Intelligent Reception of Frequency Hopping Signals Based on CVDP
Jan Mietzner
Round 1
Reviewer 1 Report
The paper proposes a CNN-based classification technique for detecting the active frequency of a frequency-hopping transmitter within a single time slot. The paper is generally well written and the mathematical description seems rigorous. The presented simulation results - especially the BER results where the receiver performance is close to that with a known hopping sequence - are promising.
In my view, the scope of the paper should be outlined more clearly, however. Throughout, a wideband receiver is employed, which is able to receive all possible hopping frequencies, i.e., the considered problem setup does not involve the design of an "intelligent" narrowband receiver which pursues an optimal placement of its receiving band within the overall frequency range of interest. Also, no effort is made to predict upcoming frequency hops, which might further improve performance (if the frequency pattern is not completely random). Furthermore, the pre-requisites for the CNN-based classification technique should be discussed in detail. For example, what kind of prior knowledge is required (only the number of possible hopping frequencies or the entire set of possible hopping frequencies?). Also, the discussion of the results should be strengthened. For example, what kind of features do CNN-VIT1 and CVDP extract from the received spectrograms, so that a better performance is attained compared to the other algorithms? Even if it is difficult to pin-point, at least some ideas should be shared, why the underlying processing blocks of CNN-VIT1 and CVDP are better suited to address the considered task of determing the hopping frequency based on a single time slot.
A major aspect for criticism is that a proper literature review ist not provided. Instead, a large number of papers is cited, where deep-learning methods are employed in several processing steps within wireless communication receivers, but these paper barely relate to the presented work. On the other hand, highly relevant papers, where frequency-hopping receivers have been optimized using deep-learning methods, are not cited. In fact, only one really related paper is cited ([21]). Yet, there is a considerable body of related literature, e.g.
- Detection of Frequency-Hopping Signals With Deep Learning
Kyung-Gyu Lee; Seong-Jun Oh
IEEE Communications Letters, Year: 2020 | Volume: 24, Issue: 5
- Detection algorithm of frequency hopping signals based on S Transform and Deep Learning
Chun Li; Zhijin Zhao; Ying Chen
2022 16th IEEE International Conference on Signal Processing (ICSP), Year: 2022
- LSTM-based Frequency Hopping Sequence Prediction
Gao Li; Jianliang Xu; Weiguo Shen; Wei Wang; Zitong Liu; Guoru Ding
2020 International Conference on Wireless Communications and Signal Processing (WCSP), Year: 2020
Without a detailed review of these (and other) relevant prior works, the novelty of the submitted paper remains completely unclear, which is unacceptable.
A question which arises is whether the presented work contributes to compromising the "anti-interception" feature of frequency-hopping communication systems. After all, the proposed CNN-based classification technique could also be employed by an adversary. Please include a discussion on this aspect.
Further recommended changes:
- The abstract should be improved. Some introductory remarks on the overall problem setup are required before talking about "the conventional receiving system" and "the intelligent decision of frequency hopping sequence by the transmitter of the frequency hopping system"
- The following phrase from the Introduction does not really make a point: "Therefore, intelligent anti-interference of frequency hopping communication has emerged as a new research direction in the field of anti-interference of frequency hopping communication."
The use of direct and indirect articles can be improved at times (e.g. in the abstract)
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have satisfactorily addressed my earlier comments. The literature review could be further strengthend, though.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
I am satisfied with the author's answers. They prepared the final version of the paper quite clearly and satisfactorily. In this form the manuscript is acceptable for publication.
It is recommended that the authors carefully read the manuscript again and correct the spelling mistakes of some words and check the form of some sentences.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
