Monitoring Structural Displacements on a Wall with Five-Constellation Precise Point Positioning: A Position-Constrained Method and the Performance Analyses

Round 1

Reviewer 1 Report

In this paper, the authors analyzed the performance of the five-constellation PPP technique to monitor SLDW and SVDW by using semi-generated measurements and real GNSS data in different making conditions and the Asia-Pacific mid-low-latitude regions. The authors have done a lot of work, but there are still few problems as follows,

1.     Figures 8-18: Make the supplement of the ‘E/N/U’ tag on the legend or axis titles in the corresponding figures.

2.     Figure 10: Correct the legend content (A = 0.5 cm, F = 0.5 Hz) in the figure.

Author Response

In this paper, the authors analyzed the performance of the five-constellation PPP technique to monitor SLDW and SVDW by using semi-generated measurements and real GNSS data in different making conditions and the Asia-Pacific mid-low-latitude regions. The authors have done a lot of work, but there are still few problems as follows,

1. Figures 8-18: Make the supplement of the ‘E/N/U’ tag on the legend or axis titles in the corresponding figures.

Response: 

  According to the comment, we had corrected figures 8-13 in the previous manuscript and added E, N, U tags on them. Please see new figures 9-14 in the revised manuscript. The other figures are not changed because they look like so busy if we add ENU tags on them. Thanks for the consideration.

2. Figure 10: Correct the legend content (A = 0.5 cm, F = 0.5 Hz) in the figure.

Response: 

   Thanks a lot. We had corrected the typo.

 

Reviewer 2 Report

My comments for Authors:

1. abstract, all acronyms must be explained, e.g. GPS/Galileo/GLONASS/BDS/QZSS

2. abstract, please add information about obtained results from research test

3.  main body of paper,  please carefully check if all acronyms are explained in text if you used their first time.

4. Figure 3 should be improved, the quality is poor.

5. line 235-236, please add information about height of station.

6. Figure 5 should be improved, the quality is poor.

7. discussion,  in this part of paper must included the comparison between the obtained results and results from publications from bibliography. Please write why your solution is better than results from another publication, maybe from another measurement technique, etc.

8. please add chapter with Conclusions.

9. line 184, it should be ECEF frame, not system

10. Figure 2 should be improved, the quality is poor.

Author Response

My comments for Authors:

1. abstract, all acronyms must be explained, e.g. GPS/Galileo/GLONASS/BDS/QZSS

Response: 

  Thanks. We had corrected all the occurrences in the abstract.

2. abstract, please add information about obtained results from research test

Response: 

  We had added one sentence to simply describe the test result of the PCPPP model. Please see line: 22-24.

3. main body of paper, please carefully check if all acronyms are explained in text if you used their first time.

Response: 

  Thanks. We had corrected all the occurrences in the main body.

4. Figure 3 should be improved, the quality is poor.

Response: 

  The resolution and the font size of the Figure 3 in the previous manuscript had been enhanced. Please see the new Figure 4 in the revised manuscript.

5. line 235-236, please add information about height of station.

Response: 

  Thanks for the comment. The coordinate of the NCCU station is recorded at the antenna reference point. Therefore, no height is recorded in the RINEX files provided officially. Also, we had added the picture of NCCU station, as shown in Figure 1 in the revised manuscript.

6. Figure 5 should be improved, the quality is poor.

Response: 

  The resolution and the font size of the Figure 5 in the previous manuscript had been enhanced. Please see the new Figure 6 in the revised manuscript.

7. discussion,  in this part of paper must included the comparison between the obtained results and results from publications from bibliography. Please write why your solution is better than results from another publication, maybe from another measurement technique, etc. 8. please add chapter with Conclusions.

Response: 

  Thanks. We had added a new section Discussion, in which we had compared the results with some currents studies and other sensors. Please see lines: 562-572. The section Discussion in the previous manuscript had been changed as the section Conclusions. Please see lines: 574-607.

