Dictionary module and UDC: Two new approaches to enhance embedding capacity of a steganographic channel

Abstract

In this paper we present a new technique to enhance embedding capacity of a steganographic channel by preprocessing the secret data. Here we limit ourselves to textual data only. By preprocessing we don't mean compression; secret data can be compressed after it has been preprocessed by our technique. We introduce the concept of Scrambled Letters, User Defined Codes and the Dictionary Module to explain our technique. All these concepts when applied together give a phenomenal embedding capacity. Theoretical results show that we can achieve at least 25-30% increase in embedding capacity. We introduce two techniques, one in which we only apply the dictionary concept while the other in which we use the user defined codes along with the dictionary approach. The former technique is generic and can be applied to any form of textual data, whereas the latter can only be applied to pure text, i.e. without any form of graphics like images or graphs. Once the data is preprocessed by using our technique it can be easily embedded in any steganographic cover medium by using any steganographic algorithm.