NORMALS: Normal Linguistic Steganography Methodology

Text-cover of contemporary linguistic steganography approaches has numerous flaws such as incorrect syntax, lexicon, rhetoric, and grammar. Additionally, the content of text-cover is often meaningless and semantically incoherent. Such detectable noise (flaws) by both human and machine examinations can easily raise suspicion. These deficiencies render contemporary approaches highly vulnerable. Unlike all other approaches, the Normal Linguistic Steganography Methodology (NORMALS) neither generates noise nor uses noisy text to camouflage data. NORMALS employs Natural Language Generation (NLG) techniques to generate noiseless (flawless) and legitimate text-cover by manipulating the inputs’ parameters of NLG system in order to camouflage data in the generated text. As a result, NORMALS is capable of fooling both human and machine examinations. Unlike Matlist, NORMALS is capable of handling non-random series domains. The implementation, validation, and experimental results of the NORMALS methodology are demonstrated in this paper.