Channel Capacity Analysis of Spread Spectrum Audio Watermarking for Noisy Environments

In this paper channel capacity analysis of spread spectrum audio watermarking (SSW) system for transmission of hidden information through a noisy environment is performed. Channel capacity analysis of algorithm enables us to determine the most reliable transmission rate of hidden information by the watermarking system, i.e. the analysis makes a relation between capacity and robustness as two conflicting features of watermarking algorithms. Examining transmission and detection of hidden information shows that SSW behaves like a binary symmetric channel. It is proved that well known Shannon channel capacity formula for calculating channel capacity is not applicable for SSW system. Instead, analysis of channel capacity for additive color noise channels should be used. Obtaining channel capacity for SSW system using pseudo-noise random sequences with different period, different power of additive white Gaussian noise channel and different power of host signal is performed in simulation results.