Details of Research Outputs

Although steganographic transparency and embedding capacity are considered to be two conflicting objectives in the design of steganographic systems, it is possible and necessary to strike a good balance between them in Voice-over-IP steganography. In this paper, to improve steganographic transparency while maintaining relatively large embedding capacity, the authors present a (2n-1, 2n) covering code, which can hide 2n-1 bits of secret messages into 2n bits of cover messages with not more than n-bit changed. Specifically, each (2n-1)-bit secret message is first transformed into two 2n-bit candidate codewords. In embedding process, the cover message is replaced with the optimal codeword more similar with it. In this way, the embedding distortion can be largely reduced. The proposed method is evaluated by comparing with existing ones with a large number of ITU-T G.729a encoded speech samples. The experimental results show that the authors' scheme can provide good performance on both steganographic transparency and embedding capacity, and achieve better balance between the two objectives than the existing ones.