A Procedure of Text Steganography Using Indian Regional Language

Published Online August 2012 in MECS DOI: 10.5815/ijcnis.2012.08.08

• I.    Introduction

  Information hiding is the ability to prevent or hidden certain aspects from being accessible to others excluding authentic user. It has many sub disciplines. One of the most important sub disciplines is steganography [2] which is derived from a work by Johannes Trithemus (1462-1516) entitled "Steganographia" and comes from the Greek language defined as "covered writing" [3]. It is an ancient art of hiding information in ways a message is hidden in an innocent-looking cover media so that will not arouse an eavesdropper’s suspicion. Steganography diverges from cryptography in the sense that where cryptography focuses on keeping the contents of a

message secret by encryption technique, steganography focuses on keeping the presence of a message secret [20], [21].

Another form of information hiding is digital watermarking, which is the process that embeds data called a watermark, tag or label into a multimedia object such that watermark can be detected or extracted later to make an assertion about the object. The object may be an image, audio, video or text only [22], [23]. A hidden channel could be defined as a communications channel that transfers some kind of information using a method originally not intended to transfer this kind of information. Observers are unaware that a covert message is being communicated. Only the sender and recipient of the message notice it. Steganography works have been carried out on different media like images, video clips, text, music and sound [24], [21].

In Image Steganography method the secret message is embedded into an image as noise to it, which is nearly impossible to differentiate by human eyes [25], [26], [7]. In video steganography, same method may be used to embed a message [6], [5]. Audio steganography embeds the message into a cover audio file as noise at a frequency out of human hearing range [7]. One major category, perhaps the most difficult kind of steganography is text teganography or linguistic steganography because due to the lack of redundant information in a text compared to an image or audio. The text steganography is a method of using written natural language to conceal a secret message as defined by Chapman et al. [4].

Figure 1: Types of Steganography

• A.    Quantum Steganography

Comparatively very little research work has been done in quantum steganography also. The idea of hiding secret messages as the error syndromes of a quantum errorcorrecting code (QECC) was introduced by Julio Gea-Banacloche in [28]. In his work Alice and Bob use the three-bit repetition code to transmit messages to each other using a shared secret key. All the noise in the channel that Eve perceives is because of these deliberate errors that Alice applies. In his model he assumes that Alice and Bob share a binary-symmetric channel. This work does not address the issue of whether the errors would resemble a plausible channel, nor does it consider the case where the channel contains intrinsic noise. Natori gives a simple treatment of quantum steganography which is a modification of super-dense coding [29]. Martin introduced a notion of quantum steganographic communication based on a variation of Bennett and Brassard's quantum-key distribution (QKD), hiding a steganographic channel in the QKD protocol [30]. Curty e.al. proposed three different quantum steganographic protocols [31].

• B.    Quantum gate [19]

Quantum circuit model of computation in quantum computing [10] [1] [9], a quantum gate or quantum logic gate is a basic quantum circuit which operates on a small number of qubits. They are the building blocks of quantum circuits, like classical logic gates are basically for conventional digital circuits. Quantum logic gates are reversible like other classical logic gates. However, classical computing can be performed by the help of only reversible gates. Quantum gates are represented as matrices. A gate which acts on k qubits is represented by a 2^kx2^k unitary matrix. The number of qubits in the input and output of the gate is equal. There are various types of quantum gates are represent the qubits. They are Hadamard gate, Pauli-X gate, Pauli-Y gate, Pauli-Z gate, Phase shift gates, Swap gate, Controlled gates, Toffoli gate, Fredkin gate, etc. Here we use Controlled gates to represent the qubits and control the operations.

