Simulation and Design of an Underwater Lidar System Using Non-Coaxial Optics and Multiple Detection Channels
Round 1
Reviewer 1 Report (Previous Reviewer 3)
The authors have addressed the raised points of criticism appropriately and adapted the manuscript accordingly. Still minor issues remain and should be tackled before publication of the manuscript. The main points are:
(i) I appreciate your comments w.r.t. to the far-field only applicability of the laser-radar equation in the rebuttal letter. However, your corrected statement does not entirely clarify matters. Please state clearer that the laser-radar equation applies in the far-field only and that, e.g., your non-coaxial design inherently considers this to a certain extent as the overlap factor O gets zero in the near-field. With your configuration, O gets greater than zero after a few meters, which is still near-field. However, the simulated near-field waveforms match the measured waveforms surprisingly well. This could be used as a confirmation for the applicability of the laser-radar equation also in the near-field, although theory suggests that the equation only holds in the far-field.
(ii) Experimental results: You report precision and mean deviation. I assume that the mean deviation is based on the comparison of the real and measured distances. You do not discuss the reasons for this bias, nor did you try to correct the measurements in this respect. Add a short discussion concerning the bias in Section 5.
(iii) Please find minor editorial suggestions in the attached commented PDF.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 2 Report (New Reviewer)
This manuscript presents an underwater lidar system for target detection which uses multiple detection channels to increase the dynamic range of the system. Especially, a gating controlled PMT is equipped to compress the strong near-field signal, which can greatly improve the detecting ability of the detector for the weak far-field signal. Generally, the manuscript is well written.
1. Line 341 to 342, the units of both attenuation coefficient (c) and absorption coefficient (a) are missing. The same problem also appeared in line 421 and 422.
2. According to Eq.(4), βπ is needed to simulate the lidar signal. However, the in-situ measurements of water properties in both pond and swimming pool only provide attenuation coefficient (c) and absorption coefficient (a). Please supplement how βπ is derived in the simulation.
3. Is this prototype water resistant? If so, what is its water-resistant depth?
4. Line 404，“The simulated waveform is consistent with the measured signal”. It can be seen from Fig.8(b), similar as Fig.8(a), the measured signal also has a higher intensity than the simulated waveform at around 100 ns. The possible reason for this issue should be explained in the manuscript.
5. In both pond and swimming pool experiments, what is the delay of the gating signal? How long is the response time the gating controlled PMT? It’s better to state it in the manuscript.
6. In Fig. 10 (h) and (i), the signal of channel 3 has two peaks, which are also appeared in the signal of channel 1. According to the signals shown in figs (a) to (g), the first peaks in channel 3 should be those from the targets. Therefore, why there are the second lower peaks in channel 3? This should be explained as well.
Generally, the English expressions in this manuscript are fine.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
In the paper, authors proposed a novel paraxial multichannel detection strategy to improve the accuracy and the range of the underwater laser detection technique. The simulations and experiments are both conducted to verify the effectiveness of the proposed method. While there are potential implementations on underwater sensing and mapping, the main novelty is unclear, and even confusing.
1. The title is “Simulation and Design of an Underwater Lidar System”, but in the abstract and conclusions sections, authors emphasized “a novel paraxial multichannel detection strategy”. Thus, the reviewer is confused about the main novelty of the paper, to design a system or propose an algorithm ?
2. Although authors always mentioned “the novel paraxial multichannel detection strategy”, the relevant part (Multiple Channels Detection Model) only occupies 18 lines (from Line 182 to 200). If the multichannel detection strategy is main novelty of the paper, then the novelty is very limited.
3. In Conclusions section, line 468, authors stated that the reliability of the simulation was verified. Hence, again, what is the MAIN novelty of paper ? If that is the multichannel detection strategy, the simulation method should be used to validate the proposed strategy, rather than being verified by experiments.
4. The literature review part need to be extended to provide broader landscape of the research.
5. Check spellings, e.g. in line 471, “These findings suggests”.
In the reviewer’s opinion, the paper is very interesting with rigorous experiments. However, the methodology part is trivial, and the main novelty of the paper is confusing and limited. It is not adequate to be published in the journal, at least, at current stage.
