Passive Positioning and Adjustment Strategy for UAV Swarm Considering Formation Electromagnetic Compatibility

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article addresses the “passive” positioning and formation correction method of unmanned aerial vehicle swarms within the framework of electromagnetic compatibility. Although the subject is theoretically remarkable, the study contains a number of fundamental problems and has not reached a satisfactory maturity in terms of application.

My comments on this study are as follows:

1. The article remains completely theoretical. Simulations were performed under ideal conditions, with simple geometries and in a noiseless environment. Real-world conditions were not taken into account. There is no hardware verification, field test or experimental data.

2. Although passive positioning is mentioned in the article, it is stated that some UAVs should actively send electromagnetic waves. This contradicts the concept of passivity. The system actually has a selectively active structure and is not a “fully passive” solution. This situation stands out as a fundamental problem in the structure of the publication from the title onwards.

3. The simulation results obtained are very smooth (noiseless) and ideal. In reality, such a situation is very difficult to occur. Neither filtering nor error tolerance mechanisms are defined. How the system behaves in real life with signal distortions or measurement errors has not been evaluated. Electromagnetic propagation is modeled only as a function of energy that decreases with distance. Complex factors such as reflections, directionality, and environmental conditions in real EM environments are ignored. Modeling parameters are too simple and far from reality to be left for future studies.

4. The method has not been compared with existing passive localization techniques in the field (AoA, TDOA, SLAM). The literature review is quite superficial and no evaluation is provided on whether the current method is competitive.

5. The position of each UAV is calculated centrally. No concepts such as local decision making, adaptability in case of error, or swarm intelligence are included. This distances the study from a real swarm behavior. The method has only been tested for a circular formation of 10 UAVs. No evaluation is provided on how the system will perform in different geometries, numbers of UAVs, or dynamic mission scenarios.

6. The method performs position correction via grid-based proximity measurement and angle difference at each iteration. However, the effects of these operations in terms of computational time, energy consumption and real-time operability have never been discussed.

Although the article has an interesting motivation, it has serious deficiencies in terms of methodology, concept and implementation. I believe that the study is not suitable for publication in its current form due to reasons such as unclear concepts, lack of experimental validation, ignoring real-world scenarios and not comparing with the literature. It does not seem possible for the authors to bring this article to a sufficient level with major revisions: the literature review should be expanded, realistic parameters should be added to the method and expanded, the findings should be expanded, and benchmarking should be done with similar/competitor studies. For these reasons, I recommend that the article be rejected.

Author Response

Comments 1: [The article remains completely theoretical. Simulations were performed under ideal conditions, with simple geometries and in a noiseless environment. Real-world conditions were not taken into account. There is no hardware verification, field test or experimental data.]

Response 1: [We sincerely thank you for your constructive comments. Our paper combines theoretical research with practical considerations. In the latest version of the article, we have added some references in Section 2.2 (first paragraph) on potential electromagnetic interference caused by mutual electromagnetic radiation. In the literature review section, we contextualized this study within the existing AOA technological background, thereby establishing the experimental foundation and practical feasibility of the work. The revised sections are highlighted in red in the updated manuscript.]

 

Comments 2: [Although passive positioning is mentioned in the article, it is stated that some UAVs should actively send electromagnetic waves. This contradicts the concept of passivity. The system actually has a selectively active structure and is not a “fully passive” solution. This situation stands out as a fundamental problem in the structure of the publication from the title onwards.]

Response 2: [Thank you for pointing this out. The term “passive positioning technology” in the text refers to the drone formation minimizing electromagnetic wave interactions with the external environment. Active positioning methods, such as satellite positioning technology, should be avoided. However, electromagnetic communication within the drone formation is necessary for the localization techniques proposed in this study.]

 

Comments 3: [The simulation results obtained are very smooth (noiseless) and ideal. In reality, such a situation is very difficult to occur. Neither filtering nor error tolerance mechanisms are defined. How the system behaves in real life with signal distortions or measurement errors has not been evaluated. Electromagnetic propagation is modeled only as a function of energy that decreases with distance. Complex factors such as reflections, directionality, and environmental conditions in real EM environments are ignored. Modeling parameters are too simple and far from reality to be left for future studies.]

