A Further Study of Optimal Matrix Construction for Matrix Embedding Steganography

Abstract

Matrix embedding is a general approach that can be applied to most steganographic schemes to improve their embedding efficiency. In order to apply matrix embedding to voice-over-IP (VoIP) steganography better, this paper analyses the means to realizing fast matrix embedding. For small payloads, we discuss the feasibility of combining several Hamming codes into parity check matrix (PCM) construction and propose a novel PCM structure. On this basis, a corresponding optimization algorithm is proposed. It can adaptively generate specific PCMs to accommodate to the given cover and provide the best performance while guaranteeing the allowable computational complexity. For large payloads, another PCM structure is presented by combining the PCM of syndrome trellis codes (STCs) and several referential columns. The corresponding optimal construction algorithm is also given. Experimental results show that compared with existing methods, two novel matrix embedding methods achieve higher embedding efficiency and faster embedding speed.