Detection of Malicious Primary User Emulation Based on a Support Vector Machine for a Mobile Cognitive Radio Network Using Software-Defined Radio

Round 1

Reviewer 1 Report

In this paper, the authors presented a hybrid method for the detection of malicious primary user emulation. Although authors gave many figures and tables, this manuscript should be revised significantly before potential publication in the journal.

The following comments and suggestions are expected to be helpful for improving this manuscript.

1. It looks more concise and clear if merge the figure 2 and figure 3 into one figure. Figure 5 need to be modified to exhibit signal transmission and the relationship between different parts.

2. Please provide the structure of the support vector machine of the detection model.

3. In the section of results, the results of the test are given as a figure (e.g., figure 7, figure 8, figure 9 and figure 10), but the manuscript provides an inadequate discussion of these results. It’s difficult to understand the necessity and significance of these data. Please specify the meaning of these results and prove the advantages of the detection model.

4. In line 404 to 405, the detection accuracy was compared with reference [27], [32-34], but only for -10dB. Please give the detection accuracy of reference [27], [32-34] in this manuscript to show their difference, and compare the results under more conditions. 5. The literature review on SVM is not up-to-date. Please refer to the following: Online Ash Fouling Prediction for Boiler Heating Surfaces based on Wavelet Analysis and Support Vector Regression

Author Response

Dear Reviewer 1:

Thanks for your revision, comments, and suggestions to our paper. We expect to cover it in this version of the document. We made a point by point revision and include the second version of the document and the response to your comments in the attachment PDF.

Best regards,

Ernesto Cadena Muñoz

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper provided very interesting results relating detection in the CRN. I would recommend for acceptance with minor corrections. Following points required attention:

1. What is meaning here ?
1. We propose a Support Vector Machine 16 (SVM) based technique, which separates no linear signals of signal to noise ratio (SNR)
2. Figure 1: define CBS
3. Provide the performance (e.g. sensitivity, NF and IDR) of the SDR unit in order to measure correct received power. Following recent paper will be helpful to get an understanding on this:
1. Cheema, A.A.; Salous, S. Spectrum Occupancy Measurements and Analysis in 2.4 GHz WLAN. Electronics 2019, 8, 1011.
4. Provide an equation and explanation to calculate the PD for the proposed model.

 

Author Response

Dear Reviewer 2:

Thanks for your revision, comments, and suggestions to our paper. We expect to cover it in this version of the document. We made a point by point revision and include the second version of the document and the response to your comments in the attachment PDF.

Best regards,

Ernesto Cadena Muñoz

Author Response File: Author Response.pdf