Response 3: [The angle-aware iterative simulation in this paper is based on establishing an idealized mathematical model. As you rightly noted, real-world scenarios may involve various challenges such as measurement errors and signal distortion, but these aspects fall outside the scope of this study. This work aims to provide a passive localization and position adjustment strategy for drone formations, offering a methodological framework for real-world deployment.]

 

Comments 4: [The method has not been compared with existing passive localization techniques in the field (AoA, TDOA, SLAM). The literature review is quite superficial and no evaluation is provided on whether the current method is competitive.]

Response 4: [We have revised the fifth paragraph of the literature review section to compare this study with existing passive localization technologies (AOA, TDOA, SLAM). The text now explicitly elaborates on the innovative aspects and competitive advantages of our approach relative to other methods. The modified content is highlighted in red in the latest version of the manuscript.]

 

Comments 5: [The position of each UAV is calculated centrally. No concepts such as local decision making, adaptability in case of error, or swarm intelligence are included. This distances the study from a real swarm behavior. The method has only been tested for a circular formation of 10 UAVs. No evaluation is provided on how the system will perform in different geometries, numbers of UAVs, or dynamic mission scenarios.]

Response 5: [The paper does not explicitly specify whether the position of each drone is calculated centrally or distributed, as this depends on the location of the processor. If each UAV is equipped with onboard computing capabilities, every drone can independently adjust its position upon receiving angular signals. If only the central UAV has computing capacity, other drones must transmit angular information to the central unit for processing. We have revised Section 3.1 to elaborate on this process in greater detail. Additionally, we expanded the discussion in Section 3.3 to address formations with varying geometric configurations. The modified sections are highlighted in red in the latest manuscript.]

 

Comments 6: [The method performs position correction via grid-based proximity measurement and angle difference at each iteration. However, the effects of these operations in terms of computational time, energy consumption and real-time operability have never been discussed.]

Response 6: [Establishing a mathematical model is an abstract process and cannot account for every real-world variable. This study abstracts away from factors unrelated to the core strategy, such as adjustment time per iteration and energy consumption, to focus on the mathematical treatment of formation reshaping and electromagnetic signal transmission.]

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This well-written paper presents the determination of UAV location in terms of other UAVs'
several positions through electromagnetic interference between them. The calculated angles between their mixed ECM signals offer the localization efficiency with condition that in the end their radius and angular deviations are being converged. The novelty is adequate as well as the paper contribution with minor issues might  be solved by the authors in the revised version.

1) the introduction might include more related work within the passive and the active localization, its applications on drones /UAVs in general, communications between drones/UAVs etc.

1) The equation of electromagnetic signal energy must be named explicitly. Name it correctly according to electromagnetic theory not only with "e".

2) No adequate references for a journal paper. The authors might add additional references with proper citations in order 
to be the paper completed. At least 36-40 references might exist. These might possibly refer to:
i) electromagnetic theory which determines energy signal, the interference between signals, the law of inverted square by which  the ECM signal energy is reduced according to the distance.
3) the equations might be written with smaller font size.
4) the table named "Optimizing UAV Formation Alignment Using Passive Localization Method" might be written as a completed latex  algorithm with symbols and functions as "start...end", "if...then..end", "for..end", "while..end". 
5) Additional matrices in appendix might be added with not only qualitative but quantitative capacity of detected signal energy of the interfered  ECM signals of drones.
6) the formation of the tables might be corrected in latex format.

Author Response

Comments 1: [The introduction might include more related work within the passive and the active localization, its applications on drones /UAVs in general, communications between drones/UAVs etc.The equation of electromagnetic signal energy must be named explicitly. Name it correctly according to electromagnetic theory not only with "e".]

Response 1: [We sincerely thank you for your constructive suggestions. The revised introduction (first paragraph and second paragraph) now includes a comparison between passive and active localization technologies, as well as relevant content on communication mechanisms. In Section 2.2 (first paragraph), we have added explicit descriptions of the specific physical meaning of "e." The modified sections are highlighted in red in the latest version of the manuscript.]

 

Comments 2: [No adequate references for a journal paper. The authors might add additional references with proper citations in order to be the paper completed. At least 36-40 references might exist. These might possibly refer to: electromagnetic theory which determines energy signal, the interference between signals, the law of inverted square by which  the ECM signal energy is reduced according to the distance.]

