Underwater Coherent Source Direction-of-Arrival Estimation Method Based on PGR-SubspaceNet

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The main novelty of this paper is the proposed PConv-GAM Residual SubspaceNet, which embeds the global attention mechanism (GAM) into residual blocks that fuse PConv convolution. This model is able to capture richer cross-channel and positional information, improving the accuracy of direction-of-arrival estimation in complex underwater conditions.

The reviewer's main concerns and queries are as follows:

1. The introduction of PConv-GAM residuals into the improved PGR-SubspaceNet model architecture begs the question of whether it has increased the algorithmic complexity. The author should provide appropriate elaboration on this matter.

2. What different features or details does the improved PGR-SubspaceNet model extract that enable it to achieve better performance?

 

3. The reviewer suggested that the author further distinguish the line types in Figure 6.

Comments on the Quality of English Language

Despite the presence of minor issues, the overall grammar level of the paper is acceptable and does not hinder the understanding of the main ideas. It is recommended to have the paper checked and revised by a professional editor or a native English speaker.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

To overcome the SNR limitations in the field of underwater acoustics, the authors propose an improved model based on SubspaceNet, called PConv-GAM Residual SubspaceNet (PGR-SubspaceNet).

The paper is correctly developed and well structured in its stages. Authors are only recommended to cite equations that are not their own, make the contributions clear in the introduction and leave a schedule-type paragraph at the end of the introduction.

Comments on the Quality of English Language

The paper has good English writing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The article presents an improved method that determine the direction of arrival (DOA) in an underwater environment. Even if DOA is a general topic and more articles from this field are introduced in Reference list, in my opinion only one article about underwater it too little.  

In the following, some observations about the article are presented.

The output of this algorithm is the angle, but there is no an expression that calculates this quantity.

Figure 5 presents the result that RMSPE decreases if SNR increases from -7 to 7 dB. But comparing the results of Tables 2 and 3, it follows that RMSPE for SNR of 5 dB is higher than RMSPE for SNR of -5 dB.

Why the results of Pool experiment are presented in a different manner compared to the simulation?

What does represent the quantity Amplitude in dB? What is the reason of the peak of about -50 degrees?

The experiment uses a 10 uniform line array. Then something is said about the first array. Are there more arrays?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear author, thank you for sending this article through, this is suitable for the current journal applied for.

Quality of English is very good. I would only address the authors to the following topis: can the results be compared with a site measurement survey? please clarify how the varibailities can affect the change of directivity based on wave current flow, presence of other source in the water, etc. that can create diffraction of the sound trajectory.

The abstract is good to anticipated the performance of the results. good.

Another thing is related to the conclusions, that should address future research studies based on the limits of your research. 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf