Perceptual audio watermarking driven by human auditory system

Abstract

The aim of this work is the design of a novel audio watermarking technique based on the Linear Predictive Coding approach and on the psychoacoustic model defined in the MPEG-I standard. The embedding imperceptibility is achieved by exploiting the frequency masking effect. The evaluation of the perceived audio quality is performed by means of the standardized ITU-R BS.1387 quality assessment method. The effectiveness of the proposed method is tested in presence of attacks.