Two-Dimensional Real-Time Direction-Finding System for UAV RF Signals Based on Uniform Circular Array and MUSIC-WAA

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents a two-dimensional real-time direction-finding system based on radio frequency signals for unmanned aerial vehicles. It employs a six-element uniform circular array and an improved MUSIC algorithm combined with a weighted average algorithm, optimizing the computational efficiency of direction angle estimation and achieving real-time measurement of azimuth and elevation angles. Experimental results show that within a range of 30 to 200 meters in distance and 20 to 90 meters in height, the system has an average azimuth angle error of 7.0° and an elevation angle error of 7.7°, demonstrating effective dynamic tracking capabilities. The system features omnidirectional coverage, dual-band operation, and high cost-effectiveness, making it suitable for low-altitude safety monitoring.

My comments are as follows:

1. The literature review is comprehensive, covering recent studies on RF-based UAV direction findings, but the connection between the studies lacks sufficient cohesion, and the relevant expressions should be refined.
2. In the performance comparison experiment between the WAA algorithm and other heuristic algorithms, there is only one set of experimental data. To better support the experimental conclusions, multiple sets of comparative experiments should be added.
3. The structure of this paper is complete, but there are some typographical errors, such as in the titles of Section 4 and Section 5, which need to be proofread and adjusted.
4. Some pictures lack clarity, such as Figure 3. Check the resolution of the picture and improve the clarity of the picture.
5. In the theory part of the MUSIC algorithm, the derivation of parameters in formula (9) is lacking. Therefore, the derivation of relevant formulas should be supplemented to improve the theoretical section.
6. The design description of the synchronous clock and trigger module is relatively simple, and more details should be added to enhance the explanation of the hardware synchronization calibration principle.
7. The analysis of the elevation angle sensitivity for UCA arrays is insufficient, and more information needs to be shared.
8. The experimental results of the paper show that the system has direction-finding errors when dealing with signal interference. It is suggested to discuss the challenges that the system deployment needs to address in the conclusion section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors developed a direction-finding system method for identifying UAVs, using the MUSIC algorithm as software and software defined radio platforms as hardware. The proposed system is validated experimentally using a DJI drone as target. The authors should take into account the following comments and suggestions in order to improve the overall quality of the paper.

Comments regarding technical aspects:

• A reference for the WAA algorithm should be included;
• In figure 4, it should be specified that the cables connecting the antennas to the SDRs are of equal lengths;
• References for the commercial hardware that was used should be included (the 10 MHz square wave generator, the FPGA board);
• More details, including some block diagrams, should be given regarding the Synchronous Trigger Module and the Host Controller; It should be mentioned if GNU Radio python scripts or GNU Radio Companion graphical environment was used;
• A capture of the GUI generated by the GUI module from the host controller should be included;
• Some comments should be given regarding the bandwidth of only 2MHz set for the HackRFOne SDRs, which is smaller than the bandwidth of the signal generated by the drone (10 MHz). Doesn’t this affect the performance of the detection?
• It should be mentioned if the calibration process is necessary after each reboot of the system or not;
• More details regarding the characteristics of the signal generated by the DJI AIR 2S target drone should be given (protocol, possibility of adjusting the bandwidth, aso);
• A 3D representation would be useful for the two traces included in Fig. 9;
• More details regarding the computational aspects mentioned in rows 400-402 should be given;
• Some comments would be useful regarding the significant difference between DF and flight log from figure 12a, between 800 and 900 seconds;

Comments regarding editing aspects:

• Some abbreviations should be explained the first time when they are used (for example UAV in the abstract);
• Sections shouldn’t start with a figure (see 3.1);
• Sections shouldn’t start directly with a subsection (see Section 5);

Comments regarding grammar/typos:

• Considering… instead of Based on… (row 106);
• 5 dB instead of 5 (row 242);
• …are shown… instead of …is shown… (row 315);
• The description of Figure 10 doesn’t reflect its contents.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper proposes a UAV direction-finding system combining UCA with a hybrid MUSIC-WAA algorithm, achieving real-time azimuth/elevation estimation with significantly reduced computational complexity. While the work is promising, the following critical issues require attention:

1、The current title is overly generic. It is recommended to explicitly mention core techniques (e.g., UCA, MUSIC-WAA) and key innovations to enhance specificity.

