An ADS-Based Sparse Optimization Method for Sonar Imaging Sensor Arrays
Round 1
Reviewer 1 Report
Explain more clearly the ADS algorithm. Attach some figure [ 85-102 ]
Explain here that it is w[n]. It is introduced later in the text [92]
The letter “d” is used to express the separation between sensors, when here it means d/landa, look for another letter for this concept [ formula 6 ]
Explain in more detail how to obtain this formula [ formule 10 ]
How to obtain this expression: 0.8488 + 1.128 Log N [128]
Change Avid by avoid [152]
Why has this ADS sequence been chosen? Are there others? [179]
To better explain the concept of number of cycles. How many different sequences can be formed? [200]
It is not logical that the width of the main lobe does not increase as the azimuth angle increases. Justify why it does not increase when the array aperture decreases. Include a table with the width of the main lobe for the 5 pointing angles evaluated [213].
Include a table with the PSLL values for the three Mill Cross algorithms, ADS and CDS, for the various steeering angles. The graphs are fine, but they do not allow a quick analysis of the results [220].
Why isn't the beamwidth used to evaluate the resolution in azimuth? Justify why the MVDR algorithm is used [239].
Include a summary table with the resolution in azimuth for the 4 cases [260].
In figure 9, display the angular response from -5 degrees to 5 degrees, so that the resolution is clear. Indicate that the array is steering to zero degrees in elevation or azimuth
In figure 10, put the x-axis in Khz and the PSLL graph show it between -15 db and -10 db, to see in detail the variation. Is the author sure that in 400 khz the wolves do not start appearing grating lobes?
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The paper entitled "The ADS-based sparse optimization method for sonar imaging sensor array" aims to optimize the cross-plant network on sonar sensors without any degradation in image quality. The authors propose an approach based on the ADS-method in which 4 criteria are analyzed : effect of cycle index, beam performance at different azimtuh angles, azimtuh resolution, frequency performance.
The article seems well-structured to me, and the analysis empirical study is generally well conducted. A few elements need to be better explained.
The most recent references are from 2016. Are there no more recent references in the field?
I didn't understand whether the authors were working with simulated or real data. In the latter case, it should be made clearer. Moreover, it would be interesting to illustrate, for example by a diagram, the experimental protocol. Otherwise, is there a way to test on real data, with real sonar.
Figure 4 : the authors should define the two axes
line 208 : diagrams are given in figure 6 when anglas are 0°, 30° and 45° respectively. In figure, the authors present results with angles 60° and 75°. Is there an interest to present these results ?
minor corrections
line 66 MIMO --> Multiple Inputs and Multiple Outputs
line 68 can't--> cannot
line 152 avid --> avoid
line 232 can't --> cannot
line 245 it's --> it is
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper seems to be an interim report, with poor English
The authors acknowledge the need for further guidance (line 233)
I suggest that after this has been done they re-submit, taking more care with the English
I suspect that incorrect words are being used
eg line 12 decrease (decreasing?)
line 73 popularized (populated?)
line 77-81 a long sentence
line 152 avid (avoid?)
line 175 are reflected (be displayed??)
line 284 declined (be reduced?)
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
I approve of the changes
