Distortionless 1/2 Overlap Windowing in Frequency Domain Anti-Jamming of Satellite Navigation Receivers

Round 1

Reviewer 1 Report

The paper presents interesting research in terms of anti-jamming technology. The authors propose a new method which can effectively solve the signal energy loss. The method is theoretically deduced, and the effectiveness of the proposed method is verified by simulation and measured data

Although the Authors write: “The method proposed in this paper has been widely used in Beidou satellite navigation system and achieved good results”

but for some readers the influence of the proposed method on accuracy Beidou positioning would be very welcome.

The manuscript is well written and should be published.

Author Response

Thanks to the reviewers for acknowledging the work of this paper.

Author Response File: Author Response.docx

Reviewer 2 Report

An interesting paper about improvement of interference mitigation effect on SNR.

Detailed comments:

Line 41: “…whose orbital altitudes exceed 30,000 kilometers…” Ok if we consider EGNOS, WAAS, BEIDOU geostationary satellites and QZSS, but most of GNSS satellites have their altitude between 19 000 km and 23 000 km.

 

 Line 41-45: “30 dB lower than noise”, for a bandwidth of 20 MHz.

 

Line 95: “digital down conversion”, you should add (DDC) which in reference to figure 1.

 

Line 104: “and supress” you should give some example of suppression methods (filtering, etc.)

 

Line 114: “ Therefore, XX proposes…” It’s a typo.

 

Line 135: “Taking the BPSK(10)…” This reference is a GALILEO or Beidou signal. You should specify. Figure 5 is obtained by your own calculation I guess.

 

Figure 7: it not clear. The scale in the inside figures in the figure (which are zoomed I presume?) are not easy to read. Enlarge or make a series of zoomed figures besides.

 

Line 197-205: your proposal is to change the delay windows function. Could you specify between which blocks in figure 6? I suppose it is where it is written “50% overlap”.

 

In (1), what is N? Is it the total number of considered samples for windowing?

 

Line 225: you should add a reference to figure 11.

 

Hanning delay and punctual functions are very close (Figure 10 shows a few 1e-3 of difference). Thus, if we consider the equation of Hanning Lu approach (i.e. (3)), which is a mean of these very close functions, we do not expect a very large difference between them (confirmed by figure 12 and 14). On the zoom view on figure 14, the scale is different from the zoom view on figure 10. But the mean effect is obvious : considering the max : 3e-3 for figure 10 (Hanning delay and punctual) and 15e-4 for figure 14 (Hanning-Lu).

 

Figure 15 is not relevant. If the scale is correct, 1e-14 dB of scale means there is no significant difference in terms of SNR variability between functions delay and their “Lu” version. The variations observe in (c) and (d) are only linked to the way calculation rounding are led. You should only keep (a) and (b) to show that they are the same.

 

Line 273: “experiment”

 

Experiments are carried out on PRN1 of Beidou B3I. You should recall here the main characteristics of this signal (in term of bandwidth). It is of importance since we thus know what proportion of bandwidth is occupied by interference (2 MHz).

 

Line 308-310 : This result is interesting but it is difficult to make the link with the previous section that explains the 0.5 dB difference. It is not clear in the text (even if I guess) if you compare the Hanning-Lu and the standard Hanning approach or the Hanning-Lu and the Hanning Distortionless Delay Window.

General comments:

We do not know which suppression algorithm is used (but I could have missed it). You should specify it.

The Hamming Lu shows an improvement of 2 dB with your method (better than 0.5 dB of Hanning Lu). You should recall this in your conclusion.

Which is more questionable is what happen for lower JNR with Hanning approch. If I consider figure 19, We observe that for JNR = 40 dB and 50 dB, the Hanning-Lu is underperforming of a few tenth of dB. Which means that Hanning-Lu is recommended for high interference level situation. You should add a word about this.

Author Response

Thanks to the reviewers for acknowledging the work of this paper. 

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript is devoted to the actual problem of GNSS electromagnetic compatibility and contains original and interesting results. Unfortunately the structure and quality of the manuscript is low and need to be rebuilt significantly. Hence I recommend major revision and re-submission. Hope that my comment below will help to the authors make the manuscript better. Attached please see the file with comments.