Response 2: [Thank you for pointing this out. We have incorporated sufficient references to electromagnetic theory for energy signals, inter-signal interference, and the inverse-square law governing ECM signal energy attenuation with distance. The total number of references now reaches 38.]

 

Comments 3: [the equations might be written with smaller font size.]

Response 3: [We agree with using smaller font sizes for equations. All equations in the text have been adjusted to an appropriate size.]

 

Comments 4: [the table named "Optimizing UAV Formation Alignment Using Passive Localization Method" might be written as a completed latex  algorithm with symbols and functions as "start...end", "if...then..end", "for..end", "while..end". ]

Response 4: [I am more proficient in using Microsoft Word than LaTeX for editing and have adjusted the formatting to be more formal and visually polished.]

 

Comments 5: [Additional matrices in appendix might be added with not only qualitative but quantitative capacity of detected signal energy of the interfered  ECM signals of drones.]

Response 5: [In Appendix B of the manuscript, we have added quantitative descriptions of ECM signal energy to better support the qualitative conclusions. The revised sections are highlighted in red in the latest version.]

 

Comments 6: [ the formation of the tables might be corrected in latex format.]

Response 6: [Thank you for your suggestions. As noted, the document formatting has been refined using Word to enhance its professional appearance.]

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Formulas 8-11 etc. are in a different font - please fix it!

Author Response

Comments 1: [Formulas 8-11 etc. are in a different font - please fix it!]

Response 1: [Thank you for your recognition of our work. We have standardized the formatting of all formulas.]

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Comment 1: The article remains completely theoretical. Simulations were performed under ideal conditions, with simple geometries and in a noiseless environment. Real-world conditions were not taken into account. There is no hardware verification, field test or experimental data.

Authors’ Response Evaluation: The authors cite newly added references and a theoretical discussion of EMC, but this does not address the original concern. The simulations are still entirely idealized, without any empirical validation or noise modeling.

What Must Be Done:

• Include at least one simulation scenario incorporating realistic signal noise, measurement errors, or reflection artifacts.

• Discuss a plan for hardware validation, or provide a basic feasibility analysis (e.g., sensor requirements, response times, onboard compute load).

• Add a dedicated section discussing real-world deployment challenges, even if preliminary.

Comment 2: Although passive positioning is mentioned in the article, it is stated that some UAVs should actively send electromagnetic waves. This contradicts the concept of passivity. The system actually has a selectively active structure and is not a “fully passive” solution. This situation stands out as a fundamental problem in the structure of the publication from the title onwards

Authors’ Response Evaluation: The authors attempt to clarify that the system avoids external signal emission but still uses inter-UAV communication. However, this contradicts the use of "passive positioning" in its strictest sense.

What Must Be Done:

• Clearly redefine the system as “semi-passive” or “internally passive, externally silent,” both in the title and abstract.

• Explicitly distinguish the system from fully passive methods (e.g., pure AOA with no emission).

• Reframe claims about stealth and passivity more cautiously and precisely.

Comment 3: The simulation results obtained are very smooth (noiseless) and ideal. In reality, such a situation is very difficult to occur. Neither filtering nor error tolerance mechanisms are defined. How the system behaves in real life with signal distortions or measurement errors has not been evaluated. Electromagnetic propagation is modeled only as a function of energy that decreases with distance. Complex factors such as reflections, directionality, and environmental conditions in real EM environments are ignored. Modeling parameters are too simple and far from reality to be left for future studies.

Authors’ Response Evaluation:The authors acknowledge the idealized modeling but claim that realistic conditions are out of scope. This is insufficient, as the credibility of the method relies heavily on real-world feasibility.

What Must Be Done:

• Introduce simplified noise models into the simulations (e.g., Gaussian noise on angle measurements).

• Discuss the limitations of using inverse-square energy models alone without considering multipath, attenuation, or directionality.

• Propose how filtering mechanisms (e.g., Kalman filter) could be incorporated into future work.

Comment 4: The method has not been compared with existing passive localization techniques in the field (AoA, TDOA, SLAM). The literature review is quite superficial and no evaluation is provided on whether the current method is competitive.

Authors’ Response Evaluation:The literature review has been extended, and existing methods are briefly mentioned. However, the comparison remains qualitative and superficial.

