Multipath Effects Mitigation in Offshore Construction Platform GNSS-RTK Displacement Monitoring Using Parametric Temporal Convolution Network
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors- Is the multi-path elimination method proposed in this paper specific to static scenarios? If it is for static scenarios, how significant are the benefits of such processing? If it is for dynamic scenarios, what are the specific effects of the model?
- In the "Data input" column of Figure 4, what specific data is input for "Near field environment"?
- What do the horizontal and vertical axes represent in Figure 6, respectively?
- How are the true values of multi-path errors obtained in Figure 11? Are they from carrier phase multi-path or pseudorange multi-path?
- Specific numerical values should be provided for the horizontal and vertical axes in Figure 11.
- What does the horizontal axis represent in Figure 12?
- Eliminating multi-path errors also serves the purpose of positioning, and the paper lacks comparative experiments on positioning accuracy.
Author Response
Thank you for your review. Please refer to the attached PDF for our detailed responses to your comments.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper is well-structured, and the figures are of high quality, demonstrating a thorough investigation into the topic. However, I have identified several areas that require clarification and improvement. Please address the following points in your revision:
1. The title and keywords emphasize the RTK model, but the abstract does not provide any information about RTK. Additionally, the abstract contains an excessive number of GNSS-related terms, which may obscure the main focus of the study.
2. In L47, "Fig. 1" should be changed to "Figure 1" to maintain consistency in figure numbering throughout the manuscript.
3. In L41, the term "ambiguity resolution (AR)" is used without providing its full form on first mention.
4. Equations 2 and 3 refer to the use of wavelengths, but it is unclear whether the proposed method is specifically designed for CDMA or if it can also be applied to FDMA.
5. Figure 3 shows that GPS and BDS data are processed together, despite the fact that these systems have different orbital periods and satellite constellations.
6. Figure 4 mentions the use of RTK fixed solutions, but in real-world scenarios, float solutions often occur due to cycle slips and loss of lock. Additionally, the term "reflection area definition" is introduced without a corresponding formula or detailed explanation. The paper only considers water surface reflection, but other environmental factors are not addressed.
7. Figure 11 presents three subplots for each system, but it is unclear whether these subplots correspond to ENU (East, North, Up) directions or have another meaning. The discrepancies between predicted and actual values in some subplots are significant, and the trends do not align well. It is also unclear whether the predictions are multi-step and whether the subplots represent the same time period. The PRN numbers of the satellites used in each subplot are not provided, and the conversion process for the "amplitude" on the y-axis is not explained.
8. The y-axis of Figure 12 is labeled as "variation," with units of m², but it is unclear what specific variation is being measured. The x-axis appears to represent time, but this is not explicitly stated.
9. Each subplot in Figure 13 contains three smaller plots, but their meanings are not clearly explained. Typically, GNSS multipath analysis involves Fourier transforms to analyze power and frequency, but this approach is not discussed in the paper.
Author Response
Thank you for your review. Please refer to the attached PDF for our detailed responses to your comments.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsCorrections are given in the PDF document.
Comments for author File: Comments.pdf
Moderate editing of English language required.
Author Response
Thank you for your review. Please refer to the attached PDF for our detailed responses to your comments.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThe manuscript presents an innovative approach to mitigating GNSS multipath effects using a parametric Temporal Convolution Network, with promising implications for offshore construction monitoring. While the methodology is robust, certain sections could benefit from clearer explanations, improved statistical validation, and enhanced visual clarity to strengthen the overall impact and reproducibility of the work. Below are my comments for the authors which needs a careful consideration:
Line 13–14: Keywords are repetitive with the title and do not add significant value. Kindly change them and use succinct and adequate terms.
Line 24–27: The RTK method's description lacks clarity regarding its limitations in mitigating multipath in dynamic environments.
Line 233–235: Dataset acquisition process lacks detail on how noise and anomalies in the data were handled.
Line 264–269: Environment perception methods are vague and do not sufficiently justify the choices made for GNSS positioning configurations.
Line 314–324: The explanation of TCN architecture lacks a rationale for choosing this specific network over other temporal models.
Line 394–396: The attribution of residual trends to orbit and path errors is speculative without additional evidence.
Line 411–414: The limitations of the para-TCN network predictions are acknowledged but not systematically analyzed.
Figure 11 (Line 405): Predictions versus ground truth plots do not show statistical metrics to assess model performance comprehensively.
Line 417–421: The variance reduction discussion lacks statistical validation, making the conclusions appear qualitative rather than quantitative.
Line 443–454: The power spectrum analysis does not compare results with alternative mitigation methods, limiting the interpretive depth.
Line 471–474: The conclusion reiterates the methodology without sufficiently emphasizing the significance of the findings or potential applications.
Author Response
Thank you for your review. Please refer to the attached PDF for our detailed responses to your comments.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsRegarding the revised version, while it has addressed most issues, there is still one matter that requires clarification:
- If, as stated by the author, this method is equally applicable to dynamic scenarios, is there any experimental data from dynamic settings that support this conclusion?
Author Response
Thank you for your comment. You may reffering the attached word for our detailed response.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsI have no quetions about the new manuscript
Author Response
Thank you for your effort in reviewing. May you succeed in all your endeavors.
Reviewer 4 Report
Comments and Suggestions for AuthorsI have no comments. All my queires were well-addressed by the authors.
Author Response
Thank you for your effort in reviewing. May you succeed in all your endeavors.
