Testbed Emulator of Satellite-to-Ground FSO Downlink Affected by Atmospheric Seeing Including Scintillations and Clouds

Round 1

Reviewer 1 Report

In this work a self-developed hardware channel emulator, which representing an FSO channel was presented.

It is an overall very well written and very interesting and innovative manuscript. The introduction section is comprehensive and the appropriate references have been used. Methodologies are described adequately and maybe such a concept like the presented, can be an important tool for the evaluation of FSO systems.

I only have some minor comments/clarifications, which could be also added to the manuscript.

Why authors decide to emulate satellite links? is it possible to use the specific hardware for other channels? for terrestrial links for example, etc.

Even the statistical distributions Lognormal and Gamma-Gamma represent a wide range of turbulence conditions, why authors considered these two specific distributions? Could any other distribution of the known ones be used?

Moreover, would be possible to consider more performance degradation parameters such as pointing errors for future works?

Author Response

Thank you extremely much for your positive comments regarding our manuscript entitled “Testbed emulator of satellite-to-ground FSO downlink affected by atmospheric seeing including scintillations and clouds”. We highly appreciate the time and the effort to provide us with such valuable and helpful remarks regarding our research article. We have carefully evaluated and answered to all of the them. We also implemented the proposed improvements in our manuscript. Finally, all revisions to the manuscript are marked up using the “Track Changes” function.

• Why authors decide to emulate satellite links? is it possible to use the specific hardware for other channels? for terrestrial links for example, etc.

Author response: Thank you very much for your comment. This is certainly a very important remark. In the submitted manuscript a related remark (the specific hardware was used for characterization of deep space FSO links) appeared in Section 1. In the revised version, we have used the reviewer’s comment (thank you) and extended the remark accordingly. Similar to clouds, the provided hardware can be used for emulation of fog effect which is the most important regarding terrestrial links.

Author action: We updated the manuscript by adding that that the provided hardware emulates well also terrestrial FSO links which are mainly degraded due to fog effect. (line 113)

• Even the statistical distributions Lognormal and Gamma-Gamma represent a wide range of turbulence conditions, why authors considered these two specific distributions? Could any other distribution of the known ones be used?

Author response: Thank you very much for your comment! It is completely right that there are other statistical distributions used for atmospheric turbulence modelling.  Such a model is M or Málaga distribution that is a generalized statistical model which covers all types of irradiance fluctuation conditions in the presence of atmospheric turbulence. While Malaga model is well proven and possesses number of advantages including closed-form and mathematically-tractable expressions, Lognormal and Gamma-Gamma modeling is very well presented in the literature with possibility for straightforward implementation in all types of FSO scenarios.   

Author action: A few sentences discussing this issue are now added. In particular, the irradiance fluctuation due to atmospheric C_n² parameter are modelled with number of statistical distribution. While new generalized statistical models such as Malaga distribution exist, hereby a straightforward modeling based on Log-Normal and Gamma-gamma distribution is applied. (line 221)

• Moreover, would be possible to consider more performance degradation parameters such as pointing errors for future works?

Author response: This is an important issue. In general, along with atmospheric effects, FSO channel modeling includes other problems such as pointing errors, background noise, etc. While the current work is focus on atmospheric effects, future works will consider also other issues including pointing errors.

Author action: We updated the manuscript with the necessary clarification which addresses the issue accordingly. (line 465)

Reviewer 2 Report

Hi, I have no critical comments to make. The paper is very well structured. There is still scope of improving the language quality  of the paper.

Author Response

Thank you extremely much for your positive comment regarding our manuscript entitled “Testbed emulator of satellite-to-ground FSO downlink affected by atmospheric seeing including scintillations and clouds”. We highly appreciate the time and the effort to provide us with such valuable and helpful remark regarding our research article. All revisions to the manuscript are marked up using the “Track Changes” function.

There is still scope of improving the language quality  of the paper.

Author response: Thank you very much for your comment! The English was carefully revised and improved.

Author action: In accordance with the reviewer’s comment, the overall English writing was improved.

Reviewer 3 Report

The authors propose to emulate the wavefront distortion of an optical beam due to atmospheric turbulence by means of a hardware-based optical fiber setup.

