extraction part of 'the algorithm. Host audio sequence. is
analysed in time domain and a minimum or a maximum
amplitude is determined at each block of 7.6ms in order to use
forward and backward masking properties of HAS.and place
the watermark amplitude below the masking curve. The
shaping filter filters the pn-sequence in order to adjust it to
masking curves of the HAS in frequency domain. The main
goal is to adapt the PN-sequence to such form that the energy
of the watemiark ,is maximized under the restriction of
keeping auditory distortions to a minimum. The frequency
characteristic of the filter is the approximation of the
threshold in quiet curve of the HAS. Despite the simplicity of
the shaping process of the PN-sequence in frequency domain,
the result is a perceptually-tuned watermark as the largest
amounts of the shaped watermark's power are concentrated in
the frequency sub-bands with lower HAS sensitivity