A Simple Neural-Network-Based Decoder for Short Binary Linear Block Codes
Round 1
Reviewer 1 Report
This paper describes a method of identifying the correct de-codification of the transmitted information after the message had been contaminated with random noise. It is based on a shallow neural network as a classifier and soft decision decoding. Its performance seems comparable with the OSD method for the study case of BCH, but it is not clear in the case of QR. Further experiments are required to identify the robustness of the technique as well as its limitation and advantage.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
This study focusing on network architecture of multivariate classification. It compare with two different network architectures of binary classification and noise elimination. Overall, the paper is fine. However, there are many issues that must be addressed before the publication.
MAJOR
1. The motivation of this work is missing in the abstract.
2. Contribution is not clear. Add clear contribution at the end of the Section 1.
3. Need more explanation on Section 2.
4. Figures not mentioned in text, also no explanations for that.
5. It is not clear the difference between the Chase II algorithm (Section 2.2.1) and OSD (Section 2.2.2).
6. Need clear justification for the results experimental results for BCH.
MINOR
7. Figures qualities are low. Improve all figures, especially in chapter 6.
8. There are some references are very old. If possible, use latest reference instead/along with that. Add more references for year 2021, 2022 and 2023.
9. Many places the text is ambiguous. E.g. Abstract : Why the 1st line in (In this study) is bold.? Section 2, line 85 , 87 etc.
10. Poor formatting E.g Algorithm 1 (Line 368). MUST Improve overall presentation of manuscript.
11. There are many instances where the language interferes with comprehensibility. The author should proofread and edit their manuscript for clarity.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper presents a network architecture of multivariate classification that can be used to compare the two different network architectures of binary classification and noise elimination. It could be accepted after checking the typos and spelling. Good work.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
A general improvement in the paper. Several typos in the diagrams. References should be reordered in order of appearance. The P meaning is still not defined.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The reviewer would like to thanks authors for correction the manuscripts issues.
This version is much improved and it can be accepted.
Good work and All the best.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