The paper has major flaws:

1. It is not clear to the reviewer and the authors fail to clearly demonstrate, that an optical fibre based hardware emulator is able to reproduce the effect of atmospheric turbulence on the wavefront distortion of an optical beam, which is what they claim (see, for example, the sentence on line 127-129). The hardware proposed based on an optical fibre system and variable optical attenuator simply accounts for optical losses due to attenuation, which may be time-dependent, and it is not demonstrated how this setup is able to reproduce scintillation effects let alone wavefront distortion or scattering.
2. The limitations of the hardware and its validity as a means to emulate turbulence effects should be clearly stated.
3. Figure 12 presents experimental and simulation results for the BER. Although the atmospheric modelling effects are well treated in previous sections, it is not clear how the simulations of the BER were obtained.
4. No comparison with real FSO data is reported for validation.

Additional comments follow:

• English writing is very poor, major sentences require rewriting:
  • frequently definite/indefinite articles are missing and prepositions are wrongly used.
  • some sentences are very long and hard to read: e.g. sentences starting at lines 108 and 363.
  • it is suggested not the use of the term "self-developed" (line 14 "... a self-developed hardware channel..."; it seems that the hardware developed itself. Just say "a hardware was developed".
  • the word prototype is misspelt several times ("porotype")
  • In describing the structure of the paper in line 119, it is said "On contrary, in section 3 ...". It should read "On the contrary, ...", yet this construction does not make sense at this point, since there is nothing to disagree with. 

  • line 200 "In other words, ..." I believe what the authors mean is "Alternatively,..."

  • line 163 sentence " In order to evaluate the performance of the built hardware, set attenuation versus induced optical attenuation into optical fibre is measured...", does not make sense.
  • it suggested the authors seek professional advice in reviewing the English writing.

• The quality of some pictures/plots should be improved:
  • Figure 3.
  • Font size in the inset of figures is very small and hard to read (Fig. 8 and 12)
  • Figure 11 is very small
  • Figure 13, expand the height of the eye diagrams
• The structure of the paper should be improved:
  • It is suggested to swap sections 2 and 3, not jump from the hardware and modelling/simulation topics.
• line 133

• • "Besides optical attenuation, coherent detection systems are vulnerable... In general, those disruptions are internally compensated and can be addressed here by using an additional MZM."
  • If the hardware is based on intensity modulation/direct detection technique how can it detect coherent modulation and what is the relevance, since the proposed system used NRZ modulation?

•  line 178, it is said that cloud and atmospheric turbulence effects are simulated, but no description of the simulation is presented apart from the basic formulation of the theory. How is the simulated BER obtained?
• line 333, explain how the simulated joint probability was obtained.
• line 380, P_th is not defined.
• line 384, it is said that the incoming RF data is converted to NRZ. Is it really an RF signal or simply binary data?
• line 402, reducing the BER does not increases the receiver sensitivity except if the receiver bandwidth is reduced accordingly. Has this been performed in the experiments?

Author Response

Thank you extremely much for your comments regarding our manuscript entitled “Testbed emulator of satellite-to-ground FSO downlink affected by atmospheric seeing including scintillations and clouds”. We highly appreciate the time and the effort to provide us with such valuable remarks regarding our research article. We have carefully evaluated and separately answered to all of the them. We also implemented the required changes and improvements in our manuscript which we hope to be positively approved. Finally, all revisions to the manuscript are marked up using the “Track Changes” function.

1) It is not clear to the reviewer and the authors fail to clearly demonstrate, that an optical fibre based hardware emulator is able to reproduce the effect of atmospheric turbulence on the wavefront distortion of an optical beam, which is what they claim (see, for example, the sentence on line 127-129). The hardware proposed based on an optical fibre system and variable optical attenuator simply accounts for optical losses due to attenuation, which may be time-dependent, and it is not demonstrated how this setup is able to reproduce scintillation effects let alone wavefront distortion or scattering.