C. Reversible Classical Logic [19]

The first concepts of the reversibility of computation were raised in the 1970s. There were two issues which are logical reversibility and physical reversibility, both were intimately connected. Logical reversibility reconstructs the input from the output of a computation or gate function. The NAND gate is explicitly irreversible, it has two inputs and one output, while the NOT gate is reversible (it’s own inverse). In case of Physical reversibility the NAND gate has only one output, one of it’s inputs has effectively been erased in the process, whose information has been irretrievably lost. The change in entropy that we would associate with the lost of one bit of information is ln 2, which, thermodynamically, corresponds to an energy increase of kT ln 2, where k is Boltzman’s constant and T is the temperature. The heat dissipated during a process is usually taken to be a sign of physical irreversibility, that the microscopic physical state of the system cannot be restored exactly as it was before the process took place. Reversible logic gates are symmetric with respect to the number of inputs and outputs. The reversible NOT gate, whose truth table is given in Fig. 2. It can also write this in the form of a matrix, or as a graphic. The matrix form lists the lines in the truth table in the form 0 , 1 . t he matrix field with 1's and 0's such that each horizontal or vertical line has exactly one 1, which is to be interpreted as a one-to-one mapping of the input to the output. A two-bit gate closely related to the NOT gate is the two-bit Controlled-NOT (or C-NOT) gate. Controlled-NOT gate shows in Fig. 3, performs a not on the second bit if the first bit is 1 , but otherwise has no effect. The C-NOT is sometimes also called XOR, since it performs an exclusive OR operation on the two inputs bits and writes the output to the second bit.

C-NOT	<00>	<01>	<10>	<11>
|<00>				LI
<01>
<10>
<11>

Figure 3: NOT Gate Truth Table

The Controlled NOT gate (also C-NOT or CNOT) is a quantum gate that is an essential component in the construction of a quantum computer. The proof of operation [32] is given below:

Let ^ 1 0) =

basis.

11 =

be the orthonormal

Figure 2: NOT Gate Truth Table

Let | ^ ) = _x |0) + y\ 1)

and ф = y | 0) + x | 1)

qubit of | ^ ^ .

x

y

y

x

| ф be the flip

Recall that ф ) ® | в = Ф )1 в = ^’ в ■••(!)

i. When control qubit is 0

First,

we

shall

prove

that: CNOT | 0, ф = 10, ф

Before we compute, however, note that our specific definition of CNOT assumes an eigen basis of

100) =

1		0		0		0
0	,|01 =	1		0	.KO=	0
0		0	,|10> =	1		0
_ 0 _		_ 0 _		_ 0 _		_ 1 _

Then, it's not difficult to verify

x

that 1 0, ф = x | 0)| 0) + y 0)| 1) =

Then

CNOT\ 0, ф =

x

y

x

y

y 0

..(2)

=|0ф) .(3)

Therefore CNOT doesn't change the ф ) qubit if the first qubit is 0.

ii. When control qubit is 1

Now, we shall prove that CNOT 1, фф = 1 1, ф , which means that the CNOT gate flips the ф ) qubit.

Similarly

to the

first demonstration,

we

have

1 ¹^Ф =

0 0 x ^y

.

As we can see that

Then CNOT| 1, ф =

,

using these on the equation above gives

CNOT 1, ф = x | 1,1 + y | 1,0) = |1) ( x | 1 + y | 0) ) = 11, ф . • ⁽⁵⁾ Therefore the CNOT gate flips the ф ) qubit into | ф if the control qubit is set to 1. A simple way to observe this is to multiply the CNOT matrix by a column vector, noticing that the operation on the first bit is identity, and a NOT gate on the second bit.

D. Text Steganography

The affluence of electronic documented information available in the world as well as the exertion of serious linguistic analysis makes this an interesting medium for steganographic information hiding. Moreover the Text is one of the ancient media used in steganography. Letters, books and telegrams hide secret messages within their texts in earlier time i.e. before the electronic age comes. Text steganography refers to the hiding of information within text i.e. character-based messages. There are three basic categories of text steganography (Fig. 1) maintained here: format-based methods, random and statistical generation and linguistic methods. [8]

i. Format-based methods [8]: This methods use the physical formatting of text as a space in which to hide information. Format-based methods usually modify existing text for hiding the steganographic text. Insertion of spaces or non-displayed

Figure 4: Mechanism of Text Steganography characters, careful errors tinny throughout the text and resizing of fonts are some of the many format-based methods used in text steganography.

• ii.    Random and statistical generation method [8]: This avoid comparison with a known plaintext, steganographers often resort to generating their own cover texts. Character sequences method hide the information within character sequences.

• iii.    Linguistic methods [8]: The affluence of electronic documented information available in the world as well as the exertion of serious linguistic analysis makes this an interesting medium for steganographic information hiding.

Fig. 4 shows the mechanism of text steganography. Firstly, a secret message will be covered up in a covertext by applying an embedding algorithm to produce a stego-text. The stego-text will then be transmitted by a communication channel to a receiver.