9. line 184, it should be ECEF frame, not system

Response: 

  Thanks. We had corrected all the occurrences.

10. Figure 2 should be improved, the quality is poor.

Response: 

  The resolution and the font size of the Figure 2 in the previous manuscript had been enhanced. Please see the new Figure 3 in the revised manuscript.

 

 

Reviewer 3 Report

Comments for remotesensing-2153630

 

Performance analyses of five-constellation precise point positioning technique for monitoring structural long-term and vibrational displacements on a wall

 

This paper focuses on monitoring structural long-term and vibrational displacements on a wall by using five-constellation PPP technique.

 

The manuscript is well organized and presented. The Figures and Tables give pertinent information. Some details need to be clarified in the manuscript to help better understand the results. In the current form, minor revision of the manuscript is needed in order to better clarify their contribution.

 

1.     Are BDS-2 or BDS-3 satellites is adopted in the processing, why B1I and B2b channels are adopted in the processing, not other combinations, such as B1I and B3I?

2.     In Figure 3, the visible numbers of satellites for each epoch and each constellation are better to be added in the manuscript at different cutoff elevation angle.

3.     In Line 275, how do you define the coordinate system origin when transforming to local system (N, E, U)?

4.     In this research, the performance of SLDW and SVDW is analyzed by comparing the reference results, which are also calculated by GNSS measurements. It is better to compare them with the reference results from non-GNSS equipment.

Author Response

1. Are BDS-2 or BDS-3 satellites is adopted in the processing, why B1I and B2b channels are adopted in the processing, not other combinations, such as B1I and B3I?

Response: 

  Of course, B1I and B3I also can be used for the PPP techniques and the results generally are the same as those of the combination of B1I and B2b, as long as the related error process are well done, such as the error propagation, phase-center offset correction and so on.

2. In Figure 3, the visible numbers of satellites for each epoch and each constellation are better to be added in the manuscript at different cutoff elevation angle.

Response: 

  Thanks. Considering the reviewer’s comment and the conciseness of the figure, we had added the average number of visible satellites for each elevation cutoff. Please see the new Figure 4 in the revised manuscript.

3. In Line 275, how do you define the coordinate system origin when transforming to local system (N, E, U)?

Response: 

  The origin of the local system refers to the point of observation (i.e., the observation station). In the revised manuscript, we had added a reference [32] and one sentence to describe the origin. Please see lines:195-196.

4. In this research, the performance of SLDW and SVDW is analyzed by comparing the reference results, which are also calculated by GNSS measurements. It is better to compare them with the reference results from non-GNSS equipment.

Response: 

  Thanks. We had added a new section “Discussion” to talk about the comparisons with other sensors. In the Discussion, we state that five-constellation PPP is an effective means for monitoring displacements on the wall of high-rise structures because the PPP has its own advantages that other sensors cannot support. Please see lines: 562-572.

 

 

 

 

Reviewer 4 Report

 

Performance analyses of five-constellation precise point ….

I would recommend major revision for this manuscript, considering the fact I was unable to analyze the numerical results because; the theoretical part of the manuscript was not clear to me. I hope given comments help authors to improve this manuscript. Please find below my comments:

 

Minor comments:

 

Considering that English is not my mother language, following comments may improve this manuscript:

1- line 39: commonly behaviors sounds wrong. I guess it should be common behaviors.

2- line 47: The sentence sounds unclear to me. Perhaps by the terms uniformly separated in this sentence and evenly separated in lines 49 and 50, authors are talking about the even or uniform distribution of deformations in different parts of a structure.

3- line 124: The term “synthetic bias” is not common in GNSS literature. Please replace it with a common one.

4- line 131: Preposition “of” is clearly missing: vector L is composed of BSSD measurements ….

5- line 143, 162: Please change “designed” by “design”.