What Must Be Done:

• Include a summary table comparing the proposed method with AOA, TDOA, and SLAM, based on criteria such as energy consumption, complexity, stealth, accuracy, and real-time feasibility.

• Discuss performance trade-offs and scenarios where the proposed method would be superior or inferior.

Comment 5: The position of each UAV is calculated centrally. No concepts such as local decision making, adaptability in case of error, or swarm intelligence are included. This distances the study from a real swarm behavior. The method has only been tested for a circular formation of 10 UAVs. No evaluation is provided on how the system will perform in different geometries, numbers of UAVs, or dynamic mission scenarios.

Authors’ Response Evaluation:The authors now describe both centralized and distributed modes of computation but do not model or simulate these differences.

What Must Be Done:

• Provide simulation results (or pseudocode flowcharts) for both centralized and decentralized versions.

• Discuss how the proposed approach aligns or deviates from swarm intelligence principles (e.g., local decision-making, fault tolerance).

• Address what happens when one UAV fails or provides faulty measurements.

Comment 6: The method performs position correction via grid-based proximity measurement and angle difference at each iteration. However, the effects of these operations in terms of computational time, energy consumption and real-time operability have never been discussed.

Authors’ Response Evaluation: This comment was largely dismissed as "not part of the mathematical model," which is unsatisfactory for a practical system aimed at UAVs.

What Must Be Done:

• Provide estimates of computational complexity (e.g., time per iteration, algorithmic complexity class).

• Discuss energy consumption implications, especially if drones are low-power platforms.

• Evaluate how many iterations are typically required and whether this can be done in real-time.

Author Response

Comment 1: The article remains completely theoretical. Simulations were performed under ideal conditions, with simple geometries and in a noiseless environment. Real-world conditions were not taken into account. There is no hardware verification, field test or experimental data.

Response 1: We sincerely thank you for your constructive comments. This paper has specifically added Section 4 to discuss practical deployment challenges, addressing hardware verification plans, basic response times, and load calculation schemes. Simulation scenarios with measurement errors were designed and analyzed. The modified sections have been highlighted in yellow in the latest version of the manuscript.

Comment 2: Although passive positioning is mentioned in the article, it is stated that some UAVs should actively send electromagnetic waves. This contradicts the concept of passivity. The system actually has a selectively active structure and is not a “fully passive” solution. This situation stands out as a fundamental problem in the structure of the publication from the title onwards

Response 2: The paper has revised the abstract, explicitly defining the system as “internally active, externally silent.” This revision resolves the terminology contradiction while preserving the research's innovative aspect—balancing necessary internal communication within the formation and external silence. The newly added comparative explanation in the review section guides readers to accurately understand the technical boundaries.

Comment 3: The simulation results obtained are very smooth (noiseless) and ideal. In reality, such a situation is very difficult to occur. Neither filtering nor error tolerance mechanisms are defined. How the system behaves in real life with signal distortions or measurement errors has not been evaluated. Electromagnetic propagation is modeled only as a function of energy that decreases with distance. Complex factors such as reflections, directionality, and environmental conditions in real EM environments are ignored. Modeling parameters are too simple and far from reality to be left for future studies.

Response 3: This paper introduces a simplified noise model in Section 4 simulations, transforming the limitations of the inverse-square energy model into the final measurement error analysis. Table 5 in Section 4 includes filtering mechanisms as future work.

Comment 4: The method has not been compared with existing passive localization techniques in the field (AoA, TDOA, SLAM). The literature review is quite superficial and no evaluation is provided on whether the current method is competitive.

Response 4: In the literature review section, the paper creates a summary table comparing the proposed method with AOA, TDOA, and SLAM based on multiple criteria. The modified sections have been highlighted in yellow in the latest version of the manuscript.

Comment 5: The position of each UAV is calculated centrally. No concepts such as local decision making, adaptability in case of error, or swarm intelligence are included. This distances the study from a real swarm behavior. The method has only been tested for a circular formation of 10 UAVs. No evaluation is provided on how the system will perform in different geometries, numbers of UAVs, or dynamic mission scenarios.

Response 5: We have revised Section 3.1 to provide flowcharts for both centralized and decentralized computation. The flowcharts for centralized and decentralized computation are included in Appendix B3. In the simulation component of Section 4, we address the scenario where measurement errors occur on a single drone.