In this paper, an approach of quantum text

steganography using Oriya character mapping method has been proposed based on the use of some special character. Here the quantum truth table also mapped to increase the security level and complexity. In this method the length of the stego and the cover are same so prediction of existence of message is difficult in view of that characteristics, so this one is the unique method which has been developed in this steganography approach.

Figure 5: Mapping Technique

The proposed scheme has been enthused by the author’s previous work [11], [12], [13], [14], [15], [16], [17], [18] on various approaches of steganography methods. The quantum truth table approach has been incorporated from previous work [18].

This paper is organized into the following sections. Section II describes the proposed model. Algorithms of various processes like embedding, extracting and GUI are discussed in Section III. Analyses of the results are in section VI. The last section descries the concluded part of the work.

• II.    The Proposed Model

Text steganography, whatever this paper exactly deals with, uses regional language, specifically the Oriya text as the medium where to hide information. Here the explanation of text steganography remains wide in order to differentiate it from the more specific “linguistic steganography”. Text steganography can involve anything from changing the ASCII character from the specific position of an existing text.

The input messages can be in any digital form and are often treated as a bit stream. Pick two pair of this bit stream of message one by one. Then select the proper Oriya Text as cover and change it to it’s equivalent binary code. Then a matrix formed with the help of message length and map the C-NOT truth table (how to map it, shown in Fig. 6) from left most corner in a sequence (vertically or horizontally), after that start embedding one by one if the mapped value showing not ‘1’ value. After that secret message has been embed to the cover text by replacing the next ASCII of “ ^ ” and “9 ” by ASCII 34 i.e. [ ‘ ] and 39 i.e. [ ’ ] in Oriya language based on the mapping information given in Fig. 5 to generate the stego text. By the help of replacing technique the stego length are being same as cover. At the receiver side with the help of same mapping algorithm and other different reverse operation has been carried out to get back the original information.

Figure 6: Quantum Truth Table Mapping

1	32	in	59	117	69	120		105	115	55	32	55
154	1/	’ 69	107	32	32	107	112	126	65	39	110	113
35	Х15	107	1	77		69	35	32	93	97	155	32
118	110	1	118	4й"	34	106	115	93	32	83	69	100
107	35	107	32	105	110	32	32	113	55	162	35	125
125	126	115	105	71	93	96	77	32	79	32	125	89
35	107	106	32	93	113	112	172	71	126	49	32	115
44	118	59	118	125	32	119	100	125	39	43	96	100
52	55	52	97	52	35	52	125	101	121	43	121	52
50	115	96	32	112	118	117	32	32	32	32	125	112
53	65	161	95	153	103	93	105	71	110	82	97	154
32	118	32	121	65	32	100	164	39	118	162	121	42
?g	69	52	125	126	110	115	77	103	32	32	106	32
96	97	45	32	32	117	66	32	32	89	69	115	65
97	101	43	117	117	77	126	93	112	126	172	32	115
32	115	32	97	106	96	73	125	153	32	103	106	110
112	96	96	55	112	32	118	32	35	67	115	59	35
118	103	115	115	115	39	44	93	32	120	89	95	125
32	115	107	96	32	126	32	121	32	112	106	32	107
45	66	32	32	117	93	117	110	100	32	113	107	113

A. Solution Methodology

The proposed system involves two windows i.e. SENDER SIDE and RECEIVER SIDE. The user will be someone who is aware with the process of information hiding and will have adequate knowledge of steganography systems. The user have a duty to select a plain text message from a file, another Oriya text to be used as the carrier (cover text) and then use the proposed embedding method which will hide the message in the selected cover text and will procedure the stego text. The user at the receiver side should be able to extract the message from the stego text with the help of different reverse process in chronological manner to un-hide the message from the stego text. The GUI of the proposed solution has been shown in Fig. 7.

Figure 7: GUI Representation

• III.    Algorithms

In this section, algorithms for different processes used both in the sender side and receiver sides are described which are furnished below:

A. Algorithm for Message Embedding

• Select the message & Oriya cover. Check whether the selected text is capable of embedding. If not possible repeat this step otherwise continue.