Reviewer 2 Report
The paper proposes a novel off-axis multi-channel underwater detection strategy that incorporates optical energy splitting, gated techniques, and high-low gain settings. This approach has been shown to successfully broaden the dynamic range and suppress backscattering, while using relatively simple hardware circuit designs.
The authors give a short review of STC methods and a description of existing lidar simulators published in the literature.
It is not clear why the background light calculation is based on artificial light rather than sunlight. Is this due to the fact that the authors are targeting deep water, or they feel solar ambient, both bottom reflection and backscatter will be insignificant due to the use of the bandpass filter. If so, then this should be stated.
The dark noise Gaussian noise distribution would be interesting if it was complemented with a similar plot from the CH1, CH2 and CH3 ranging accuracy measurements, where the distribution is expected to become Poissonian for a well-designed system that is shot-noise limited. Indeed it not correct that the authors give the detector noise generation Nn term in figure 2 a Gaussian distribution when they are using PMT detectors, and it is not clear whether the model includes an actual term for detector shot noise.
No information on how the attenuation and absorption coefficients were measured, and whether they mean the beam attenuation (c) or diffuse attenuation coefficient (Kd). Indeed careful attention needs to be paid to both scattering and temperature corrections with most designs of transmissometer and this particularly important when making absolute comparisons between experimental and analytical waveforms. The number of one-way beam attenuation lengths that the system is able to obtain range measurements over would be interesting for the authors to state, as this would make this interesting work comparable with other such work in the literature.
The simulations are not capturing the effects of multiple scattering on the backscatter part of the waveform but it is not clear that the analytical model as presented will simulate the effect of multiple scattering on the target return, which seems to be included by the appearance of the tail on the pulse return part of the waveform? Monte Carlo methods are typically used to accurately capture multiple scattering effects. It would be important for completeness if the authors could include more of a discussion about the method used to simulate the multiple scattering tail of the target return?
In the comparison between experimental and analytical (simulated) in figure 8 was the simulated waveform normalized to the experimental Echo signal Max? To rationalize the results in figure 8 it would be good to know what some of the input parameters of the model were - e.g. AR, the PMT receiver area or diameter and what type of gating mechanism is being employed for the gated PMT, and does it have the same diameter as the non-gated PMT? What was the gain and the maximum average anode current of the PMTs? This is important as the alternate gating mechanisms have different impacts on the noise and dynamic range.
Spelling mistake on line 186 and 453
Reviewer 3 Report
The authors described both a method for simulating underwater lidar measurements using a non-coaxial multi-channel sensor and the design of a respective prototype instrument. The main idea is to use separate receivers for the near-field (two channels) and the far-field, where only the latter is range gated. This has a positive effect as the far field receiver does not suffer from abundant near-field volume backscattering and the near-field receivers effectively suppress the volume backscattering peak with the non-coaxial design. One motivation for doing this, is increasing the usable measurement distance (avoiding a purely range gated approach), which however has also its limits due to the non-overlap of the laser footprint and the receiver's FoV close to the instrument. The minimum detectable distance should therefore be more clearly stated and discussed in the paper.
I a part of the paper is related to the simulation process, where most of the material is taken from Abdallah et al (WaLID). WalID is designed for airborne/spaceborne platforms, an consequently does not consider reflections from the the sensor's window in an underwater setting. The respective part of the formula in the current MULS research is, however, inappropriate, as the laser radar equation is used, which is only valid in the far field and must not be used for signal interactions very close to the object. This needs revision.
The experiments and results are clearly presented. The discussion sometimes suffers from unclear statements or missing information (e.g. no target reflectance reported for the maximum detection range in the simulation). The discussion, also, needs revision.
Apart from that, the article is well structured and the English reads well. I have to state that, as a photogrammetrist with a certain but not an expert-level knowledge on electric engineering, I can rather comment on the mapping related aspects of the paper.
Please also consider the comments in the attached PDF.
Comments for author File: Comments.pdf