Major remarks:

• The introduction section reviews mix of different problems such as electromagnetic compatibility, GNSS noise resistance and radio-link power improvement. However, the exact motivation of the paper is unclear. Indeed, each of the above mentioned problems require particular solutions. For example, new design of the windowing function reduces the spectral power leakage and electromagnetic compatibility of course. But I am not sure if this is good solution to improve GNSS noise resistance or radio-link power. Which exact problem is in the focus of the research? Why do the authors think that their solution is the best for the particular problem? What is the area to apply their solution in the best manner? I would recommend to the authors specify the Introduction section much more, providing the exact problem in GNSS, application area of their solution, comparison of the existed methods to solve the GNSS problem et al.
• Sub-section 2.1 needs to be rewritten totally and join with sub-section 2.2 taking in account following:

1. 1 explains nothing: “Anti-jamming” block does not exist in the real receiver as a separate part but realizes as a part of IF filtering procedure; “Tracking” block provides carrier phase, code delay and Doppler frequency but not PNT information;
2. Fig 2 does not explain “basic principle of frequency domain anti-jamming”. I would recommend to the authors to replace it with the math model and explanations defining the algorithm of the anti-jamming.

• Lines 112-113: Please explain how “the spectrum leakage is related to the … interference intensity”? Do these things relate one to another?
• Line 120, Fig 2-4: I do not understand the logic of this text and what it proves. Please rebuild it taking in account following:

1. What is efficiency criterion to choose the windowing function for the authors goals?
2. What is the Fig 4 destiny? Why do you demonstrate Hanning window but not another ones?
3. Pseudo-Random Noise Code parameters and structure (BOC, BSPK, Alt-BOC et al) brings much more improvement in carrier-to-noise ratio (CNR) in comparison to the windowing procedure changing. The authors need to demonstrate the CNR improvement thanks to optimizing the windoving function concerning to a particular RPN code but not to the “single carrier”. It was done in Fig 5 for BPSK but what is the destiny of Figs 2-4 then?

• Fig 6 and 7 needs to be defined, described and discussed. Especially interested to understand:

1. what is the “Interference suppression algorithm”.
2. What conclusions follow from Fig 7, which window you choose for further analysis and why?

Minor remarks:

• Abstract: Please replace “navigation signal signal-to-noise” with “signal-to-noise”;
• Abstract: Please replace “anti-jamming processing process” with “anti-jamming processing”;
• Line 76: Something wrong with this sentence: “theoretically designs a window function design method without loss…”. Please rewrite it clearly.
• Line 93: “FPGA and DSP …” Please decode all the abbreviations here and after;
• Line 96: “other processing, and finally output PNT in”
• Lines 114-115: Please correct it “Therefore, XX proposes …”
• Line 145: Please rewrite this “jamming in the frequency domain of satellite navigation”. I do not understand what it means
• Fig 7, 8b, 10, 13 and 14: these are just mix of pictures. Nothing is clear. Please re-draw each picture in several panels;
• Probably sub-section 3.1 and 3.2 are better to join in the entire section?
• Line 171: what does in mean “The effect of ratio…”?
• Please check and correct English. There are a lot of grammar mistakes and unclear sentences in the text.

Comments for author File: Comments.pdf

Author Response

Thanks to the reviewers for acknowledging the work of this paper. 

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

I am very appreciate authors for their step-by-step comments on the reviewer's remarks.

Unfortunately I do not see almost any improvement in the modified manuscript version (exluding some figures corrected).

As the authors answer to reviewer's remarks and provide their comments more for the readerf than for a reviewer, I strongly recommend the authors to include their comments into the correspondent text of the manuscript. I do not think it is so difficult to insert the ready comments into the text correspondently.

Author Response

Dear reviewer,

    The submitted manuscript adopts the “revision mode” according to the requirement of "Remote Sensing" Journal. Firstly, open the reviewed word file, click the drop-down menu in the "Review" column, and enter "Revision".

    If the manuscript still need to be modified, please contact us.

 I am sincerely hoping for your help and reply.

 

Thank you and best regards.

 

Yours sincerely,

Zukun Lu

 

Author Response File: Author Response.docx