2、Abbreviations should be standardized by providing full terms at first mention only. For example, “MAD” is defined twice in Sections 5.3 and 5.4.

3、In the Introduction, the extended discussion on UAV applications (agriculture, logistics, etc.) in paragraphs 1-2 is redundant. This section can be condensed to 1-2 sentences, focusing on the motivation for direction-finding systems.

4、The “Related Work” section can be integrated into the Introduction to improve structural coherence. The analysis of prior works focuses on “array configurations, algorithmic enhancements, and hardware implementations”—aspects tangential to this study. To strengthen relevance, the authors can critically review the literature addressing “elevation angle estimation, direction-finding range, and computational efficiency,” explicitly identifying shortcomings in these domains.

5、The transition condition between the exploration and exploitation phases in Eq. (16) lacks an explicit definition.

6、In Figure 2, parameter k₁ is referenced as “f(iter) ” in Eq(16); please unify the notation. Additionally, decision boxes lack “Yes/No” label, and parameters k₂ and k₃ remain unexplained.

7、Variable names in Eq (12)-(15) and (17)-(19) are overly verbose and should be simplified.

8、Sections 4 and 5 share identical titles (“DF System Description”); these should be revised for distinction.

9、Consider presenting the parameter settings from Sections 5.1-5.2 in a table format, enhancing readability while preserving technical completeness.

10、Replace “floating point” with “hovering position” in 5.3.

11、Figure 7 can be removed. The hover position annotations in Figure 10 can be integrated into Figure 9 to eliminate redundancy while preserving informational clarity, allowing either Figure 9 or 10 to be retained. Table 1 is expendable, while Table 2 duplicates the data in Figure 11. Retain Figure 11 or Table 2 for visual clarity.

12、Comparative simulations with existing methods are recommended to validate the accuracy and performance improvements.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have properly addressed all my previous comments and suggestions.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The author has provided clear responses to the initial review comments one by one and made corresponding revisions to the article. Here are specific evaluations and suggestions for the current version:

1、The title should be corrected to adhere to proper capitalization rules: “Two-Dimensional Real-Time Direction-Finding System for UAV RF Signal Based on Uniform Circular Array and MUSIC-WAA Algorithm”.

2、The literature review in the Introduction lacks a structured technical narrative and hierarchical organization. For instance, when discussing ULA-based algorithms, the authors failed to frame their analysis around key advantages (e.g., computational efficiency) to establish logical connections between references [16]–[19]. Critical comparisons of performance differences among these studies were omitted. Additionally, while criticizing [19] for “failing to address full 3D airspace monitoring”, the proposed method also does not achieve 3D monitoring. The significant angular error (18.65°) reported in [20] lacks analysis of its underlying causes. It is recommended to reorganize the literature review by technical categories (ULA vs. UCA) and/or limitations (coverage, resolution, dimensionality).

3、The experimental results report azimuth and elevation errors of 7.0° and 7.7°, respectively, but the ULA-based methods in [16]–[19] can achieve higher estimation accuracies (e.g., [17] achieves a lower azimuth error of 1.39°). More convincing explanations are needed to prove the superiority of the algorithm.

4、The claim of “reducing computational complexity by more than 99.9% compared with spectral traversal” and the claim of “a 5-element UCA paired with MUSIC requires 3.24 million spectral calculations for 0.1° resolution” is misleading, as the comparison assumes different resolutions (7° and 7.7° in this work vs. 0.1° in traditional MUSIC). The authors must either compare complexity at the same resolution (0.1°) or explicitly justify the trade-off between resolution loss and real-time performance.

5、Redundant phrasing persists, such as the repetitive structure in Section 2.2 describing parameter 𝑘2: "When 𝑘2 = 1, Equation (21) is employed..." Rewrite this as: "The candidate solution updates are controlled by 𝑘2 ∈ {1,2,3}, applying Equation (21)/(22)/(23) respectively." Similar issues in other sections require further simplification.

6、Inconsistent Equation Referencing: The manuscript inconsistently formats equation citations, mixing "Equation (x)," "equation (x)," and "Eq. (x)" (e.g., "Eq. (x)" in Figure 2). Consistently use "Equation (x)" throughout to align with journal guidelines.

7、Please indicate the units in Equation 31 and Table 1. The typo "SLB" on line 186 has been corrected to "STB," but a full-text review is needed to eliminate similar errors. In line 186, SLB should be STB.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf