HIGH-CAPACITY MULTI-LAYER FRAMEWORK FOR HIGHLY ROBUST

A novel approach has been proposed to improve security and imperceptibility to achieve secure communication. Information hiding has been extensively functional on diverse fields due to enormous use of internet, with support of two major branches, digital watermarking and Steganography. The security issues become prime and necessary concern and energized requirement for better techniques for securing info. In this paper we have proposed a novel text steganography technique by using its technical properties. Proposed technique is storing of text files require less memory and it works faster and makes communication preferable to other mediums of steganography. Paper aim is to develop model to improve capacity, imperceptibility and robustness by multilayer textual steganography with combination of encryption, compression and embedding approach for data obscurity. Various functions are used to encrypt secret word for number of times in our proposed method. Compression is achieved through Lemple-ZIV compression method which as a result reduces size of secret word and embedding is achieved with help of permutation method. Experimental results show that proposed scheme has improved stego features as early research uses less embedding capacity approach which affects other two features. INTRODUCTION In the field of data communication, security issues are the prime concern and therefore are given top priority. Classical cryptography is one of the methods to secure the data over the communication channel and furthermore, security is augmented by introducing the concept of steganography [1]. Steganography is the art of hiding secret information in cover message without getting attention of the third party as shown in Fig 1. Image, audio, video, text and network protocols can be the carrier of secret messages [2]. With the development of computer networks much information has been developed and exchanged on the internet [3]. Since dawn of written communication, people have been concerned to obscuring the content of communication by using cryptography; and obscuring the fact that communication takes place with the help of steganography [4]. Steganography hides the information in such a way that no one apart from the sender and intended receiver, suspects the existences of the message [5]. Stegnalysis is the analysis of the stego text in order to check its robustness and detect/ extract the secret messages [6]. A successful steganographic algorithm requires that steganographically hidden message cannot be detected by the malicious users [7]. Although, all digital file formats are used for steganography, but most suitable are the image and audio files, because of their high degree of redundancy [8-9]. However, text steganography is most difficult approach of information hiding because of low amount of redundancy in its nature [10-11]. Text steganography is broadly divided into two categories, one is technical text steganography and other is linguistic text steganography. Text Steganography: based on following parameters [12].  Robustness: It is measure of strength of the hidden message so that it could not be revealed to an attacker [1314].  Capacity: It measures the amount of data that can be hidden in given cover like picture, sound, and database [1516].  Perceptibility: Advanced content steganography methods are considered recognizable when there is a possibility that hidden message can be perceived by an attacker [17].  Security: When mystery data is imperceptible from the interloper during the transmission of data [18].  Integrity: The legitimacy of a transmitted message [19], course of action that guarantee that the substance of message have not been modified. The content Steganography is craft of concealing the data by utilizing the regular dialect to cover a mystery message as characterized by Chapman at al [20]. Implanting calculations help to encode [21] the message and the resultant message is called stego-message which is transmitted over the correspondence channel [22]. During the transmission it is being observed by the unapproved viewers who will just perceive the transmission of guiltless message without finding the presence of the shrouded message in it [23-24]. Content Steganography is comprehensively characterized into four types [25-26], specialized Steganography, etymological Steganography, irregular & factual and different strategy Steganography [2728]. Emoticon characters can also contain some information; however they have an inherent problem of increasing the size, when used for hiding information [29-30]. Technical steganography includes Word shifting technique, in which the secret message which is to be hidden is shifted horizontally i.e from left or right to represent bit 0 or 1 correspondingly. Line shifting involves hiding the secret message vertically by shifting the lines of the text to some extent. Network steganography includes usage of protocols like TCP/IP [223]. Other methods of text steganography include hiding message in images, audio and video files. Linguistic Steganography includes syntactic and semantic methods of data hiding. Syntactic technique involves using of punctuation marks like full stops, commas to hide the data and semantic method uses similar meaning word to hide text in that particular word without changing the context of the sentence. Random and statistical steganographic methods utilize random sequence of characters to conceal the information [7-9]. Table 1.Types of text Steganography Types Text Steganography Techniques Technical (TS) Word shifting, Line shifting, Feature coding, HTML tag, Image, Audio, Video and Network Steganography Linguistic (TS) Syntactic, Semantic, Abbreviations and Change of spelling Random & Statistical Probabilistic context-free grammar, Character Sequence, Words Sequence and Sequences-text mimicking 4452 ISSN 1013-5316; CODEN: SINTE 8 Sci.Int.(Lahore),28(5),4451-4457,2016 September-Octobe Fig.1 Steganography Process In another technique, statistical properties such as letter frequencies and word length are widely utilized to generate a resultant word which as a result will point to the same statistical properties as depicted by the actual word which is present in the cover source. The existing schemes [1-2,913,23,30] are mostly using single security layer for data transmission, which trim down robustness of the hidden information due to the absence of encryption and permutation. Data hiding using LSB technique is not a recommended option as the hidden data can be extracted by the analysis of LSB. The techniques used in [3-4,21] of text Steganography has drastic effects on imperceptibility, which is quite low than the acceptable level. As a result the pattern of alphabets are changed and statistical detection of the text dig out imperceptibility differences of natural and stego-text. In the schemes [1-2,4-5,9,11,13,16,19,22], the cover message provides small storage capacity. Usage of small scale dictionary and incomplete vocabulary restrict their hidden capacity. The paper is organized as follows: The related work and proposed methodology along with algorithms is elaborated in section 2 and section 3 respectively. Section 4 presents the experimental results and the paper is concluded in section 6. RELATED WORK Less work has been done in textual steganography as compared with other mediums as shown in Table 1 like Image, audio and video [5,7,14]. In [1] Singh et al. have shown a brand new approach on textual content steganography throughout null space inside cover information regarding camouflaging. The secret information is camouflaged via introducing extra bright and null spaces, inside plain textual content of the cover data file. The white and null spaces are adjusted according to the binary of the secret message. It is pertinent to mention that for each 1 and 0 a different white/ null space is used in LSB. Downside of suggested process: hidden text has been hidden in LSB which is more vulnerable to attacks by malicious users. Thamaraiselvan et al. [2,10] have proposed a new process, utilizing the inter-word and also inter sentence spacing with an aim to increase capacity feature of text steganography. Any change in the inter word or inter sentence by the attacker will ruin the hidden text at the decoder end. In [3] Qi et al. have suggested a technique which; uses synonyms of the secret text to be hidden in the cover text. However, an attacker can replace the chosen synonym with another synonym and therefore the hidden text will become meaningless at receiver end. Wang et al. [4] suggested new linguistic steganography scheme that based upon the fact that the hidden word indicates the same meaning as the cover word. Downside is that the hidden information can be tracked through few Natural Language processing NLPbased schemes. Mirielle. Hassan [5] Shirali-Shahreza et al.; suggested a brand new approach, which is based on hiding the data Table 2. Techniques Comparison Table Techniques LSB Based