Computational science guided soft

International Journal on Computational Sciences & Applications (IJCSA) Vol.4, No.5, October 2014
COMPUTATIONAL SCIENCE GUIDED SOFT
COMPUTING BASED CRYPTOGRAPHIC TECHNIQUE
USING ANT COLONY INTELLIGENCE FOR WIRELESS
COMMUNICATION (ACICT)
Arindam Sarkar and J. K. Mandal
Department of Computer Science & Engineering, University of Kalyani, W.B, India
ABSTRACT
In this paper, a computational science guided soft computing based cryptographic technique using Ant
Colony Intelligence (ACICT) has been proposed. In this proposed approach at first a metamorphosed
based strategy is used to produce intermediate cipher text. Finally, ACI generated keystream is used to
further encrypt the intermediate cipher text to produce the final cipher text. In this approach an ant agent
having a pheromone deposition consisting of a group of alphanumeric characters is called a key stream
and each character in the key stream is known as key. The key stream length always be less than or equal
to the plaintext to be encrypt. The keystream generation is based on distribution of characters in the
plaintext. Instead of transmitting the plain keystream to the receiver, further encryption is done on
keystream and encrypted keystream get transmitted to the receiver. Parametric tests are done and results
are compared with some existing classical techniques, which show comparable results for the proposed
system.
KEYWORDS
Ant Colony Intelligence (ACI), Key Generation, cryptography, soft computing
1.INTRODUCTION
These days a variety of techniques are available to secure data and information from
eavesdroppers [1, 6, 7, 8, 10] with some merits and demerits. Most of the cryptographic
algorithm needs large memory and energy [2, 3, 4, 5, 9]. There are few applications
where soft computing is used in encryption/decryption purpose. In recent days
cryptographic protocols are also get deployed in wireless communication. Wireless
devices have the problem of memory and energy constraints. In this paper, a novel soft
computing based technique has been proposed for encryption/decryption in wireless
communication to address this problem.
The organization of this paper is as follows. Section 2 of the paper deals with the proposed ACI
based key generation technique. Encryption and Decryption Process has been discussed in section
3 and 4 respectively. Example of ACI based key stream generation is discussed in section 5.
Results are described in section 6. Conclusions are drawn in section 7 and that of references at
end.
DOI:10.5121/ijcsa.2014.4505 61

62
2. THE ACI BASED SESSION KEY GENERATION TECHNIQUE
The key stream generation process using ACI technique is illustrated using the following steps:
Step 1. In ACI based approach an ant agent is used to denote a key stream (collection of
alphanumeric characters). Each Ant can have multiple dimensions. Each dimension denotes an
individual key within that key stream. The dimensions in the key stream can be filled or unfilled.
For example if the ceiling of dimension of each Ant is equal to 192 then it is represented by
Anti or Key streami = (Key1, Key2, … , Key192).
Which actually signify a key stream comprises of 192 keys i.e. 192 alphanumeric characters. Key
stream length can be obtained by counting number of dimensions are filled in the key stream.
Generally key stream length is less than or equal to the plain text. With 192 alphanumeric
characters multiple key stream can be generated of predetermined fixed length by permutation of
these predetermined fixed length characters ordering all feasible ways without any reappearance.
So, for example if total number of alphanumeric characters = 192.
If key stream length = 128 then among 192 alphanumeric characters 128 alphanumeric characters
are elected such a way so that by ordering all possible ways without any duplication these 128
characters forms multiple key stream having fixed length i.e. 128. For an example if 5 characters
A, C, M, H, R are taken to form key stream of length 4 among 192 alphanumeric characters. Then
there are 120 possible ways of obtaining key stream.
ACMHR, CAMHR, MACHR, HACMR, RACMH
ACMRH, CAMRH, MACRH, HACRM, RACHM
ACHMR, CAHMR, MAHCR, HAMCR, RAMCH
ACHRM, CAHRM, MAHRC, HAMRC, RAMHC
ACRMH, CARMH, MARCH, HARCM, RAHCM
ACRHM, CARHM, MARHC, HARMC, RAHMC
AMCHR, CMAHR, MCAHR, HCAMR, RCAMH
AMCRH, CMARH, MCARH, HCARM, RCAHM
AMHCR, CMHAR, MCHAR, HCMAR, RCMAH
AMHRC, CMHRA, MCHRA, HCMRA, RCMHA
AMRCH, CMRAH, MCRAH, HCRAM, RCHAM
AMRHC, CMRHA, MCRHA, HCRMA, RCHMA
AHCMR, CHMRA, MHACR, HMACR, RMACH
AHCRM, CHMAR, MHARC, HMARC, RMAHC
AHMCR, CHRMA, MHCAR, HMCAR, RMCAH
AHMRC, CHRAM, MHCRA, HMCRA, RMCHA
AHRCM, CHAMR, MHRAC, HMRAC, RMHAC
AHRMC, CHARM, MHRCA, HMRCA, RMHCA