6- line 167: change the term “transit matrix” to “transition matrix”

7- line 167: change “composed of” to “is composed of”

8- line 171: change “ is given with a diagonal” to “is a diagonal”

9- lines 218-219 & 220: The term“local level system” is not common in Geodesy. Please replace it with the “local geodetic system”. As you might know, local level system is a local coordinate system for the attitude determination. The local level system has nothing to do with the Earth’s gravity field but the local geodetic system is based on the gravity field of the Earth. Therefore, it is important to clearly distinguish between the two coordinate systems.

 

Major comments:

1- As authors indicate in the first section of their manuscript, performance analysis of the precise point positioning using all GNSS constellations is not new. According to authors, their contribution to the safety analysis of buildings using a constrained PPP (i.e. PCPPP) model. This is not seen in the title of this manuscript. Therefore, I believe that the title in its current form does not fully represent the scope of this manuscript.

 

2- line 41: It is a pretty well known fact that the displacements are small and cannot be inspected visually. This is why, monitoring systems are mounted on the structures whose safe operation is crucial. Therefore, I would recommend to remove this sentence.

 

3- line 76 is quite obvious and its inclusion within the manuscript adds nothing to this research. I strongly recommend to delete this sentence.

 

4- lines 87 to 89: Structures may be considered as a rigid body as along as the relative deformations of different parts are negligible. Since rigid body deformations, i.e. the movements of a structure as a single monument, does not change the expected or design state of stress in a structure; they are not required in SHD analysis. Therefore, deformation monitoring systems are used wherever the relative changes of deformations may be crucial to the safe operation of a structure. For example, for monitoring the vertical deformations in the turbine hall of a nuclear power plant; a wall point is mount on each side of a joint. Considering this, I recommend you to revise these sentences.

 

5- line 99: Application of wall points (points mounted on walls) is common for analyzing the deformations of high rise buildings. This sentence does not provide any kind of details concerning the measurements made for this research.

 

6- Equations 1 & 2: I expect to see inspected of in these equations but, authors insist on the later case as indicated in sentence 122.

 

7- Dots used in Equation 4 are not necessary. In linear algebra and matrix notation dot is used for inner product only but here, we have both matrix and scalar products. Please also consider this comment in the other equations as well.

8- The considering the matrix , state matrix misses the identity matrix for the applied constraints . The correct form of this matrix is:

 

 

 

 

9- It is not clear to me where the values given in lines 175 and 177 are coming from and why they are required.

 

10- Given information in lines 184 to 191 look trivial for all readers therefore, I would recommend either to delete or to move them to an appendix.

 

11- When the transition matrix is the identity matrix, Kalman Filter is nothing else but the recursive adjustment and the recursive adjustment is a well known technique to the geodesits. Taking this into account, to me it is not clear why I should consider the PCPPP as a new idea or your innovation in this research.

 

12- The mathematical model 23 looks quite arbitrary. Obviously, by changing the signal parameters, i.e. the frequency or period, amplitude and phase of this signal, results are changed. Therefore at least two questions remain:

12-1- What are the physical evidences which justify this specific kind of model for short term vibrations. Is there any relationship between this equation and the vibrations that are imposed by the P and S wave that are generated by an earthquake?

12-2- Why do you consider the same magnitude for the amplitude of vibrations on all directions in a local frame?

 

 

 

 

 

 

Author Response

Performance analyses of five-constellation precise point ….

I would recommend major revision for this manuscript, considering the fact I was unable to analyze the numerical results because; the theoretical part of the manuscript was not clear to me. I hope given comments help authors to improve this manuscript. Please find below my comments:

 

Minor comments:

 

Considering that English is not my mother language, following comments may improve this manuscript:

1- line 39: commonly behaviors sounds wrong. I guess it should be common behaviors.

Response: 

Thanks. We had corrected “behaviour” to “behavior” for all the occurrences.

