Capital Shape Alphabet Encoding Based Text Steganography

A Capital Shape Alphabet Encoding (CASE) Based Text Steganography Approach

Abstract

Steganography is the art and science of writing hidden messages in such a way that no one, apart from the sender and intended recipient, suspects the existence of the message. Steganography is a form of security through obscurity. Steganography algorithms uses cover media such as text, image, audio and video etc. to hide the data. User relies on change in the structure of these mediums and features of the target medium in such a manner as is not identifiable by human. In this paper we also present and evaluate my contribution to design the new approach for text Steganography and named it as CASE (Capital Alphabet Shape Encoding) approach. This approach is a combination of random character sequence and feature coding method.

Keywords- Steganography; features; encoding; decoding; cover text.

Steganography means conceal communication. It is derived from a work by Johannes Trithemus (1462-1516) titled “Steganographia”. The word Steganography comes from Greek and meaning of steganography is “concealed writing”. Steganography is used to transmit a message through some innocuous carrier i.e. text, image, audio or video over a communication channel in order to effectively conceal the existence of the message. Text steganography is a process to hide the secret information within text (i.e. character based) messages. Text steganography is the most difficult kind of steganography [1]. Text steganography is considered as difficult one is due to the lack of redundant information in a text file, while there is a lot of redundancy in a picture or a sound file, which can be used in steganography [1] [2].

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

For text steganography there are many methods available [3] [4]. Some method change format of text while some method change actual word to hide secret data. White space is used to hide secret data, in open space methods [5], while in syntactic method punctuations are used to hide secret data [5]. In acronyms method [2] and semantic method [4], actual word or phrase replacement is used to hide secret data. By using characteristics of that particular language data is hidden in Persian/Arabic Text steganography [4] and Hindi Text steganography [6].

If we use open space methods [6] or feature coding method [4] in text steganography, and if somehow format of file is changed then it results in data loss. In acronyms and semantic method, meaning of information can be changed because these methods use actual word replacement or punctuation to hide secret data. So a method is needed by which secret data survive after changing format of file and meaning of text will not changed.

So considering some problems like format changing, changing meaning of secret data, etc. in existing text steganography methods; we have proposed new CASE approach for text steganography. We have used text features of English letters to hide secret data. Letters of English alphabet based on kind of round shape or curve, vertical and horizontal line are grouped in first approach. Like some letters in English language contains kind of close round shape or curve. Like some letters in English contains only one straight vertical line. In second approach, we are grouping numbers and symbols. In CASE categorization, based on different text features of English letters we categorize English letters into eight groups. We have studied the implementation some existing methods mentioned in paper [4] and we have compared these methods with proposed CASE approach. In the proposed CASE approach randomness is used but it aids to provide more security to secret information. For various existing methods and proposed CASE approach we have measured number of bytes hide, time overhead and memory overhead. Our results shows that, very less time overhead and memory overhead is required to implement proposed CASE approach compared to existing methods, and also we can hide more number of bytes using proposed approach. Required cover text size is also very small in proposed approach.

In this approach, we introduce new encoding technique to hide the secret message in cover text. We will call this technique Capital Alphabet Shape Encoding (CASE). In this method every character of secrete message is encoded in the form of 8-bit binary number after that the equivalent ASCII character is replace the original character. In this the left most 0th bit will represent the alphabet group or digit/symbols group (0 value for alphabet set and 1 value for digit or symbols).

In case of alphabet we made eight groups of English letters based on features of letters. While making group we consider only Capital letters of English alphabet. The left most 1st, 2nd, 3rd bit of 8 bit number represents the group number. The next 4th bit will represent the sentence case of letter. If it is 1, than it represents upper case letter and if it’s 0 than letter will be lower case letter. Table I, II, III shows eight bit encoding format of alphabets, digits and symbols. In this approach all alphabets are divided into groups and every alphabet has its position in corresponding group. This position will represent the last three bit of 8-bit number. By using this approach, we can hide all eight bits of one letter of secret message into one letter of cover text at a time.

Table1. 8-Bit encoding format in case of alphabet

Table2. 8-Bit encoding format in case of digit

Table3. 8-Bit encoding format in case of symbols

As shown in Table1, in first group, we include neither those letters which have round shape or any curve and nor vertical and horizontal straight line. We can use any letter from this group to hide “000” bit. Candidates for this group are V, W, X, Y. In second group, we include those letters which have one or two vertical straight line. We can use any letter from this group to hide “001” bit. Candidates for this group are K, M and N. In the third group, we include those letters which have only one or more horizontal straight line. We can use any letter from this group to hide “010” bit. Candidates for this group are A and Z. In forth group, we include those letters which have both one and more than one straight vertical and horizontal line. We can use any letter from this group to hide “011” bit. Candidates for this group are E, F, H, I, L and T.

In fifth group, we include those group, we include those letters which have only curve or round shape. We can use any letter from this group to hide “100”bit. Candidates for this group are C, O, Q, S and U. In sixth group, we include those letters which have both curve and straight vertical line. We can use any letter from this group to hide “101”bit. Candidates for this group are B, D, P and R. In seventh group, we include those letters which have curve and straight horizontal line. We can use any letter from this group to hide “110” bit. Candidate for this group is G. In last but not the least eighth group, we include those letters which have curve, and both straight vertical and horizontal line. We can use any letter from this group to hide “111” bit. Candidate for this group is J. For example if the secret letter is H then by using CASE approach it will be encoded as 00111010 and its ASCII equivalent is 58 which is given by ‘:’. After encoding, now letter ‘:’ will be mixed up with the cover text. And this letter will hides all the 8 bits of original letter H into cover text.

2.1. Hiding the Message

In the CASE text stenography approach, first we encode all the characters of the secret message with new proposed encoding technique which is based on the shape of the alphabet characters. Second we hide this message with the cover text by mixing it with the contents of cover text, i.e. first we encode the secret character and then we perform process of finding the ASCII equivalent of the 8-bit format of the secret character and hide it with the contents of cover text.

To hide it or embed it with the cover text, we made a new technique. In this technique we encode the first three letter of the cover text by using CASE approach and then count the bits having value 1. This count value is the key value for hiding the data. After calculating the key value message is mixed up, one character of message comes after key number character of cover text and this process of embedding are repeated until whole message is hidden in the cover text.

For implementation of CASE approach we develop two algorithms for hiding and retrieving which are implemented in JSP using NetBeans IDE.

3.1. Pseudo code for message hiding

Procedure CASE_steno_hide (String msg, String covertext)

begain;

For i=0 to msg.length()

ch=msg(i); encode_msg=encode_msg+char(encode(ch));

End For

key_msg=covertext.substring(0,3);

For i=0 to key_msg.length()

ch=key_msg(i);

en_key=en_key+encode(ch);

End For

keyarray []=en_key.toByteArray();

key=0;

For i=0 to keyarray.length()

IF keyarray[i] == 1 then

key++;

End IF

End For

IF key < 5 then

key=5;

End IF

For i=0,j=0 to i

hidden_msg=hidden_msg+covertext.substring(i,i+k)+encode_msg(j);

i=i+k+1;

j++;

End For

return hidden_msg;

End Procedure

3.2. Pseudo code for message retrive