63
ARCMH, CRAMH, MRACH, HRACM, RHACM
ARCHM, CRAHM, MRAHC, HRAMC, RHAMC
ARMCH, CRMAH, MRCAH, HRCAM, RHCAM
ARMHC, CRMHA, MRCHA, HRCMA, RHCMA
ARHCM, CRHAM, MRHAC, HRMAC, RHMAC
ARHMC, CRHMA, MRHCA, HRMCA, RHMCA
Using 192 characters total number of generated possible key stream is given in following
equation.
− » »
192 ! /(192 )! 192 ! ( ) 192 !* 2 .718
192
1
=
c e
c
Step 2. According to Ant Colony Intelligence technique each ant should have an allied energy.
The ACI technique also offers energy for each and every ant or key stream. The energy value of
the ant agent is computed by taking the number of characters in the key stream occurring in the
plain text divided by the key stream length. The pheromone deposition of the ant agent with a
maximum energy value greater than a specified threshold value is the solution and the key stream
is chosen for encryption. Energy value for each ant agent is computed by the following equation.
Energy (Anti) = count (keyj Î plain text)/ lengthof(Anti). ,
where j=1, 2, …, lengthof(Anti).
Step 3. If (Energy(Anti) threshold value) then
return (Anti with Energy(Anti) = max energy value)
Step 4.If (Energy(Anti) threshold value) then
repeat change the key stream i.e. update the pheromone deposition of ant agent and
perform the following steps
Step4.1.Evaluate Energy value for each ant agent in the current trial.
Step4.1.Select the ant agent where,
Energy(Anti in current trial) threshold value) then
return (Anti in current trial with Energy(Anti) = max energy value in current trial )
until Energy(Anti in a trial) threshold value)
3. ENCRYPTION TECHNIQUE
Step 1. If the length of the plain text is more than length of the key stream then the values of the
key stream are added to a preset value to produce the keys for the characters in the level 2
encoded plain text which is at a position grater than the length of the key stream.
Predetermined value is calculated using the following equation

International Journal on Computational Sciences Applications (IJCSA) Vol.4, No.5, October 2014
64
Predetermined_value= lengthof(plaintext)/2
Step 2. For the very first plain text block keys are form by the values of the characters in the key
stream.
Step 3. For the consecutive plain text blocks keys are generated by adding preset value with the
keys of the previous block for sinking the key storage load that in turn reduces the
space complexity.
Keyforblock(i)= Keyforblock(i-1) + Predetermined_value, where i=2
Step 5. Now, those characters in the key stream are appearing in the plain text represent them
using their corresponding ASCII code value and those characters are not appearing in
the plain text represent them using their corresponding ASCII value. Then perform
XOR operation between plain text and key stream to generate the encrypted cipher text.
4. DECRYPTION TECHNIQUE
Step 1. Receive the cipher text from the sender.
Step 2. Compute the predetermined value based on cipher text length.
Step 3. Using predetermined value and keys in the key stream receiver generates the keys for the
portion of the text exceeding the length of the key stream.
Step 4. Generate plain text by performing XOR operation between encoded cipher text and key
stream.
5. EXAMPLE OF ACI BASED KEY STREAM GENERATION
Consider the text to be encrypted is “antcolonyintelligence” threshold value is assumed to be
0.65. Each ant agent has a pheromone deposition comprising of characters representing the
keystream. The energy level of the ant agent is a count of the characters in the key stream
occurring in the plain text divided by the length of the keystream. The ant agent with a maximum
energy level greater than the specified threshold value is chosen as the keystream for text
encryption. Table. 1 shows the pheromone deposition of ant agents denoting the keystream and
their corresponding energy value. Since the second ant agent in Trial II has the maximum energy
value 0.66 which is greater than the threshold value. The keystream ueigunscaoblyt
corresponding to that ant agent is chosen for encryption. Each character in the keystream is
chosen as the key for encryption
Table 1.Pheromone deposition table
Iteration I Energy Iteration II Energy
ckyaptseifdorgq 0.46 cyusadkleownjgm 0.53
anwghqbcletzduo 0.53 ueigunscaoblyt 0.66
yurtdfbnczfsvam 0.33 tedcbkhouesxvaq 0.40
rqewcalkygtxifo 0.60 ivbjtwaxrdgnzpu 0.33
Highest energy 0.60 Highest energy 0.66