 

2- line 47: The sentence sounds unclear to me. Perhaps by the terms uniformly separated in this sentence and evenly separated in lines 49 and 50, authors are talking about the even or uniform distribution of deformations in different parts of a structure.

Response: 

To make it clear, we only use the word “uniformly”. Thanks.

 

3- line 124: The term “synthetic bias” is not common in GNSS literature. Please replace it with a common one.

Response: 

To make it more common. We had changed the name “synthetic bias” to “lumped parameters”, and corrected all the occurrences. It can be found in Teunissen and Montenbruck (2017), i.e. [26] in the manuscript.

 

4- line 131: Preposition “of” is clearly missing: vector L is composed of BSSD measurements ….

Response: 

Thanks. We had corrected the typo.

5- line 143, 162: Please change “designed” by “design”.

Response: 

Thanks. We had corrected all the occurrences.

6- line 167: change the term “transit matrix” to “transition matrix”

Response: 

We had corrected the usage according to the reviewer.

 

7- line 167: change “composed of” to “is composed of”

Response: 

We had corrected the typo according to the reviewer.

 

8- line 171: change “ is given with a diagonal” to “is a diagonal”

Response: 

We had changed it according to the reviewer.

 

9- lines 218-219 & 220: The term“local level system” is not common in Geodesy. Please replace it with the “local geodetic system”. As you might know, local level system is a local coordinate system for the attitude determination. The local level system has nothing to do with the Earth’s gravity field but the local geodetic system is based on the gravity field of the Earth. Therefore, it is important to clearly distinguish between the two coordinate systems.

Response: 

Thanks. We had changed it to “local geodetic system” for all the occurrences.

 

 

Major comments:

1- As authors indicate in the first section of their manuscript, performance analysis of the precise point positioning using all GNSS constellations is not new. According to authors, their contribution to the safety analysis of buildings using a constrained PPP (i.e. PCPPP) model. This is not seen in the title of this manuscript. Therefore, I believe that the title in its current form does not fully represent the scope of this manuscript.

Response: 

Thanks for the good comment. We had changed our title as “Monitoring structural displacements on a wall with five-constellation precise point positioning: a position-constrained method and the performance analyses” to enhance the contribution.

 

2- line 41: It is a pretty well known fact that the displacements are small and cannot be inspected visually. This is why, monitoring systems are mounted on the structures whose safe operation is crucial. Therefore, I would recommend to remove this sentence.

Response: 

We had removed the redundant words according to the comment in the sentence. Please see lines 41-43.

 

3- line 76 is quite obvious and its inclusion within the manuscript adds nothing to this research. I strongly recommend to delete this sentence.

Response: 

We had removed the sentence "The GNSS is composed of constellations developed by respective countries" in the revised manuscript. 

 

4- lines 87 to 89: Structures may be considered as a rigid body as along as the relative deformations of different parts are negligible. Since rigid body deformations, i.e. the movements of a structure as a single monument, does not change the expected or design state of stress in a structure; they are not required in SHD analysis. Therefore, deformation monitoring systems are used wherever the relative changes of deformations may be crucial to the safe operation of a structure. For example, for monitoring the vertical deformations in the turbine hall of a nuclear power plant; a wall point is mount on each side of a joint. Considering this, I recommend you to revise these sentences.

Response: 

Thanks for the valuable comments. Indeed, some applications assume that the structure is a rigid body. Therefore, we had added an assumption in the sentence. Please see lines: 89-91.

 

5- line 99: Application of wall points (points mounted on walls) is common for analyzing the deformations of high rise buildings. This sentence does not provide any kind of details concerning the measurements made for this research.

Response: 

Thanks, we had rephrased the sentence. In the sentence, we simply added the way how we make measurements in the analysis. Please see lines: 100-102.

 

6- Equations 1 & 2: I expect to see inspected of in these equations but, authors insist on the later case as indicated in sentence 122.

Response: 

Sorry, I cannot clearly know this question. Probably, we can improve the question in the second round. However, we had carefully checked the PPP equations in the revised manuscript, not to make any mistake or typo. Thanks.

