Exploring Performance of a Spread Spectrum-Based Audio Watermarking System Using Convolutional Coding

This paper presents a spread spectrum-based audio watermarking system, where the message embedding is performed in the time domain. Similarly, as in a communication system, a channel coding is used to improve the performance of the audio watermarking system. Our goal here is to explore the performance using convolutional coding in a Spread Spectrum-based audio watermarking system to get the boundaries and the best configuration in the system. To this end, data payload, imperceptibility, and robustness are tested. Usually the number of bits used for quantization, improves the performance. However, we demonstrated here that the increasing number of bits more than 8 will not affect the performance. This is important result which may result in better use of computational resources.