65
5.1 EXAMPLE OF CHARACTER TREE GENERATION
A character tree is generated for the plain text “antcolonyintelligence”. For plain text
“antcolonyintelligence” figure 5 shows corresponding tree. Characters ‘a’, ‘y’ and ‘g’ occur once
and character‘t’, ‘c’, ‘o’ and ‘i’ occurs twice. Characters ‘l’ and ‘e’ occurs thrice. Character ‘n’
occurs four times. Each character code can be generated by preorder traversal. Left and right
branches of the tree are labeled as ‘0’ and ‘1’ respectively. Table 2 depicts the code value of a
particular character.
1
21
1
20
1
0
3 17
e
1
0
2
15
1
0
1 14
1
0
3
11
i
y
l
1
0
2 9
o
Figure 1: Character code tree
0
g
1
0
2
7
1
0
2 5
1
0
4
1
c
t
n a
If (Code_Tree_value (Char) ==0) then Value(Char)= ASCII(Char) + ASCII (Code_Tree_value
(Char));
If Code_Tree_value (Char) 32 then Value(Char)= ASCII(Char) + Code_Tree_value (Char);
If Code_Tree_value (Char) 255 then Value(Char)= 255-
Position_of_even_or_odd_number_series(sum_of_digit(Code_Tree_value (Char))) –
number_of_1’s_in (Code_Tree_value (Char));

66
Table 2.Code Tree table
Plain text
Code
Value
Modified value
g 0 0 103 + 48 = 151
e 10 2 110 + 2 = 112
i 110 6 105 + 6 = 111
y 1110 14 117 + 14 = 128
l 11110 30 112 + 30 = 142
o 111110 62 62
c 1111110 126 126
t 11111110 254 254
n 111111110 510 255 - 3 - 8 = 244
a 1111111110 1022 255 - 3 - 9 = 243
5.2EXAMPLE OF METAMORPHOSED TREE BASED ENCRYPTION
TECHNIQUE
From the original tree metamorphosed tree is derived using mutation. Table 3 depicts the
metamorphosed code value. For identical entry in a table, length of the corresponding code is
summed up to the character.
1
0
1
20
g
21
3 17
e
2
15
Metamorphosed at level 2
1 14
3
11
i
y
l
2 9
o
2
7
2 5
4
1
c
t
n a
0
0
1
1
0
1
0
1
1
1
0
0
1
0
1
0

