Audio Stego Intrusion Detection System through Hybrid Neural Tree Model

Introduction

The aggressive environment, in which the electronic connection happening between two parties becomes susceptible, has led to a keen awareness to insist on communication security in networks. The emerging possibilities of contemporary hostilities by the adversaries have forced the use of new methodologies to protect resources and data from vulnerabilities, disclosures and also to safeguard the systems from network based attacks (Katzenbeisser et al., 2000). The significance of information confidentiality and integrity has led to an explosive growth in the domain of information hiding.

The two important aspects of information security are - Cryptography (Katzenbeisser et al., 2000) and steganography (Bender, 1996). Even though cryptography is a fundamental method of securing valuable information by making the message indecipherable to outsiders (Katzenbeisser et al., 2000), steganography is one step ahead by rendering the total communication invisible (Ozer et al., 2006). Steganography is thus the art of hiding the existence of communication by embedding secret messages into naive, innocuous looking cover files, such as digital images, videos, sound files etc. (Ozer et al., 2006). Clearly the purpose of steganography is to evade drawing suspicion towards the transmission of hidden information. Many creative methods have been devised and employed in hiding process that reduces the detectable artifacts of the stego documents. A message could be hidden information in any form – raw text, cipher text, audios, images, i.e., anything that can be represented as bit stream (Katzenbeisser et al., 2000). This message is hidden inside a cover- object to create a stego-object. The process may be represented as:

Just like any other technology, even steganography has been observed to be misused by criminals and anti-social elements across the internet (Westfeld et al., 1999) (Siwei et al., 2006). With the extensive use and profusion of free steganography tools on the internet, law enforcement authorities have serious concerns over the trafficking of superfluous and unsolicited material in the form of web page images, audio and other files. Hence methods which detect such hidden information and understand the overall organization of this technology is vital in revealing these activities. To accomplish the intention of exposing such activities, steganalysis is being employed.

Steganalysis is the art and science of finding the presence of secret messages hidden using steganography (Westfeld et al., 1999). The aim of steganalysis process is to collect ample evidence about the presence of embedded information and to crack the security of its carrier object, thus defeating the primary purpose of steganography. The significance of steganalytic techniques that can reliably find the presence of hidden information in digital media is increasing. Steganalysis is thus fitting itself as an inevitable element into cyber warfare, computer forensics, internet-based criminal activity tracking, and collecting enough evidence for investigations particularly against the anti-social constituents (Westfeld et al., 1999; Cachin, 1998).