Author response: Thank you for your comment. This is certainly an important issue. Atmospheric turbulences disrupt the wave front of the optical wave and among other effects (i.e., synchronization issues due to phase fluctuations), mainly lead to fluctuation of the optical power in the plane of the receiving aperture [11], [22]. In the current work is clearly stated that the turbulence fluctuations causing wave front distortions (i.e., phase and irradiance random changes) are evaluated based on their influence over the collected instantaneous FSO power valid for direct detection techniques combined with intensity modulations. However, the proposed hardware channel emulator cannot fully reproduce coherent FSO detection systems where a synchronization issue between Local Oscillator (LO) and incoming signal due to wave front distortions should be also taken into account based on additional MZM. Nevertheless, this limitation is allowed because in near- and deep- space FSO feeder links mostly OOK or DPSK are implemented and rarely coherent detection systems including LO [17].

Author action: We added clarification regarding the limitations of the developed hardware FSO channel emulator. In particular, the channel emulator is limited to a near-space FSO systems having direct detection technique combined with OOK or DPSK modulation, while the rarely used near-space FSO systems equipped with coherent detection scheme are not considered. Moreover, we have made several changes in the text of Section 2 in order to deliver the above described message in a clearer way. (line 130 – line 142)

2) The limitations of the hardware and its validity as a means to emulate turbulence effects should be clearly stated.

Author response: Thank you for your comment. We completely agree. We added a relevant remark regarding the limitations in section in Section 2.

Author action: We added clarification regarding the limitations of the developed hardware FSO channel emulator. In particular, the channel emulator is limited to a near-space FSO systems having direct detection technique combined with OOK or DPSK modulation, while the rarely used near-space FSO system equipped with coherent detection scheme are not considered. (line 135)

3) Figure 12 presents experimental and simulation results for the BER. Although the atmospheric modelling effects are well treated in previous sections, it is not clear how the simulations of the BER were obtained.

Author response: True. This is certainly an important issue. The BER is obtained based on the experimental setup given in Figure 11 with the considered Agilent N4906B BER testers. In particular, atmospheric turbulence-induced fading is introduced in the experimental setup by using the FSO channel emulator as well as the RAOB-based simulated attenuation data for Vienna, Austria (Figure 5). The simulations are accomplished at 100 Hz fading frequency (response time) that is set up in the developed attenuator GUI (Figure 2).

Author action: We updated the manuscript by adding a statement regarding the obtained BER. We explicitly mentioned that atmospheric turbulence-induced fading is introduced in the experimental setup by using the FSO channel emulator as well as the RAOB-based simulated attenuation data for Vienna, Austria (Figure 5). In addition, we made a comment that the measured and approximated BER performance, based on the experimental setup applying Agilent N4906B BERT (Figure 11), is provided in Figure 12. (line 399 and line 412)

4) No comparison with real FSO data is reported for validation.

Author response: Thank you very much for your comment. This is certainly a very important remark. Only a few ground-space bidirectional links (e.g., Teide Observatory - ARTEMIS GEO satellite FSO link) offering real evaluation of atmospheric turbulence and cloud effects over transmitted optical beam have been established [15]. While those initiatives provide the whole picture by means of testing real FSO systems, this is a costly procedure and real measured data are available only for few special locations. Those FSO links exclude the selected scenario for Vienna, Austria. Nevertheless, the used Radiosonde Observation (RAOB) approach is well verified in [7]. Being developed as one of the very first prototypes for verification of high-demanding SNSPD-based (Superconducting Nanowire Single-Photon Detector) deep-space FSO channels, its design, given in [19] and [20], allows verification of other scenarios including terrestrial and near-space links. Its operation is hereby translated to a platform for evaluation of near-Earth SatCom links implementing 10 Gbps APD-TIA photodetector (PD) installed in future OGSs (Optical Ground Stations) for LEO/GEO data transmission [21]. Offering capabilities for testing COTS components as well as technologies, the testbed implements a state-of-the-art fiber-based hardware FSO channel emulator representing atmospheric turbulences using RAOBs [7], [8].

