kumoh national institute of technology
Networked Systems Lab.

Imperceptible Digital Data-Hiding in Audio File using Novel Rounding Modulation Technique
By : RIZKI
Date : 2019-11-01
Views : 408

Based on the enclosed set of reviews, I regret to inform you that your manuscript has been rejected for publication.

Please note that according to IEEE Signal Processing Society policy "Handling of Rejected Papers" (http://signalprocessingsociety.org/volunteers/policy-and-procedures-manual), the Society strongly discourages resubmission of rejected manuscripts more than once. Authors should carefully review the aforementioned policy before resubmitting their manuscript.

If you have any questions regarding the reviews, please contact me. Any other inquiries should be directed to Kathy Jackson.

Thank you for submitting your work to the IEEE Transactions on Audio, Speech and Language Processing. We hope you consider us again in the future.

Sincerely,

Prof. Stefan Bilbao
Associate Editor
IEEE Transactions on Audio, Speech and Language Processing
sbilbao@ed.ac.uk

Kathy Jackson
Coordinator Society Publications
IEEE Signal Processing Society
k.jackson@ieee.org

Reviewer Comments:

Reviewer: 1

Recommendation: R - Reject

Comments:
The paper has presented an interesting approach for audio data hiding. However, some major issues have made the evaluation very hard. Specifically,
- The paper lacks focus. The term data hiding is a general term that covers several different techniques including steganography, watermarking, and fragile watermarking. It is not clear what is the focus of this work. Does it propose a watermarking algorithm, or a steganography method? This delineation would make a huge difference in terms of the evaluation criteria.
- Aims and significance of the work are not clear. Specifically, what are the problems that the proposed method is trying to address? And why are they important?
- Novelty of the paper is another issue that needs to be addressed. Authors are encouraged to spell out the novelty of their proposed method and how it is different from other works. For example, what are differences between this work and the reference 22, which is by the same authors and on a similar topic?
- The paper lacks a smooth and logical flow. This is in part due to continuously jumping between watermarking and steganography methods.
- Writing of the paper needs major revision and proofreading.
- In the experiment section, very important pieces are missing. This is in part due to a lack of focus. If the method is a watermarking algorithm its robustness to different attacks should be investigated. If it is a steganography its security should be investigated.
- Using a single secret data is not enough and average performance of the system over multiple messages should be reported.
- Specifications of the dataset is missing. How many files? What types of audio files were used? How many speeches and how many music samples were there in your dataset? How were they acquired? What was the history of the files? Were they transcoded from a different format, or were they in the raw format?
- No comparison is done with the best and up to date alternatives!
- Information in table II seems to be incorrect! Assuming audios with 44100 Hz and 16bit resolution, the bit rate of the host would be about 705kbs. How could you hide 5488kbs in a 705kbs host???!!!! This is a direct violation of the information theory!
- The references are not adequate, and the related work section lacks a fair comparison with existing approaches. You may consult with existing great review papers on audio steganography, audio steganalysis, and audio watermarking.

Some minor issues:
- Abstract, The performance of the proposed system has . Often, the term the proposed system refers to the work that you are proposing in the paper. Probably, existing or previous is a better choice.
- Abstract, introductory part of the abstract should be reduced. Instead, state the rational of the project, its significance and then present the method and give some objective numbers showing the performance of the system in comparison to the best existential alternatives.
- The wavelet domain paragraph is repeated twice!
- Note that cosine and sine are also frequency domain transformation. Putting them in the same paragraph as wavelet adds to the confusion of the readers.
- while the robustness of the secret data can be measured from the bit error ratio (BER). Please note that robustness is defined in the presence of active attacks! What you are measuring is the correct recovery rate, not robustness.


Additional Questions:
1. Is the topic appropriate for publication in these transactions?: Poor Match

1. Is the paper technically sound?: No

2. How would you rate the technical novelty of the paper?: Not Novel

Explain: Novelty of the paper is not clear at all. Specifically, this paper seems to have significant overlap with another work published by the same authors. Yet, the differences between these two works are missing in the manuscript.

