A Hybrid Approach for Ensuring Security in Data Communication

Jan Zizka (Eds) : CCSIT, SIPP, AISC, PDCTA - 2013
pp. 133–144, 2013. © CS & IT-CSCP 2013 DOI : 10.5121/csit.2013.3614
A HYBRID APPROACH FOR ENSURING
SECURITY IN DATA COMMUNICATION
Shailender Gupta1
, Bharat Bhushan2
, Surbhi Singhania3
and
Jeetesh Gulani4
1,2,3
Department of Electronics Engineering, YMCA University of Science and
Technology, Faridabad, India
1
shailender81@gmail.com, 2
bhrts@gmail.com, 3
surbhidec24@gmail.com
4
Indian Institute of Technology Kharagpur,India
4
jeetesh1993@iitkgp.ac.in
ABSTRACT
For a very long time, various forms of steganographic and cryptographic techniques have been
used to ensure security in data communication. Whereas steganography is the art of hiding the
fact that any communication is taking place, cryptography on the other hand ensures data
security by changing the very form of the data being communicated by using a symmetric or an
asymmetric key. But, both the methods are susceptible to being weakened by a challenger. In
steganography, there is always a possibility of detection of the presence of a message by the
opponent and most of the cryptographic techniques are vulnerable to disclosure of the key. This
paper proposes a hybrid approach where in image steganography and cryptography are
combined to protect the sensitive data thereby ensuring improved security in data
communication. To find the impact of the same, a simulator was designed in MATLAB and
corresponding time complexities were recorded. The simulation results depict that this hybrid
technique although increases the time complexity but ensures an enhanced security in data
communication.
KEYWORDS
Cryptography, Image Steganography, information hiding, time complexity
1. INTRODUCTION
With an increase in data communication in the recent past, the security of data over a
communication channel is of paramount importance. Various Steganographic [1-2] and
Cryptographic [3] algorithms have been proposed to ensure data security in communication.
Steganography is a smart strategy which calls for hiding the sensitive information in other data
which is generally considered to be ordinary by an unaware viewer. This cover data may be in the
form of text, image, audio, video, etc. The security of the data communication depends on the
ingenuity in the alteration of the cover such that the adversary is not able to detect the presence of
a secret message. The receiver uses the reverse approach to extract the data from the modified
cover.

134 Computer Science & Information Technology (CS & IT)
Cryptography refers to a science of transforming the messages into an unreadable form by means
of a set procedure hence making the core data inaccessible to an interfering individual. In this
approach, the original message called the plain text is converted into an encoded message called
the cipher text with the help of an encryption algorithm. This cipher text is then sent over a
communication channel. On the receiver side a decryption algorithm is used to recover the plain
text.
Despite the above mentioned techniques being advanced and sophisticated, none of them provide
a fool proof safety in communication. Steganographic techniques used deteriorate the image
quality [4-5] hence presence of data in images can be easily detected by malicious nodes. On the
other hand cryptographic techniques [3] has the disadvantage that if an intruder is able to figure
out the key then he will be able to decrypt the data. Hence, none of the methods are completely
guaranteed in a stand-alone fashion.
This paper proposes a novel scheme to provide a comparatively high security to the data. In this
scheme, a combination of the above mentioned techniques offers advantages of both
steganography and cryptography. The encryption methods used are: Data encryption standard
(DES) algorithm [3] and Diffie-Hellman algorithm [3], whereas the size of the key implemented
is 32 bit and 64 bit. The encrypted data is embedded in the images of different sizes using three
different algorithms namely: Pseudorandom Substitution, Transform Domain (DCT) and
Distortion. To check the efficacy of the proposal the time complexity of the process at the sender
and receiver side was recorded.
The rest of the paper has been organized as follows: Section 2 provides the current
steganographic techniques used in this paper, Section 3 provides the problem identification,
Section 4 provides the proposed work for problem identified, Section 5 gives the simulation and
results of the proposed model, Section 6 gives the conclusion followed by references.
2. STEGANOGRAPHIC TECHNIQUES USED
2.1. Pseudo Random LSB- Substitution Method
The least significant bit substitution [6-11] method is the oldest method used for steganography.
A simple substitution method involves replacing the least significant bit of a first few pixels of
the image with that of the bits of the data. The change incurred in the pixel value is that of only
+/- 1. Therefore, the chances of these changes being noticed in the embedded image are very
minor.
This method being fairly banal and is susceptible to attacks. Therefore, we have introduced a
modification of the same by introducing pseudorandom substitutions. Here, instead of replacing
the bits of the ordered pixels the sender chooses a series of random pixels and replaces the LSB of
these bits with the data bits. The receiver must also choose the same random pixels to extract the
data from the LSB of these pixels. For this purpose the sender and receiver share a key. This key
is used as a seed to generate the same random numbers for embedding and extracting purpose.
This method can be further made robust by sharing a second seed which will help the sender and
receiver to choose the data-bits randomly from the complete set of data bits to be embedded. Thus
embedding random data bits in random pixel and as a result, improving security. Even if a hacker
is able to get hold of one key the absence of the second will make his effort go vain.
This method is good for transmitting large data as the image can hold the binary data of length up
till the total number of pixels of the image. Hence an RGB image can hold data as much as three

