[WM]: Re: What algorithm to use?

["Jae-Won CHO" <cho@creatis.insa-lyon.fr>]

Hi, Nikos Mitropoulos,

You can simulate the 'echo hiding' as a simple audio watermarking.
It embeds watermark two (or more) kinds of echo kernels into host audio signal.

Let's aussme that x(t) is the original audio signal, and, e0 and e1 are echo kernels to embed watermark '0' or '1', respectively.
Note that each echo kernel is just a version of the delayed original signal.

That is, if we want to embed wateramrk '0', 
x'(t) = x(t)*(this is convolution, else is just multiplier)(delta(t)+initial_amplitude*delta(t-d0))=x(t)+initial_amplitude*x(t-d0)
Else if we want to embed watermark '1'
x'(t) = x(t)*(this is convolution, else is just multiplier)(delta(t)+decay_rate*delta(t-d1))=x(t)+initial_amplitude*x(t-d1)
Where, 'initial_amplitude' and 'decay_rate' are some kinds of stength factors of watermarks.

In other words, we embed watermark into host audio data by introducing two another kinds of echo signals.

You can reference a paper, D.Gruhl, and W.Bender, 'Echo hiding', Proceedings of Information Hiding Workshop, Cambridge, UK, Dec.
1996, pp. 295-315.

Note that echo hiding is one of simple and efficient data hiding technique for audio signals.

J.W. CHO

============================================================
International Cooperative Ph.D Student
Between Yeungnam Univ.(KOREA) and INSA de Lyon(FRANCE)
------------------------------------------------------------
# Address : 
(KOREA) MSP Lab. Material Engineering Building Room #213,  
School of Information & Communication Engineering,
Yeungnam University, 214-1, Kyung-San City,  
Kyungbuk, Korea. 712-749
(FRANCE) INSA-Lyon CREATIS CNRS UMR 5515,Inserm U630
Building Blaise Pascal 
20 Avenue Einstein 69621 Villeurbanne cedex FRANCE
# E-mail : ram56@yumail.ac.kr; cho@creatis.insa-lyon.fr
# Phone : 
(KOREA)
- Office : +82-53-810-3939
- CDMA : +82-16-822-1806
(FRANCE)
- GSM : +33-6-27-66-18-88
# CV : http://chunma.yu.ac.kr/~m0456037/
==================================u.ac.kr/~==========================

----- Original Message ----- 
From: "Nikos Mitropoulos" <zeroout2003@yahoo.gr>
To: <watermarking@watermarkingworld.org>
Sent: Tuesday, April 19, 2005 5:25 PM
Subject: [WM]: What algorithm to use?


> Hi all, 
> 
> since i am new to this field i w

> Hi all, 
> 
>ould like you to help me in choosing a suitable algorithm to immplement watermarks for audio files
> (specifically .mp3). 
> 
> i would like to implement an algorithm imp3). 
>n C or Java, but i want it to be easy to understand.
> 
> Any links or papers, will be grately appreciated.
> 
> Greetings... 
> 
>


______________________________________________________________________________

Watermarking Mailing List - http://www.watermarkingworld.org/ml.html
To unsubscribe send email to "majordomo@watermarkingworld.org" with
"unsubscribe watermarking YOURMAIL" in the body.
______________________________________________________________________________