Comment 6: The method performs position correction via grid-based proximity measurement and angle difference at each iteration. However, the effects of these operations in terms of computational time, energy consumption and real-time operability have never been discussed.

Response 6: We have revised Section 4 to discuss the impact on energy consumption and evaluated the number of iterations required. When measurement errors are present, the number of iterations increases as expected, with our test results showing convergence achieved at 59 iterations.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

Deficiencies in Author Responses and What Needs to Be Done
Although the authors have significantly improved the article compared to its previous version, their responses are still insufficient on some basic issues. Noise modeling is limited to a simple scenario, filtering mechanisms are only stated as “left to future studies”, which is a negative situation for a study carried out only in a simulation environment. Real electromagnetic propagation conditions such as reflection, multipath, directionality and their effects on the model are not discussed. Elements such as expected adaptive behaviors, local decision making, and fault tolerance for swarm intelligence are not included in the system. In addition, no concrete evaluations have been made in terms of computation time, energy consumption and real-time, only platform suggestions have been presented. This situation causes the claims that the system is ready for real-world applications to remain weak.

What Needs to Be Done:
- Noise modeling should be made more comprehensive. Different levels of angle deviation, signal interference scenarios should be added.
- Discuss the limitations of the "1/r²" model clearly; specify the effects of real EM propagation conditions.
- Discuss to what extent your system complies with Swarm Intelligence principles and where it falls short. Limitations should definitely be mentioned in the text.

- Assess how the system will be affected in the event of a malfunctioning UAV and how you can handle this. Limitations should definitely be mentioned in the text.

- After explaining the hardware context in which the simulation is run, an average iteration time should be written. The suitability of the total system response in terms of real time should be analyzed.

- The hardware context in which the simulation is run should be explained and the computational load and/or energy consumption should be supported with numerical data.

- All assumptions and limitations stated in previous revision comments should be addressed in the results and discussion sections.

Author Response

We sincerely thank you for your constructive comments. The following modifications have been made as required:

Comment 1: Noise modeling should be made more comprehensive. Different levels of angle deviation, signal interference scenarios should be added.

Response 1: To expand the modeling scenarios for interference conditions, this study expanded simulations from 1% to 2% interference levels. The results demonstrate that even after multiple iterations, the formation cannot be adjusted to the desired configuration, indicating the algorithm's inapplicability under strong external interference. These modifications have been highlighted in yellow in the final paragraph of Section 4.

Comment 2: Discuss the limitations of the "1/r²" model clearly; specify the effects of real EM propagation conditions.

Response 2: Electromagnetic wave reflections at medium interfaces (e.g., buildings, ground surfaces) increase signal path lengths, causing energy attenuation that deviates from ideal models. When signals arrive at receivers through multiple paths (direct, reflected, and scattered), path differences create phase interference and multipath effects. In our formation environment, these effects are negligible, with detailed modifications presented in Section 2.2.

Comment 3: Discuss to what extent your system complies with Swarm Intelligence principles and where it falls short. Limitations should definitely be mentioned in the text.

Response 3: The final paragraph of Section 4 clarifies that this algorithm is not suitable under strong external interference conditions. An additional note in Figure B3 specifies that the model becomes inapplicable when any formation drone is missing.

Comment 4: Assess how the system will be affected in the event of a malfunctioning UAV and how you can handle this. Limitations should definitely be mentioned in the text.

Response 4: As evident from Section 2 and Figure B3, there exists strong interdependence among the drones within the formation. When a drone malfunctions, the system can no longer maintain the original positioning strategy and must immediately switch to an alternative approach; otherwise, the formation will significantly deviate from its intended relative positioning.

Comment 5: After explaining the hardware context in which the simulation is run, an average iteration time should be written. The suitability of the total system response in terms of real time should be analyzed.

Response 5: This paper includes Table A4 to present the simulation iteration time and computational load data.

Comment 6: The hardware context in which the simulation is run should be explained and the computational load and/or energy consumption should be supported with numerical data.

Response 6: The hardware environment specifications are provided in the notes of Table A4, where the data demonstrates the computational load conditions.

Comment 7: All assumptions and limitations stated in previous revision comments should be addressed in the results and discussion sections.

Response 7: We have implemented corresponding revisions in the Conclusion section.

Author Response File: Author Response.pdf