[WM]: MPEG 1, Audio, Psychoacoustic model 1

["Roger A. Flor 1987-31140" <rflor@eee.upd.edu.ph>]

Hello everybody!

I am trying to implement the audio watermarking scheme proposed by 
Swanson et al (M. D Swanson, B. Zhu, A.H. Tewfik, L. Boney,"Robust audio 
watermarking using perceptual masking",Signal Processing, 
66(1998)337-355). I will be using C.

Some questions regarding the psychoacoustic models:

1. Hann window (Step 1, calculation of the power spectrum)

According to Swanson et al, h(n)=(((8/3)^(1/2))/2)*(1-cos(2*pi*n/N)), 
sampling rate = 32 kHz,  this is in agreement with ISO-IEC 13818-3. 
However, ISO-IEC 11172-3 says that  h(n)=(1/2)*(1-cos(2*pi*n/N)), 
sampling rate = 44.1 kHz. Painter and Spanias also uses this value (T. 
Painter and A Spanias, "Perceptual coding of digital audio", Proceedings 
of the IEEE, April 2000, p462). Painter and Spanias also mentioned that 
the hann window should have a 1/16 overlap but Swanson et al and the ISO 
standards did not mention about the overlap. Why the differences?



Thanks


Roger A. Flor
DSP Lab
Department of Electrical and Electronics Engg
University of the Philippines    


______________________________________________________________________________

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