Next Article in Journal
Characterizing the Impact of Communication on Cellular and Collective Behavior Using a Three-Dimensional Multiscale Cellular Model
Next Article in Special Issue
Sound Recognition Method of Coal Mine Gas and Coal Dust Explosion Based on GoogLeNet
Previous Article in Journal
Surrogate-Assisted Hybrid Meta-Heuristic Algorithm with an Add-Point Strategy for a Wireless Sensor Network
Previous Article in Special Issue
Underwater Acoustic Target Recognition Based on Attention Residual Network
 
 
Article
Peer-Review Record

Automatic Modulation Classification for Underwater Acoustic Communication Signals Based on Deep Complex Networks

Entropy 2023, 25(2), 318; https://doi.org/10.3390/e25020318
by Xiaohui Yao, Honghui Yang * and Meiping Sheng
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Entropy 2023, 25(2), 318; https://doi.org/10.3390/e25020318
Submission received: 11 January 2023 / Revised: 3 February 2023 / Accepted: 7 February 2023 / Published: 9 February 2023
(This article belongs to the Special Issue Entropy and Information Theory in Acoustics III)

Round 1

Reviewer 1 Report

The authors propose two physical signal processing layers based on DCN, which include deep complex matched filter (DCMF) and deep complex channel equalizer (DCCE). It is well structured and very interesting. However, there are some minor technical defects.

1.      The author needs to further highlight the innovation of the article in the abstract. Why based on DCN? Why include DCMF and DCCE?

2.      In introduction, many features have been mentioned, such as instantaneous statistics features, constellation features, spectrum features but some new nonlinear dynamic features are missing, such as 10.3390/fractalfract6100544 and 10.1016/j.chaos.2022.112400, which are more consistent with the journal Entropy.

3.      To prove the superiority of the proposed method, it is suggested that the author increase comparative experiments from different aspects, such as data and similar methods.

4.      In the conclusion, the author needs to summarize the key points and contributions of the conclusion, and some detailed descriptions about experiments and methods need to be simplified.

5.      Minor grammatical errors need to be eliminated.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this paper, the authors adopted DCN into AMC of underwater acoustic communication signals. DCN has advantages over real-valued DNN for processing complex baseband signals. Since DCN is designed in the rule of complex arithmetic, it is more convenient to construct physical signal processing layer using DCN than real-valued DNN. In order to overcome the influence of underwater acoustic channel and improve the AMC performance, the authors proposed DCMF and DCCE based on DCN and embedded them in DCN to conduct a hierarchical network. Real world underwater acoustic communication scenario was fully considered in this paper. The authors simulated several underwater acoustic channels using real world ocean observation dataset and ambient noise. The paper is interesting and well-written but needs revision:

1. How was the accuracy calculated? This point is not clear.

2. The authors should present the limitations of the proposed approach.

3. The authors should present a comparison with related approaches.

4. The authors use generated signals. What is about results on real signals? This point should be presented.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have solved all my concerns and the article can be published.

Reviewer 2 Report

The authors address all my concerns. The paper can be accepted.

Back to TopTop