Author action: A part discussing the missing possibility for implementation of real measured data (i.e., atmospheric turbulence and clouds data) is now added in Section 1 and Section 5. Unfortunately, although a room for further comparison with real measurements is left, to best of authors knowledge such atmospheric data are still not available for Vienna, Austria scenario. (line 93 and line 438)

Additional comments follow:

5) English writing is very poor, major sentences require rewriting:

• • frequently definite/indefinite articles are missing and prepositions are wrongly used.
  • some sentences are very long and hard to read: e.g. sentences starting at lines 108 and 363.
  • it is suggested not the use of the term "self-developed" (line 14 "... a self-developed hardware channel..."; it seems that the hardware developed itself. Just say "a hardware was developed".
  • the word prototype is misspelt several times ("porotype")
  • In describing the structure of the paper in line 119, it is said "On contrary, in section 3 ...". It should read "On thecontrary, ...", yet this construction does not make sense at this point, since there is nothing to disagree with. 
  • line 200 "In other words, ..." I believe what the authors mean is "Alternatively,..."

Author response: Thank you very much for your comments! The English was carefully revised and improved.

Author action: In accordance with the reviewer’s comments, all proposed corrections are accomplished. In addition, the overall English writing was improved.

6) The quality of some pictures/plots should be improved:

• Figure 3.

Author response: Thank you very much for your comment! The quality of Figure 3 is slightly improved. The resolution depends on SPIRICON PYROCAM III beam camera and its software, which cannot provide better quality.

Author action: In accordance with the reviewer’s comment, Figure 3 was improved.

• Font size in the inset of figures is very small and hard to read (Fig. 8 and 12)

Author response: Thank you very much for your comment!

Author action: In accordance with the reviewer’s comments, the Font size of Figure 12 was improved. Unfortunately, the large amount of text within Figure 8 and Figure 12, do not allow bigger font size.

• Figure 11 is very small

Author response: Thank you very much for your comment!

Author action: In accordance with the reviewer’s comment, Figure 11a was slightly increased. Figure 11b is just a photo that proves the carried out experiment and doesn’t provide some important information.

• Figure 13, expand the height of the eye diagrams

Author response: Thank you very much for your comment!

Author action: In accordance with the reviewer’s comment, the height of eye diagram is increased.

7) The structure of the paper should be improved: It is suggested to swap sections 2 and 3, not jump from the hardware and modelling/simulation topics.

Author response: Thank you very much for your comment! While revising the paper, we examined a potential swapping of sections, but finally realized that such change would not improve much the paper quality. The hardware and modeling/simulation parts are equally important topics. We defined first the hardware channel emulator (Section 2) and after that the required atmospheric turbulence and cloud attenuation data (Section 3). These two sections are completely different from Section 4 where the performance of APD-TIA is evaluated.

Author action: No visible action taken.

8) line 133, "Besides optical attenuation, coherent detection systems are vulnerable... In general, those disruptions are internally compensated and can be addressed here by using an additional MZM." If the hardware is based on intensity modulation/direct detection technique how can it detect coherent modulation and what is the relevance, since the proposed system used NRZ modulation?

Author response: Thank you for your comment. We completely agree. This is the main limitation of the proposed hardware. We added a relevant remark regarding the limitations in section in Section 2.

Author action: We added clarification regarding the limitations of the developed hardware FSO channel emulator. In particular, the channel emulator is limited to a near-space FSO systems having direct detection technique combined with OOK or DPSK modulation, while the rarely used near-space FSO system equipped with a coherent detection scheme are not considered. Moreover, we have made several changes in the text of Section 2 in order to deliver the above described message in a clearer way. (line 135)

9) line 178, it is said that cloud and atmospheric turbulence effects are simulated, but no description of the simulation is presented apart from the basic formulation of the theory. How is the simulated BER obtained?

Author response: This is an important issue. In the submitted manuscript a related remark appeared that atmospheric effects are simulated in order to provide the required test data for optical attenuation. The RAOB data together with the applied probabilistic model (the equation part) are used to generate the atmospheric turbulence-induced attenuation (The built normalized histogram of the simulated turbulence attenuation data is shown in Figure 5). Moreover, the cloud attenuation is statistically modelled based on empirical approach using Log-Normal distribution (i.e., equation (18)).

Regarding the BER measurements, they are only accomplished by applying the simulated atmospheric turbulence attenuation data. While optical losses due to clouds are also simulated, the cloud attenuation is considered to be static one in comparison to fast atmospheric turbulence fading. Consequently, clouds attenuation isn’t used during the BER measurements.

