Advances in Text Steganography Theory and Research: A Critical Review and Gaps

Introduction

Steganography is a way in which a secret (private or sensitive) data is concealed into other non-secret text or data at the source, for the purpose of avoiding the tendency of attack from a third party; that the text can only be spotted by its intended recipients and to protect the legitimate data. Later, embedded data is sent to the destination through insecure channels of the Internet; and it is then extracted at the destination.

The data hiding or steganography is a skillful technique of covert communication. (Dhawan, 2020), an unremarkable cover medium is used to disguise the text so as not to doubt it. The use of this technique was witnessed in ancient times. People were using a type of ink to create images on the body and embed the text in it. It could only be understood by the people who believed to be the trusted parties on the other side. With the evolution of computer power, the combination of the Internet and e-processing turned manual Steganography into digital (Anand, 2020).

Steganography is the process of hiding in a trusted carrier of critical data without sharing it with third parties with the knowledge that information exists. Steganography has two forms: one in which the message is securely retrieved and the medium is shielded (reversible). Secondly, the message may be properly retrieved, but the cover medium could be skewed (irreversible) (Douglas, 2018). Cryptography provides the data with confidentiality, although steganography does not reveal that even the message exists. By combining it with cryptography, the power of steganography can be increased (Alajmi, 2020).

The section of Background and Literature survey, discusses the origin of Steganography, contemporary steganographic methods, characteristics of strong Steganography, text steganographic forms, Cryptography and Steganography, and Steganalysis types.

In this paper, Text Steganography is chosen for our analysis because it is the most difficult form of steganography; this is due to the absence of repetitive bit sequences in comparison to the image and sound (Kamaruddin, 2018).

Linguistic steganography is one of the modes of text steganography that this paper focuses more on. It is a collection of techniques that enables any digital information to be concealed within texts based on certain linguistic features (Lingyun Xiang, 2020).

To conceal the secret, the corresponding covert text should not be invisible but grammatically accurate and semantically cohesive. The methods of linguistic steganography use fonts, alphabets, lexicon, word structure, phrase structure, and word order to camouflage the message.

In Format based Steganography methods, the data hiding is done by character space assignment, line shifting, and word shifting, changing font are some sole characteristics (Xiang, 2017).

This paper explores few Steganographic approaches, paper by (Mahato, 2020), proposes a steganography technique by using Microsoft word document to adjust tracking function.

Low-tech steganography operations, which do not use digital transfer (Nag, 2019) use only manual, are the second method talked about. To create and interpret it needs some form of human intelligence. Low-tech methods, therefore, are often difficult to detect using only automated techniques and, indeed, maybe improvisational. These low-tech methods include Semagrams and concealment Ciphers.

Another type of steganography mentioned here in this paper is to use mathematical formulas or a particular number system to convert text into some kind of numbers (Mandal K. K., 2020).

The last method discussed in this paper is based on text steganography within image data (Joshi, 2018). This method is simple and has more capacity for hiding data. Finally, a comparison of text steganography methods concerning to their merits and demerits is given as table format.