A Low-throughput Wavelet-based Steganography Audio Scheme

Steganography literally means "covered writing", which is derived from the Greek στεγοσ (roof) and γραπηια (writing). Cryptography, watermarking and stenography are closely related [Johnson and Jajodia 1998]. This paper introduces the basis of a wavelet-based low-throughput secret key steganography system that requires the exchange of a secret key (stego-key) prior to communication. We try to take apart the fact that the message can only be read with a secret key by using a standard cryptosystem [Schneier 1996] before hiding a text in an audio file. An interesting recent survey lists about 32 different audio stego-tools (.wav 50%, .mp3 28%, midi 6%, others 16%) [Dittmann and Kraetzer 2006]. Possible strategies for embedding data inside the host audio include the use of frequencies inaudible to humans [Gopalan and Wenndt 2004], embedding data using the LSB [Dumitrescu et al. 2003], transform embedding techniques [Lin and Delp 1999]. This paper presents the preliminary of a novel scheme of steganography, and introduces the idea of combining two secret keys in the operation (Figure 1). The first secret key encrypts the text using a standard cryptographic scheme (e.g. IDEA, SAFER+, etc.) prior to the wavelet audio decomposition [de Oliveira 2007]. The way in which the cipher text is embedded in the file requires another key, namely a stego-key, which is associated with features of the audio wavelet analysis.