• Map the quantum C-NOT gate to the matrix MATMSG (N x N) vertically or horizontally & Put the message value by replacing ’0’ in the matrix MATMSG.

• •    Check the message sequence and pick first two bit sequence (MSG). Start from the first character of the cover text (TX).

• •    Start checking & embedding.

o If MSG=‘11’ & character “fit ” from the TX then change from ASCII (109,254) to (109,39)

o Else If MSG=‘10’ & character “^ ” from the TX then change from ASCII (109,254) to (109,34)

o Else If MSG=‘01’ & character “Q ” from the TX then change from ASCII (107,254) to (107,39)

o Else If MSG=‘00’ & character “ Q ” from the TX then change from ASCII (107,254) to (107,34)

• •    Repeat the above step for the remaining bit sequence of the message and prepare the stego text.

• B.    Algorithm for Message Extracting

• •    Select the stego text put in MATMSG (N x N) matrix and map the quantum C-NOT gate to the matrix vertically or horizontally.

• •    Extract the message value from ‘0’Th position of C-NOT. Pick values one by one from MATMSG and create MSG.

• •    Select the stego text TX.

o If “ ®” and next ASCII is 39, MSG=’11’.

o Else If “ ® ” and next ASCII is 34, MSG=’10’. o   Else If “⁹” and next ASCII is 39, MSG=’01’.

o   Else If “ 9” and next ASCII is 34, MSG=’00’.

• C.    Algorithm of GUI in Sender side

• •    Select the Cover Text from the set of Text files.

• •    Check whether the selected text is capable to do the embedding or not. If not possible then error.

• •    Select the message in text form.

• •    Embed the encrypted message in the cover text to form the stego text.

• D.    Algorithm of GUI in Receiver side

• • Receive the text with embedded message along with positions.

Extract the encrypt form of message from the Stego Text.

• IV.    Results Analysis

500

0.98884

1000

0.97772

2000

0.95561

Figure 8: Cover Text

Figure 9: Secret Message eisiQ geea eeabiaiaa gg«e s°6aa 6 eee aeeia 1 as aaieea аееёее дйот де eea es№i eeqiea sae oiee oasaa aiaase- w Seamoia eiee 1 <9 eascira asm aeai esa eqia easee, eea e 9°6®ae eaa eaiOT See® eea tigei eiaoi 1 eieiy geea eeaitiiaiaa a^eiia етад <з see gaeia 1 as aaissa aieeses ^siae as sea awi seaiea eqs ее оаеаа ai^ase- w esaitiioiei eige 1 69 saeqiai asm aeai eea Sara eaeaa eeg 6 o=eaae eaa eai54 esea eea beei eiaai 1 sisiq aaea eeeitiiaiaa ayie e°6®a 6 бее aeeia 1 as aaieea aieeooa eyaoa go sea aeaai esaiaa ego oiea oaeaa aieaee- 49 Seanniaiai eiSe 1 69 eaeqiai aeaie aeai eea @qia eases, бед 6 e^eaae eaa 6ai@4 eeea eee beei eiaai 1 aioiq sees Seaiuaiqa q^eiie e°6gq 6 esc qeeia 1 as qaiqea aieeose qeiae gs eea asa@i eeqiea age aiea oaeaa oiaaes- 45 eeatiieiai eige 1 <9 eaeaiai aeaie aeai eea §qiq easye. seg ti s^aqs east eai@4 ossa see ueei eiaai 1 eieiq gesa Ssaiiiiaqa q^ie 6”6®q 6 eee qeeia 1 as qaieea aieeeee ^eiaa ge eea aeaai eeqiea Sqe eiee oaeaa 0№se- v5 esaiiiiaiai ei5e 1 69 sasgiaj asm aeai eea Sara eaese. eeg <3 e=egqe Sae eai64 ossa eee tieei eiaai 1 eieiQ qeea eeaitiiaiqa qqaia e°6®q <з бее qaeia 1 09 qeieea aiesess %aa® go eea asa® eeqiea eqs aee oasaa aigaee- 49 esaiuaiai ei5e 1 ее eaeqieg aeaie aeai eea Sara eases, eeg 6 e'eaqe eaa 6@i§4 oeea oee №ei eiaa 1 eioiQ qesa «аивиа qqw e^gq <3 see qeeia 1 ae qaiesa assess ^«100 go eea aeaei eeqiea ego aiee oaeaa aieaea- 49 eeai^iarai eige 1 ее eaeqieg aoaie aeai esa 5qiq еоб99, бед 6 e°6aqe eaa 6aie4 ossa eee tigei eiaai 1 eiety geea eeaitiiaaa agaie