 

7- Dots used in Equation 4 are not necessary. In linear algebra and matrix notation dot is used for inner product only but here, we have both matrix and scalar products. Please also consider this comment in the other equations as well.

Response: 

We had corrected all the occurrences. Thanks for pointing out our mistakes.

 

8- The considering the matrix , state matrix misses the identity matrix for the applied constraints . The correct form of this matrix is:

 Response: 

We had added a new equation (12) to describe the design matrix for the PCPPP model in the revised manuscript. Thanks.

 

9- It is not clear to me where the values given in lines 175 and 177 are coming from and why they are required.

 Response: 

Since both the parameter of the ionospheric delays and ZTD generally change slowly in time, the elements for them are in practice given with empirical values. We have added new content and two related references in the revised manuscript. Please see lines: 184-187.

 

 

10- Given information in lines 184 to 191 look trivial for all readers therefore, I would recommend either to delete or to move them to an appendix.

Response: 

We had removed the related equations, and rephrased the content. Please see lines: 194-198. Thanks.

 

11- When the transition matrix is the identity matrix, Kalman Filter is nothing else but the recursive adjustment and the recursive adjustment is a well known technique to the geodesits. Taking this into account, to me it is not clear why I should consider the PCPPP as a new idea or your innovation in this research.

Response: 

The PCPPP model is proposed to decrease the uncertainty of positional parameters (XYZ) in the classical PPP model. To achieve this, the PCPPP model additionally adopts positional pseudo-observations of the wall point in the classical PPP model. As far as we know, the PCPPP model can be a new idea because this kind of idea has not been presented.

  We had rephrased the related content to make the novelty more convincing in the revised manuscript. Please see lines: 148-152.

 

12- The mathematical model 23 looks quite arbitrary. Obviously, by changing the signal parameters, i.e. the frequency or period, amplitude and phase of this signal, results are changed. Therefore at least two questions remain:

12-1- What are the physical evidences which justify this specific kind of model for short term vibrations. Is there any relationship between this equation and the vibrations that are imposed by the P and S wave that are generated by an earthquake?

12-2- Why do you consider the same magnitude for the amplitude of vibrations on all directions in a local frame?

Response: 

Thanks for the comment. In the analysis, the given amplitude and vibrational frequencies are based on general vibrational behaviors of high-rise buildings and bridges (i.e. only affected by surrounding environments, such as wind force, pedestrians, trains and so on). Therefore, earthquakes are excluded. Xiong and Niu (2018) indicated that the amplitudes in the horizontal direction can be as small as a value less than 1 cm and the vibrational frequency by wind force could be as low as 0.2 Hz when monitoring a building. Similarly, Yu et al. (2016) indicated the amplitudes can be less than 1 cm in the vertical direction, but the vibrational frequency could be higher than 1.6 Hz when monitoring a bridge. In case of a train passing by, the vibrational frequency can reach at least 5 Hz on a building (Eitzenberger, 2008). Considering these reference values, in the analysis, the amplitude is given with 0.5 cm (< 1 cm); the vibrational low frequency is given with 0.5 Hz (i.e., approximate to 0.2 Hz); the high frequency is given as 4.5 Hz, which is as high as possible according to the sampling rate of our test GNSS data.

(The above three references are also [16][18][7] in the manuscript)

  Since this analysis attempts to compare the performance of the two PPP models in monitoring SVDW, a simple case, whose amplitude and vibrational frequencies are time-invariant in the test period, is assumed. In order to clearly distinguish the performance of monitoring SVDW in the horizontal and vertical directions, respectively, we gave the same values on the three direction (NEU).

  In the revised manuscript, we had added new sentences and rephrased the related content to supply the use of the semi-generated measurements. Please see lines: 405-411 and 418-422.

Round 2

Reviewer 2 Report

I accept the paper