Mathematical model of steganography using suboptimal decisions in data compression algorithms

Abstract

Background. Protecting access to information in the digital age requires the development of digital tools to encrypt and hide the information from everyone who is not meant to be able to access it. While encryption is great at preventing unauthorized people from accessing the information, it lets people know that there is something hidden behind the transformation. Steganography instead allows us to obscure the fact of information concealment in the first place. Unfortunately, common types of steganography rely on altering the source signal, leaving subtle footprints of data manipulation that can be traced and detected. The goal of this research paper is to provide a model that has the potential of preserving the source signal in its entirety, instead relying on changing the way the source signal is represented in digital media in a way that allows us to encode secret data into the output stream.
Methods. A theoretical analysis of steganography approaches on data compression algorithms was conducted. Methods of preserving source data stream were investigated.
Results. A new model has been developed that uses decision-making processes in data compression algorithms to encode steganographic data in a resulting data stream. In lossless data compression schemes, it is possible to achieve perfect reproduction of source data stream, making detection through analysis of underlying signal encoded in data compression algorithms useless.
Conclusions. The need for information security has been increasing over time, with tensions between countries resulting in new armed conflicts around the world. The ability to embed and covertly send data through steganography may provide a competitive economic, political, and/or military edge. The developed model can be applied to further develop specific methods of steganographic data encoding that is resilient to analysis and detection by existing approaches that rely on statistical analysis of underlying signal stream.