Precise RF Phase Measurement by Optical Sideband Generation Using Mach–Zehnder Modulators
Round 1
Reviewer 1 Report
Dear authors,
1. The abstract is too short to fully describe the important research contents or conclusions of the paper. E.g, the theoretical characterization of MZM should be introduced in this part.
2. What is the basis original of Eq.(1) ? I suggest a reference or some references should be cited here. Further more, I don’t understand the meaning of K, which is defined as the insertion loss in Line 74. Why does the square of K have the dimension of intensity
3. The format of the descriptions of X-axis and Y-axis in Figs 2-5 can be improved. The units are missing for Y-axis in Fig. 2, Fig.3, Fig. 4 (a), and Fig.5 (a). The units are missing for X-axis in Fig.3, Fig. 4 (a), and Fig.5 (a). It is better to describe the Y-axis in Fig. 2 (a) and Fig. 2 (b) as “Bias phase (degree)” and “RF phase (degree)”, respectively.
4. Finally, there are a few mistakes in English expression. E.g., In Line 93, “because their intensity is in general larger than that of high-order components”. In Line 126, “can help to determine the proper solutions”. In Line 176, “Relationship between measured bias phase and applied DC bias voltage”.
Author Response
Point 1: 1. The abstract is too short to fully describe the important research contents or conclusions of the paper. E.g, the theoretical characterization of MZM should be introduced in this part.
Response 1: Thank you for your suggestions. The abstract part has been reviewed and modified. The MZM characterization is metioned and the final achievement is induced.
Point 2: What is the basis original of Eq.(1) ? I suggest a reference or some references should be cited here. Further more, I don’t understand the meaning of K, which is defined as the insertion loss in Line 74. Why does the square of K have the dimension of intensity
Response 2: The reference for Eq. (1) has been added. K is defined as the intertion loss of the hole MZM. We have changed the expression here, the “intensity ” here means the output power of optical sideband. Basically, K is the parameter describe the ratio.
Point 3: The format of the descriptions of X-axis and Y-axis in Figs 2-5 can be improved. The units are missing for Y-axis in Fig. 2, Fig.3, Fig. 4 (a), and Fig.5 (a). The units are missing for X-axis in Fig.3, Fig. 4 (a), and Fig.5 (a). It is better to describe the Y-axis in Fig. 2 (a) and Fig. 2 (b) as “Bias phase (degree)” and “RF phase (degree)”, respectively.
Response 3: Yes, the format of description in figures should be improved. Fig. 2, Fig.3, Fig. 4 (a), and Fig.5 (a) have been modified. The units of X-axis and Y axis have been added.
Point 4: Finally, there are a few mistakes in English expression. E.g., In Line 93, “because their intensity is in general larger than that of high-order components”. In Line 126, “can help to determine the proper solutions”. In Line 176, “Relationship between measured bias phase and applied DC bias voltage”
Response 4: Thank you for your advices. We have asked for the professional editing of English language and style from Editage. The certificate of editing is uploaded in attached file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper presents a novel method to measure optically RF phase using Mach-Zehnder modulators. The method is interesting but the manuscript presents some serious flaws.
Indeed, the authors claim that they can determine the absolute error of their measurement to be less than 10 degrees. Nevertheless, this absolute error (if I understood correctly) is determined as the absolute difference between the phase obtained through the authors' method and that determined by a cross-domain analyzer (CDA). The problem is that the authors state that the "CDA has a variance of +-15 degrees" (lines 215-216).
Therefore, the claim of an error of less than 10 degrees cannot be supported by the data, since the uncertainty of the CDA is higher than that.
Other comments on the rest of the paper:
- All acronyms should be stated in full before being used (lines 29, 30, 37, 38)
- The last part of the introduction (lines 49 to 54) can be clearer. The connection with the rest of the manuscript is too weak.
- Chapter 2.1 can be clearer. sometimes it is difficult to follow the authors' stream of thoughts (e.g. lines 105-106 the sentence is truncated). The effect of DC bias could be explicated and a scheme could be useful to help readers.
- In chapter 2.2 the brands and models of all the instruments used should be added. Moreover, the single MZM part (including the laser diode, the MZM, the bias tees, and the DC power supplies are not addressed.
- The LabView and MATLAB scripts should be available as supporting information
- In figures 3, 4, and 5 the error bars (both for X and Y) due to the instruments' uncertainty should be added. A discussion on the uncertainty of the method should be added considering all the factors (e.g. laser diode stability and noise, DC stability, etc)
- Repeatability should also be tested. For the same frequency, are the results the same for different rounds of measure?
Author Response
The main problem: The paper presents a novel method to measure optically RF phase using Mach-Zehnder modulators. The method is interesting but the manuscript presents some serious flaws.