Computer Science & Information Technology (CS & IT) 135
times a gray scale image without any significant distortion. One major assumption of the method
is that the length of the embedded data must be known to both the receiver and the sender for
successful transmission.
2.2. Distortion Technique
This technique [12-15] is similar to the substitution technique except for the fact that it is
important, for the receiver and sender to have the original cover image. A pixel is chosen using
the pseudorandom method explained before. Next, the data bit to be embedded is evaluated. If the
binary bit is ‘0’ then the pixel is left as it is. On the other hand when it is ‘1’ a small value ∆x is
added to the pixel. This stego-image is then sent to the receiver.
The receiver subtracts the pixels of the received image from the original cover image he already
has. This results in an array of ∆x and 0 values. These elements are then accessed in the same
pseudorandom manner. If the value is zero the bit is read as ‘0’ and if it is ∆x it is read as 1. This
is altogether compiled to generate the data.
The distortion technique is faster compared to the substitution technique. It can also hold large
volumes of data in accordance to the image size. But it is also more vulnerable because if a
hacker has the cover image he may be able to access the data provided he knows the other
parameters. And the distortion of the image is more prominent compared to the substitution
methods.
2.3. Transform Domain Technique (DCT)
The distortion and substitution methods are vulnerable to any modifications in the image. The
processes like cropping, compressing, image processing, etc can largely destroy the information
embedded in the image. The transform domain methods [16] hide the data in the significant part
of the images hence; make it more robust against these disturbances. DCT is one of the transform
domain methods. This method is specifically designed to resist any changes in the secret data
caused by jpg image compression and still allow successful compression. Jpeg compression can
be explained with the help of the following block diagram (see Fig 1).
Figure 1. Jpeg Compression
2.3.1. JPEG Compression
The Quantization matrix used at the step of quantization is the most important element of the
process. It is a standard matrix given by:

Referring to this matrix, the sender and receiver agree on the two indices (j1, k1) and (j2, k2)
(say). These are chosen such that the coefficients at these indices correspond to the cosine
function with middle frequencies. This ensures that the information is stored in the significant
parts of the image. According to the above shown matrix the positions (5, 2) and (4, 3) or (2, 3)
and (4, 1) are favorable for this operation.
The image is divided into an array of 2 dimensional matrices of the magnitude 8x8. The DCT
operation is performed on these matrices to obtain another set of 8x8 matrices (say D).
Starting with the first block and the first data bit, we check if the data bit is ‘1’ or ‘0’. Next,
compare the magnitude of coefficients at the chosen indices. If D (j1, k1)>D (j2, k2) and the bit to
be embedded is ‘1’ then the indices are left as it is. Else to embed ‘0’ the coefficient values are
swapped. Alternately if D (j1, k1) <D (j2, k2) and the bit to be embedded is ‘1’ then the
coefficients are swapped whereas for ‘0’ they are left as it is. Also this algorithm ensures that |D
(j1, k1)- D (j2, k2)|>x, where x>0.
This is done by increasing the value of the larger coefficient and by further decreasing that of the
smaller to maintain a constant difference between the two.
The factor of distortion depends upon the constant x and the location of the coefficient. This
method causes visibly more distortion that the previous two methods and also consumes more
time. Moreover the capacity of a particular sized image is fixed. If we try to embed the data more
than its capacity the secret data is lost due to superimposition. Yet, it is the most robust method
for obtaining compressed images and to perform other manipulations without losing the
embedded secret data. The security can be improved by choosing the blocks and data in the
pseudorandom manner.
3. PROBLEM IDENTIFICATION
Neither steganography nor cryptography is standalone solution to provide security in
communication networks. In steganography the presence of data can be easily detected by
observing distortion in the cover image. On the other hand cryptographic techniques can fail in
scenario of key disclosure by nodes having rogue intentions. In our opinion if the data embedded
in the image is encrypted, the confidentiality of the data would be maintained even if it is
extracted by an intruder node. Thus it would be beneficial if the secret data is encrypted before
applying steganography. Some researchers who have previously worked in this direction as
follows:

Sujay Narayana et. al. [17] proposed a scheme using S-DES algorithm combined with LSB
substitution technique. The image to be hidden was first encrypted using a key and the encrypted
image was embedded in the cover image.
Shailender Gupta et. al. [18] proposed a hybrid model that combines steganography and
cryptography. The Cryptographic technique used was RSA and Diffie Hellman and the
steganographic technique used was LSB substitution.
This paper has been inspired from the above mentioned literature. This paper focuses on
embedding a text file in the image unlike Sujay narayanas et. al. attempt which was centrally
based on the idea of hiding an image in a cover image. Moreover Shailender guptas et. al. used
only 32 bit key for the purpose of encryption and decryption and uses only LSB substitution
steganographic technique. The idea proposed in the current paper is an improvement over this
work.
4. THE PROPOSED SCHEME
4.1. Sender side
The sender side process is depicted in figure 2. The secret message called the plain text is
converted into cipher test using one of the two mentioned algorithms i.e. DES or Diffie-Helman
using a 32 bit or 64 bit. This cipher text is then converted to bits. These bits are embedded into a
cover image using one of the three steganographic techniques namely pseudorandom substitution,
Transform Domain (DCT) or Distortion. The resultant encrypted stego-image is sent over a
channel.
Figure 2. Block diagram of proposed scheme at sender side
4.2. Receiver side
The complete process at the receiver side is depicted in figure 3. The received encrypted stego-
image is subjected to steganographic extraction process and decryption technique corresponding
to the techniques used at transmitter side to recover the plain text.
Image file
Plain text
(secret
message)
Convert cipher
text to bits
Apply
Steganography
Encrypted
Stego
Image
Encryption Key
32 / 64 bit
1. Diffie
Hellman
2. DES

Figure 3. Block diagram of proposed scheme at receiver side
5. SIMULATION AND RESULTS
5.1. Performance metrics
The following parameters were used to record the results:
• Time complexity- Defined as the total time taken by the sender side to encrypt and
embed the secret message in the cover image, combined with the time taken by the
receiver to extract and decrypt the secret message from the transmitted image.
• Qualitative comparison of images- The original cover image and the modified images
(after cryptography and steganography) were placed side by side and subjected to visual
inspection.
5.2. Simulation setup
A simulator was designed in MATLAB 7.12.0 to simulate various combinations of the
cryptographic and steganographic techniques at different key and image sizes. The simulation
setup parameters used are as shown below:
Table 1. Setup parameters.
Component Parameter Value of parameter
Steganography Techniques Substitution, Distortion, DCT
Image Size • 1024x1024
• 1536x1536
• 2048x2048
• 3072x3072
• 4096x4096
• 5120x5120
Plain text size 1 K Byte
Cryptography Technique Diffie Hellman
Key = Gxy mod (P)
Where for
32 Bit
p = 2863311530, g=112557
x=637, y=597
64 Bit
p= 768614336404564650, g=112557
x=637, y=597
Image file
Plain text
(secret
message)
Convert
encrypted data
to bits
Apply
steganography
extraction
1. Substitution
2. DCT
3. Distortion
Encrypted
Stego
Image
Decryption Key
32 / 64 bit
Decryption
1. Diffie
Hellman 2.
RSA

