, research gaps and research objectives/contributions need to be added.
In Section 2, why did the author choose to improve the SSA as compared with other optimization algorithm? Please justify clearly.
In Table 1, please cite the benchmark functions if there are taken from the other works. Please lable the name of the functions as F1, F2 until F23.
The discussion for the results represented in radar diagram need to be added as well. Please consider adding a paragraph to dicuss the outcomes of Figure 3 thoroughly.
Please resize and improve the resolution of Figure 5 (a) and (b).
In conclusion, authors should mention whether the objectives have been achieved or not. Please another sentence of future work.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
An improved meta-heuristic algorithm is presented and compared with other algorithms to solve the problem of node localization in heterogeneous wireless sensor networks. There are some major changes that needs to be rectified:
- The English of the work needs to be been proofread to remove grammatical mistakes.
- Authors need to bring novelty and originality to their work and need to establish clear superiority of their methodology through comprehensive comparison results with very recent algorithms published in higher Impact Factor journals in case of benchmark functions. Authors must explicitly reference those references of the recent methodologies with which comparisons have been performed to show superiority in the text of the paper.
- The algorithms need to be compared in terms of complexity (time, and resources).
- Sensitivity analysis is required to determine the best settings for the internal parameters of WSN.
- The literature review is required and should be added in the introduction section.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper proposes an improved sparrow search algorithm with the aid of golden sine algorithm (Gold-SA), individual optimal strategies, and Gaussian perturbation to avoid getting stuck in a local minima. Thereafter, the proposed algorithm is applied to real world problem, which is unknown node localization of Heterogeneous Wireless Sensor Network (HWSNs).
The paper is well-written and structured. However, there are several major issues that should be addressed by the authors:
- The authors claim that "there is an urgent need for swarm intelligence of optimization algorithm with fast convergence speed and high optimization seeking accuracy to participate in calculation of the node coordinates in HWSNs." Why is there a specific need for swarm intelligence to solve such a problem? The authors should explain the drawbacks of the current approaches to solve the localization problem and the real need to exploit swarm intelligence in solving it.
- Section 2.1 (Basic SSA) is missing references. Please add specific reference for each paragraph as well as equation.
- In Figure 2, what does random1 and random 2 represent? Is it different from R1 and R2?
- Experimental results section should include the parameters chosen to run the proposed algorithm. For instance, what threshold value did the authors use to decide whether the algorithm is stuck in a local minima? All needed parameters should be supported to ensure the reproducibility of the experiments.
- Provide a reference for the CEC benchmark functions.
- Analysis for the obtained results in Table 2 is missing. Why does the proposed algorithm find the optimal value for Functions F1 to F4? What is the reason that ISSA is the best in all cases in the multimodal functions F12 and F13? More analysis should be provided.
- Why did the authors choose to use Wilcoxon rank-sum test for statistical analysis?
- Section 4: More description about the localization problem in HWSN is required before proceeding to Network model. This problem should be formulated as an optimization problem.
- Why did the authors replace the least-square method with swarm intelligence? A comparison should be conducted to argue the necessity of swarm intelligence instead of using least square method.
- More analysis for Table 4 results is required.
- Minor issues:
a. Line 140: a and be should be in italic form.
b. Line 40: sparrow search algorithms suffers from ==> suffer from
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
All the comments have been successfully addressed by the authors.
Author Response
Special thanks to you for your acknowledgment of our revised manuscript.
Reviewer 3 Report
All major issues have been addressed by the authors. However, the following issues should be addressed:
- Formally define random1 and random2 variables in the manuscript.
- Explain the reason for setting the population size to be 30. Is it possible to empirically obtain the optimum population size prior to algorithm execution?
- Define node localization as an optimzation problem formulated in an equation. Use proper symbols.
Author Response
Please see the attachment.
Author Response File: Author Response.docx