Indeed, the authors claim that they can determine the absolute error of their measurement to be less than 10 degrees. Nevertheless, this absolute error (if I understood correctly) is determined as the absolute difference between the phase obtained through the authors' method and that determined by a cross-domain analyzer (CDA). The problem is that the authors state that the "CDA has a variance of +-15 degrees" (lines 215-216).
Therefore, the claim of an error of less than 10 degrees cannot be supported by the data, since the uncertainty of the CDA is higher than that.
Response : Thank you very much for your constructive comments.
"CDA has a variance of +-15 degrees" represents the specification of CDA. It describes the worst situation but the actual measurement error might be smaller. In order to confirm the accuracy of CDA, we conducted additional phase measurement with vector network analyzer. The VNA we used can provide the accuracy of +-4 degrees in the frequency range of 8 to 20GHz.
The phase measurement for comparison was conducted under 12, 12.5 and 13 GHz. In these three frequencies, the difference in phase measurement between VNA and CDA is +-2 degrees. Based on this fact, we can claim that the variation of CDA under 12, 12.5 and 13GHz would be +-6 degrees, and an error of less than 10 degrees can somehow be supported by the data. The additional test is added as Appendix A in manuscript.
For the suggestions on extensive editing of English language and style, we requested the English editing service from the Editage. The certificate of editing is uploaded in attached file.
Other comments on the rest of the paper:
Point 1: All acronyms should be stated in full before being used (lines 29, 30, 37, 38)
Response 1: Thank your for your comments. These acronyms have been stated in full before being used right now.
Point 2: The last part of the introduction (lines 49 to 54) can be clearer. The connection with the rest of the manuscript is too weak.
Response 2: Some modifications are added to lines 49 to 54. This part is induced to give some basic information of MZM.
Point 3: Chapter 2.1 can be clearer. sometimes it is difficult to follow the authors' stream of thoughts (e.g. lines 105-106 the sentence is truncated). The effect of DC bias could be explicated and a scheme could be useful to help readers.
Response 3: Some of the contents in chapter 2.1 have been modified. (eg. lines 105-106). A few sentences to explain the DC bias are added.
Point 4: In chapter 2.2 the brands and models of all the instruments used should be added. Moreover, the single MZM part (including the laser diode, the MZM, the bias tees, and the DC power supplies are not addressed.
Response 4: Thank your for your advices. Appendix A is added to provide the information the brands and models of all the instruments used in this study. The explaination on the single MZM part is also added into chapter 2.2.
Point 5: The LabView and MATLAB scripts should be available as supporting information
Response 5: The LabView and MATLAB scripts can be available. However, the LabView script is a type of graphfical script. The situation could be quite different depending on the measurement instruments used. We strongly suggest the researchers who is interested in this study design the LabView script based on the measurement instruments they have.
The Matlab script is mainly used for MZM characterization and all the equations used in script have been included in chapter 2.1
Point 6: In figures 3, 4, and 5 the error bars (both for X and Y) due to the instruments' uncertainty should be added. A discussion on the uncertainty of the method should be added considering all the factors (e.g. laser diode stability and noise, DC stability, etc)
Response 6: "+-15 degrees" is actually the specification of CDA. It is not the measurement error we got in the experiment; thus, we don’t think adding the error bars here is appropriate.
The discussion of laser diode stability was added. The wavelength stability of laser diode is ±2.4 pm under 1550nm and the variation of optical output level is ±0.03dB. This would cause minor influence on the measurement result.
Point 7: Repeatability should also be tested. For the same frequency, are the results the same for different rounds of measure?
Response 7: The measurement showed similar results under different rounds of measurement. Actually, the measured value in fig 3,4 and 5 is the average measured value of 5 rounds. We apology for our mistakes, the statements on this point have been added into the manuscript.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors have successfully addressed the previous points.
Just few minor remarks:
- for point 6 I still think that error bars that take into account the various sources of error should be added to help readers to understand the real accuracy of what is presented.
- Line 178: there's a "The" that should be eliminated
- Line 230: I think that "hole" should be "whole"
Author Response
Point 1: For point 6 I still think that error bars that take into account the various sources of error should be added to help readers to understand the real accuracy of what is presented.
Response 1: Thank your for your comments. We accept your suggestions. The error bars have been added to the graph. A few sentences were added to illstrate the meaning of the error bars.
Point 2: There's a "The" that should be eliminated (Line 178)
Response 2: "The"in Line 178 has been eliminated right now. We apology for this mistake.
Point 3: I think that "hole" should be "whole"( Line 230)
Response 3: Yes, the "hole" has been changed as "whole".