Figure 10: Stego Text

MESSAGE LENGTH (In Character)	CORRELATION VALUE
10	0.99978
50	0.99888
100	0.99776
150	0.99665
200	0.99553
250	0.99442
300	0.99330
350	0.99218
400	0.99107

Figure 11: Correlation values of cover and stego text in different length of message

r between X and Y. The sample correlation coefficient is written r xy

n

^ ( x_t - x)(yi - y )

i =1 _________________________________

(n - 1) X Л before embedding operation. This method hides two bit per word in the cover text which reflects the high embedding capacity of the system. Although the embedding capacity of the proposed method depends upon the embedding sequence, pattern of the cover text. This method hides two bit per word in the cover text which reflects the high embedding capacity of the system. Although the embedding capacity of the proposed method is depends upon the Oriya characters. In this method the length of the stego and cover are same and unchanged. So due to the said reason the steganalysis part can also handle.

• A. Similarity Measure

For comparing the similarity between cover text and the stego text, the Correlation method for measuring similarity between two strings has been computed. The Correlation [27] is a measure of similarity between two strings.

Correlation : The most familiar measure of dependence between two quantities is the Pearson product-moment correlation coefficient [27], or “Pearson’s correlation”. It is obtained by dividing the covariance of the two variables by the product of their standard deviations. Karl Pearson developed the coefficient from a similar but slightly different idea by Francis Galton. The Pearson correlation is +1 in the case of a perfect positive (increasing) linear relationship (correlation), -1 in the case of a perfect decreasing (negative) linear relationship (anticorrelation)[27] and some value between -1 and 1 in all other cases, indicating the degree of linear dependence between the variables. As it approaches zero there is less of a relationship (closer to uncorrelated). The closer the coefficient is to either -1 or 1, the stronger the correlation between the variables. If the variables are independent, Pearson’s correlation coefficient is 0, but the converse is not true because the correlation coefficient detects only linear dependencies between two variables.

Figure 12: Representation Correlation values of Cover and Stego where x and y are the sample means of x and y, sx and Sy are the sample standard deviations of X and Y. The number of matching (but different sequence order) characters divided by two defines the number of transpositions. The Correlation score of comparing cover text and stego text is shown in Fig. 11 and the chart in Fig. 12 represent the Correlation of different stego of various length of message.

If we have a series of n measurements of X and Y written as x i and y i where i = 1, 2, . . . n then the sample correlation coefficient can be used in Pearson correlation

Figure 13: Graph of Cover Text

Figure 14: Graph of Stego Text

Figure 15: Graph of Input Message

Figure 16: Graph of Output Message

In Fig. 13 and Fig. 14 it has been observed that the cover and stego graphs both are remain same. The Fig. 15 and Fig. 16 shows that the Input and Output messages graphs are same; from this view also we can proof our technique. In Fig. 17 shows the computation time of Stego and Message generation, the graph shows in Fig. 18. To observe the computation time it is proved that the receiver side message generation system is faster than sender side stego generation system.

Message Length (In Character)	Computation Time (in Second)
	Stego Generate (A)	Message Generate (B)	Difference (B-A)
10	0.2184	0.1872	-0.0312
50	0.858	0.7956	-0.0624
100	1.8876	1.5132	-0.3744
150	2.3556	1.7472	-0.6084
200	3.2136	2.4024	-0.8112
250	3.8688	2.6364	-1.2324
300	4.2276	3.8376	-0.39
350	5.2572	3.7128	-1.5444
400	6.162	4.4304	-1.7316
500	7.2852	5.2572	-2.028
1000	13.6501	9.8437	-3.8064
2000	26.6294	19.7809	-6.8485

• V.    Conclusions

In this paper the authors presented an approach of text steganography method using Oriya language by the help of quantum truth table mapping technique. This property generates the stego text with minimum degradation. In this method the length of stego and cover remain same and this property enables the method to avoid the steganalysis also. The result shows that the performance of the technique is satisfactorily. This work can be extended by using other Indian regional language.