Method of digital-audio watermarking based on cochlear delay in sub-bands

摘要

We improved our proposed method of digital-audio watermarking based on cochlear delay (CD) characteristics by applying this method to the sub-bands of the original audio signal. A constant-bandwidth filterbank was used to decompose the original audio signal into sub-band signals. The improved method took advantage of embedding data into sub-band signals to increase embedding capacity as multiples of the number of sub-bands with that of the previous method. We investigated the possibility of inaudible embedding in sub-bands and whether embedded data could survive in sub-bands through a process of analysis and synthesis. We carried out simulations to evaluate the sound quality of watermarked signals and bit-detection rate of watermarks in the improved method. The results revealed that the improved method could greatly increase the capacity of embedded data. The improved method could be efficiently used as a hiding scheme to transmit data in digital audio signals.