DES
Key(k)=4294967295 (32 bit)
Key(k)=18446744073709551615(64
bit)
Processor Type i5-64 bit
RAM: 2 GB
speed: 2370 MHz
5.3. Simulation results
This section is dedicated to depicting the impact of the combination of the steganographic and
cryptographic techniques on images of various sizes. Before discussing time complexities two
important things should be kept in mind:
• The time shown in all cases is the total time elapsed on the sender and receiver side. It
does not include time elapsed in transfer of information from source to destination.
• The time complexities in case of Diffie- Helman (combined with stego techniques)
include the time of key calculation on the receiver as well as the sender side.
5.3.1. Results of Steganography and 32 bit cryptography
5.3.1.1. Qualitative results
Figure 4 depicts the effect of combination of Steganography and Cryptography (32 bit) on the
quality of the cover image (1024x1024). It can be observed that:
• Pseudorandom Substitution combined with Diffie-Helman produces best qualitative
result followed by pseudorandom substitution with DES.
• DCT combined with DES produces the worst qualitative result.
In the similar manner, the impact of this scheme on images of different size i.e. 1536x1536,
2048x2048, 3072x3072, 4096x4096 pixels was recorded.
Figure 4(a). Original Image Figure 4 (b). Result of applying DCT

Figure 4(c). Result of applying
Distortion
Figure 4 (d). Result of applying
Substitution
Figure 4(e). Result of DES plus DCT Figure 4(f). Result of Diffie-Helman
plus DCT
Figure 4(g). Result of DES plus
Substitution
Figure 4(h). Result of Diffie-Helman
plus Substitution
Figure 4(i). Result of DES plus
Distortion
Figure 4(j). Result of Diffie-Helman
plus Distortion
Figure 4. Impact of proposed technique on images (Image size = 1024X1024 pixels)

5.3.1.2. Qualitative results
Figure 5 depicts the total time taken by a complete process of the combination including the time
on sender and receiver side. It can be observed that:
• Distortion combined with DES has minimum time complexity
• DCT combined with Diffie-Helman has maximum time complexity
• Time complexity for all the combinations increases with an increase in the image size.
Time complexity for image size 1024 x 1024
0
20
40
60
80
100
120
140
160
180
Technique
Timecompexity
Substitutiononly
Substitution+Delfi-Helman
Substitution+DES
Distortiononly
Distortion+Delfi-Helman
Distortion+DES
DCTonly
DCT+Delfi-Helman
DCT+DES
Figure 5(a). Time Complexity (1024X1024
pixel)
0
50
100
150
200
250
300
350
Technique
Timecompexity
Substitutiononly
Substitution+DES
Distortiononly
Distortion+DES
DCTonly
DCT+Delfi-Helman
DCT+DES
Figure 5(b). Time Complexity (1536X1536
pixel)
0
100
200
300
400
500
600
Technique
Timecompexity
Substitutiononly
Substitution+DES
Distortiononly
Distortion+DES
DCTonly
DCT+Delfi-Helman
DCT+DES
Figure 5(c). Time Complexity (2048X2048
pixel)
0
200
400
600
800
1000
1200
1400
Technique
Timecompexity
Substitutiononly
Substitution+DES
Distortiononly
Distortion+DES
DCTonly
DCT+Delfi-Helman
DCT+DES
Figure 5(d). Time Complexity (3072X3072
pixel)
Time complexity for image size 4096x x4096
0
500
1000
1500
2000
2500
Technique
Timecompexity
Substitutiononly
Substitution+DES
Distortiononly
Distortion+DES
DCTonly
DCT+Delfi-Helman
DCT+DES
Figure 5(f). Time Complexity (4096X4096 pixel)
Figure 5. Impact on Time complexity of Proposed Technique
5.3.2. Results of Steganography and 64 bit cryptography
5.3.2.1. Qualitative result
Figure 6 depicts the effect of combination of Steganography and Cryptography (64 bit) on the
quality of the cover image (2048x2048). It can be observed that:

• Pseudorandom Substitution combined with Diffie-Helman produces best qualitative
result followed by pseudorandom substitution with DES.
• DCT combined with DES produces the worst qualitative result.
In the similar manner, the impact of this scheme on images of different size i.e. 3072x3072,
4096x4096, 5120x5120 pixels was recorded.
Figure 6(a). Original image Figure 6(b). Result of applying DCT
only
Figure 6(c). Result of applying
Distortion
Figure 6(d). Result of applying
Substitution
Figure 6(e). Result of a DES plus
Substitution
Figure 6(f). Result of Diffie-Helman
plus Substitution
Figure 6. Impact of proposed technique on images (Image size = 3072X3072 pixels)
5.3.2.2. Results of time complexity
Figure 7 depicts the total time taken by a complete process of the combination including the time
on sender and receiver side. It can be observed that:
• Distortion combined with DES has minimum time complexity
• DCT combined with Diffie-Helman has maximum time complexity
• Time complexity for all the combinations increases with an increase in the image size

0
100
200
300
400
500
600
700
800
Technique
Timecomplexity
Subtitution
Substitution + DES
Substitution + Deffie Hellman
Distortion only
Distortion + DES
Distortion + Deffie Hellman
DCT
DCT + DES
DCT + Deffie Hellman
Figure 7(a). Time Complexity (2048X2048
pixel)
0
200
400
600
800
1000
1200
1400
1600
1800
Technique
Timecomplexity
Subtitution
Substitution + DES
Distortion only
Distortion + DES
DCT
DCT + DES
Figure 7(b). Time Complexity (3072x3072
pixel)
0
500
1000
1500
2000
2500
3000
3500
Technique
Timecomplexity
Subtitution
Substitution + DES
Distortion only
Distortion + DES
DCT
DCT + DES
Figure 7(c). Time Complexity (4096x4096
pixel)
0
500
1000
1500
2000
2500
3000
3500
4000
Technique
Timecomplexity
Subtitution
Substitution + DES
Distortion only
Distortion + DES
DCT
DCT + DES
Figure 7(d). Time Complexity (5120X5120
pixel)
Figure 7. Impact on Time complexity of Proposed Technique
6. CONCLUSION
In this paper a new scheme of combining steganography and cryptography has been proposed.
Where steganography eliminates any chances of detection of hidden data being present in the
cover image, text, etc., cryptography ensures data confidentiality by changing the very form of
data. Thus the hybrid approach ensures a higher degree of data security compared to either of the
techniques applied alone. After analysing the results of the application of the proposal it can be
concluded:
• The time complexity of this hybrid process is much more than that of pure
steganography.
• As the size of the key used for encryption and decryption increases time complexity
further increases.
• There is no major effect on the process of steganography on adding it to cryptography
because the data being embedded is in the form of bits in either case.
• On increasing the size of the image used as a cover, the time complexity increases even
more.
• DCT Combined with Diffie-Helman has the highest time complexity in comparison to
any other proposed combination.
• The pseudorandom substitution combined with any cryptographic technique provides
least distortion in image quality.
These results can be very useful to researchers who wish to use both steganography and
cryptography to secure data in communication networks.