3. Is the contribution significant?: Incremental

4. Is the coverage of the topic sufficiently comprehensive and balanced?: Important Information is missing or superficially treated

5. Rate the Bibliography: Unsatisfactory

1. How would you rate the overall organization of the paper?: Poor

2. Are the title and abstract satisfactory?: No

3. Is the length of the paper appropriate? If not, recommend what should be added or eliminated.: No (Explain):

4. Are symbols, terms, and concepts adequately defined?: Not always

5. How do you rate the English usage?: Needs improvement


Reviewer: 2

Recommendation: AQ - Publish With Minor, Required Changes

Comments:
The paper proposes a useful extension to increase payload. Authors need to check spelling and syntax, and text formatting. On page 1, line 48, column1, for example, note '... authors theories ..." and on p3, l22, c1, and in p5 ",,, reminder..." also, "miltiple" on p4, l50, c1 On p2, c1, l33 - l42 is a repetition.
Beginning of section III on p2, c2 needs revision.
Eq. 1 needs better explanation.
H and C, and H' and C' -- distinguish.
Reference list needs editing -- Ref. 3 and 5 start with 'and'. Ref. 6 : under mp3. Ref. 8 pp is incorrect.
An example similar to Fig. 1 will be helpful to illustrate H and H'.

Additional Questions:
1. Is the topic appropriate for publication in these transactions?: Poor Match

1. Is the paper technically sound?: Yes

2. How would you rate the technical novelty of the paper?: Novel Enough for Publication

Explain: An extension of a previously published paper by the same authors.

3. Is the contribution significant?: Moderately Significant

4. Is the coverage of the topic sufficiently comprehensive and balanced?: Treatment somewhat unbalanced, but not seriously so

5. Rate the Bibliography: Satisfactory

1. How would you rate the overall organization of the paper?: Could be improved

2. Are the title and abstract satisfactory?: Yes

3. Is the length of the paper appropriate? If not, recommend what should be added or eliminated.: Yes

4. Are symbols, terms, and concepts adequately defined?: Not always

5. How do you rate the English usage?: Needs improvement


Reviewer: 3

Recommendation: R - Reject

Comments:
(There are no comments. Please check to see if comments were included as a file attachment with this e-mail or as an attachment in your Author Center.)

Additional Questions:
1. Is the topic appropriate for publication in these transactions?: Excellent Match

1. Is the paper technically sound?: No

2. How would you rate the technical novelty of the paper?: Not Novel

Explain: I will speak from the point of view of someone who has closely followed the field of digital data hiding almost since its inception in the late 1990s.

The authors propose a seemingly reasonable data hiding method for audio signals. However there is very little novelty in their proposal. Unfortunately, there are precious little signs of engagement with existing data hiding research in the manuscript. The references are either dated or of little relevance.

In essence the authors' proposal boils down to the application of Quantization Index Modulation (QIM) on the phase space of an audio signal. There is nothing new here, but it is significant that the authors do not even mention the pioneering paper by Chen and Wornell on the topic. f they had, they would have become aware that distortion-compensated QIM (DC-QIM) improves on the robustness of QIM, by making the embedding procedure perform closer to the theoretical limits first discovered by Costa in 1983 for channels with side information at the encoder (like the data hiding problem).

The one contribution this paper could have had would have been an analysis of the _perceptibility_ issues associated to their method. However the authors gloss over this important question, especially given the fact that phase information very relevant psychoacoustically. How would a "golden ear" like this method?

3. Is the contribution significant?: Not Significant

4. Is the coverage of the topic sufficiently comprehensive and balanced?: Important Information is missing or superficially treated

5. Rate the Bibliography: Unsatisfactory

1. How would you rate the overall organization of the paper?: Satisfactory

2. Are the title and abstract satisfactory?: Yes

3. Is the length of the paper appropriate? If not, recommend what should be added or eliminated.: Yes

4. Are symbols, terms, and concepts adequately defined?: Not always

5. How do you rate the English usage?: Satisfactory