An Effective Watermarking Method Based on Energy Averaging in Audio Signals

Table 4

The low BER illustrating the effectiveness of the watermarking method with error correcting code (ECC) under signal processing attack.

Music Type	Classical piano music		Vocal music		Symphonic music		Pop music
Attack	Algorithm
	w/o	w/cc	w/o	w/cc	w/o	w/cc	w/o	w/cc
MP3 compression; (128 kbits/s)	0.19	0.05	0.21	0.04	0.18	0.05	0.21	0.03
Low-pass filter; (4 kHz)	0.06	0.06	0.22	0.21	0.09	0.08	0.26	0.26
Time-scaling; (preserve pitch)	0.21	0.13	0.26	0.19	0.22	0.13	0.28	0.18
Time-scaling; (preserve tempo)	0.13	0.07	0.18	0.12	0.14	0.07	0.17	0.11
Time-scaling; (preserve both)	0.21	0.16	0.31	0.21	0.22	0.16	0.25	0.22
Resampling; (44.1 kHz to 22.05 kHz)	0.12	0.06	0.04	0.02	0.13	0.07	0.05	0.03
Resampling; (44.1 kHz to 11.025 kHz)	0.19	0.11	0.07	0.03	0.21	0.13	0.08	0.05

w/o: audio watermarking with energy averaging only.
w/cc: audio watermarking with energy averaging and ECC (cyclic code).