ACKNOWLEDGEMENTS
The authors would like to thank everyone, just everyone!
REFERENCES
[1] Neil F. Johnson and Sushil Jajodia, "Steganalysis of Image Created Using Current Steganography
Software", Workshop of Information Hiding Proceedings, Portland Oregon, USA,15-17 April,1998.
Lecture Notes in Computer Science, Vol.1525, Springer-Verlag (1998).
[2] Clair, Bryan, “Steganography: How to Send a Secret Message”, 8 Nov. 2001
www.strangehorizons.com/2001/20011008/steganography.shtml.
[3] “Cryptography and Network Security: principles and practices”, William Stallings, pearsons
education, first Indian reprint 2003.
[4] Rhodas, G. B., “Method and Apparatus Responsive to a Code Signal Conveyed Through a Graphic
Image”, U.S. Patent 5,710,834, 1998.
[5] Swanson, M. D., B. Zhu, and A. H. Tewk, “Transparent Robust Image Watermarking", in
Proceedings of the IEEE International Conference on Image Processing, vol. 3, 1996, pp. 211-214.
[6] Bender, W., D. Gruhl, and N. Morimoto, “Techniques for data hiding”, IBM Systems Journal , vol.
35, no. 3/4, 1996, pp. 131-336.
[7] Moller. S.A., Pitzmann, and I. Stirand, “Computer Based Steganography: How It Works and Why
Therefore Any Restrictions on Cryptography Are Nonsense, At Best”, in Information Hiding: First
International Workshop, Proceedings, vol. 1174 of Lecture Notes in Computer Science, Springer,
1996, pp. 7-21.
[8] Gruhl, D., A. Lu, and W. Bender, “Echo Hiding in Information Hiding”, First International
Workshop, Proceedings, vol. 1174 of Lecture Notes in Computer Science, Springer,1996, pp. 295-
316.
[9] Kurak, C., and J. McHughes, “A Cautionary Note On Image Downgrading”, in IEEE Computer
Security Applications Conference 1992, Proceedings, IEEE Press, 1992, pp. 153-159.
[10] Van Schyndel, R. G., A. Tirkel, and C. F. Osborne, “A Digital Watermark”, in Proceedings of the
IEEE International Conference on Image Processing, vol. 2, 1994, pp. 86-90.
[11] Johnson, N. F., and S. Jajodia, “Exploring Steganography: Seeing the Unseen", IEEE Computer, vol.
31, no. 2, 1998, pp. 26-34.
[12] Maxemchuk, N. F., “Electronic Document Distribution”, AT&T Technical Journal,
September/October 1994, pp. 73-80.
[13] Low, S. H., et al., Document Marking and Identifications Using Both Line and Word Shifting," in
Proceedings of Infocom'95, 1995, pp. 853-860.
[14] Low, S. H., N. F. Maxemchuk, and A. M. Lapone, “Document Identification for Copyright Protection
Using Centroid Detection”, IEEE Transactions on Communications, vol. 46, no. 3, 1998, pp. 372-383.
[15] Low, S. H., and N. F. Maxemchuk, “Performance Comparison of Two Text Marking Methods”, IEEE
Journal on Selected Areas in Communications, vol. 16, no. 4, 1998, pp. 561-572.
[16] Rhodas, G. B., “Method and Apparatus Responsive to a Code Signal Conveyed Through a Graphic
Image”, U.S. Patent 5,710,834, 1998.
[17] Sujay Nararayana and Gaurav Prasad, “ Two New Approaches For Secured Image Steganography
Using Cryptographic Technique and Type Conversions” , Signal and Image Processing: An
International Journal (SIPIJ) Vol. 1, No. 2, December 2010.
[18] Shailender Gupta, Ankur Goyal And Bharat Bhushan, “Information Hiding Using Least Significant
Steganography and Cryptography”, I. J Modern Education and Computer Science, vol. 6, pp. no. 27-
34, IJMECS-2012.

A Hybrid Approach for Ensuring Security in Data Communication

More Related Content

What's hot (20)

Similar to A Hybrid Approach for Ensuring Security in Data Communication (20)

More from cscpconf (20)

Recently uploaded (20)

A Hybrid Approach for Ensuring Security in Data Communication