Author action: In the revised version we have used the reviewer’s comment (thank you) and extended the remark regarding the simulated optical attenuation accordingly. We explicitly mentioned that atmospheric turbulence-induced fading (without cloud attenuation) is introduced in the experimental setup by using the FSO channel emulator as well as the RAOB-based simulated attenuation data for Vienna, Austria (Figure 5). In addition, we made a comment that the measured and approximated BER performance based on the introduced experimented setup applying Agilent N4906B BERT (Figure 11) is provided in Figure 12. (line 182, line 349, line 399, line 412)

10) line 333, explain how the simulated joint probability was obtained.

Author response: Thank you. This is certainly a very important remark. Having both attenuation variables (i.e. atmospheric turbulence and clouds) defined in the same probability space, simulated joint probability density function on all possible optical attenuation pairs of output is provided in Figure 9.

Author action: We updated the manuscript by adding the required clarification. (line 343)                                                                                                                          

11) line 380, P_th is not defined.

Author response: Thank you very much for your comment. The P_th is the damage threshold of APD-TIA and is already defined in Section 4. (line 361)

Author action: No visible action taken.

12) line 384, it is said that the incoming RF data is converted to NRZ. Is it really an RF signal or simply binary data?

Author response: Thank you for your comment. This point was apparently overlooked in the submitted manuscript. Binary data is the proper term.

Author action: Instead of RF signal, binary data were included. (line 398)

13) line 402, reducing the BER does not increases the receiver sensitivity except if the receiver bandwidth is reduced accordingly. Has this been performed in the experiments?

Author response: Thank you for your comment. Authors completely agree with the comment. Only measurements with two different data rates (1 Gbps and 10 Gbps) were performed during the experiment. (line 420)

Reviewer 4 Report

In free space optical communication technology, when the laser beam as the information carrier is transmitted in the atmosphere, it will be affected by atmospheric attenuation and turbulence effect, resulting in the degradation of the performance of the communication system. How to evaluate the impact of atmosphere on FSO has become a hot issue. In this paper, as an alternative to real-time evaluation of the impact of atmosphere on optical signals, a state-of-the-art laboratory test-bed based on slant avalanche photodiode FSO link is proposed. By combining self simulated scintillation data with self-developed hardware, the high-end testing of ground-space FSO links is carried out. The experimental results show that the scheme proposed in this paper can evaluate the impact of the atmosphere on FSO in real time.

Author Response

Thank you extremely much for your positive comments regarding our manuscript entitled “Testbed emulator of satellite-to-ground FSO downlink affected by atmospheric seeing including scintillations and clouds”. We highly appreciate the time and the effort to provide us with such valuable and helpful remark regarding our research article. All revisions to the manuscript are marked up using the “Track Changes” function.

Round 2

Reviewer 3 Report

The authors replied satisfactorily to all the comments. The only issue at the moment is that editing of the English language is still required, although it has been much improved. Validation of the results with real data would be a bonus.

Comments for author File: Comments.pdf

Author Response

Thank you extremely much for the second round of comments regarding our manuscript entitled “Testbed emulator of satellite-to-ground FSO downlink affected by atmospheric seeing including scintillations and clouds”. We highly appreciate the time and the effort to provide us with such valuable and helpful comments regarding our research article. We have carefully evaluated and answered to all of the them. We also implemented the proposed improvements in our manuscript. Finally, all revisions to the manuscript are marked up using the “Track Changes” function.

1)The authors replied satisfactorily to all the comments. The only issue at the moment is that editing of the English language is still required, although it has been much improved. Validation of the results with real data would be a bonus.

Author response: Thank you very much for your comment! The English was additionally revised and improved.

Author action: In accordance with the reviewer’s comments, the overall English writing was additionally improved. Moreover, the authors carefully evaluated and implemented all proposed changes by the reviewer.

2)What does the two sets of numbers mean? The attenuation is often referred in dB as a positive number. (Line 319)

Author response: Thank you for your comment. This is certainly an important issue. This is the range of the simulated cloud attenuation, namely the minimum and the maximum attenuation values in [dB].

Author action: In accordance with the reviewer’s comment, the necessary clarification is provided (line 317).