67
Figure 2: Metamorphosed Code Tree
Table 3. Metamorphosed Code Table
Plain text
Code
Value
Modified Value
g 0 0 103 + 48 = 151
e 11 3 101 + 3 = 104
i 101 5 105 + 5 = 110
y 1000 8 121 + 8 = 129
l 10010 18 108 + 18 = 126
o 100111 39 39
c 1001100 76 76
t 10011011 155 155
n 100110100 308 255 - 6 - 4 = 244
a 100110101 309 255 - 6 - 5 = 243
Now, the plaintext “antcolonyintelligence” becomes
10010111/01101000/01101110/10000001/01111110/00100111/01001100/10011011/11110100/1
1110011
5.3 EXAMPLE OF ANT COLONY INTELLIGENCE (ACI) KEY STREAM BASED ENCRYPTION
Now, those characters in the key stream are appearing in the pain text represent them using their
corresponding character code value and those characters are not appearing in the plain text
represent them using their corresponding ASCII value.
Table 4.ACC based key stream guided Encrypted text Generation Table
ACI
based
key
stream
Blocks
based on
ACI based
key
stream
Length
Value
of the ACI based
keys
Metamorphosed
Encoded
Cipher text Value
ACI
based
key
stream
u
Block1
117 (ASCII code) 243 134
e 112 (Character code) 244 132
i 111 (Character code) 155 244
g 151 (Character code) 76 219
u 117 (ASCII code) 39 82
n 244 (Character code) 126 138
s 115 (ASCII code) 39 84
c 126 (Character code) 244 138

68
a 243 (Character code) 129 114
o 62 (Character code) 110 80
b 98 (ASCII code) 244 150
l 142 (Character code) 155 21
y 128 (Character code) 104 232
t 254 (Character code) 126 128
u
Block2
117 (ASCII code) +
2 (Predefine value) =119
126 9
e
112 (Character code) ) +
2 (Predefine value) = 114
110 28
i
111 (Character code) +
151 230
g
104 241
u
117 (ASCII code) +
244 131
n
76 186
s
115 (ASCII code) +
104 29
So, ACI encoded cipher text is
10000110/10000100/11110100/11011011/01010010/10001010/01010100/10001010
01110010/01010000/10010110/00010101/11101000/10000000/00001001/0001110011100110/11
110001/10000011/10111010/00011101
6. RESULTS
For the purpose of the practical implementation, ACICT technique has also been implemented on
SYS, COM, EXE, DLL, and CPP files.
6.1 RESULT OF ENCRYPTION, DECRYPTION, CHI-SQUARE, DEGREE OF FREEDOM
Table 5.Result for Encryption, Decryption, Chi-Square, Degree of Freedom
Source
File
Encrypted
File
Source
Size
(In
Bytes)
Encryption
Time
(In
Seconds)
Decryption
Time
(In
Seconds)
Chi-
Square
Value
Degree
of
Freedom
ASPI2HLP.SYS A1.SYS 1105 0.0087 0.0083 178 171

69
DBLBUFF.SYS A2.SYS 2614 0.0213 0.0206 536 234
IFSHLP.SYS A3.SYS 3708 0.0306 0.0302 1057 239
REDBOOK.SYS A4.SYS 5664 0.0603 0.0594 1927 246
RAMDRIVE.SYS A5.SYS 12663 0.1362 0.1358 1429 253
CHOICE.COM A1.COM 5239 0.0436 0.0432 1232 237
MORE.COM A2.COM 10471 0.0873 0.0865 753 236
THELP.COM A3.COM 11072 0.0930 0.0927 2572 252
DOSKEY.COM A4.COM 15495 0.1671 0.163 4128 254
SYS.COM A5.COM 18967 0.2047 0.2044 3156 255
TCDEF.EXE A6.EXE 11611 0.0975 0.0972 3798 255
CLIPBRD.EXE A10.EXE 18432 0.1542 0.1537 4973 255
UNZIP.EXE A3.EXE 23044 0.1930 0.1926 2247 255
PING.EXE A8.EXE 24576 0.2651 0.2647 4338 251
NETSTAT.EXE A9.EXE 32768 0.3535 0.3532 8739 255
HIDCI.DLL A1.DLL 3216 0.0262 0.0254 877 225
NDDENB.DLL A2.DLL 10976 0.0917 0.0911 8273 253
NDDEAPI.DLL A3.DLL 14032 0.1178 0.1173 4705 251
ICCCODES.DLL A4.DLL 20992 0.2262 0.2253 9712 254
WINSOCK.DLL A5.DLL 21504 0.2320 0.2314 9653 254
GRADE.CPP A1.CPP 1257 0.0102 0.0096 442 71
MAX.CPP A2.CPP 4071 0.0340 0.0336 2389 79
LCM.CPP A3.CPP 4663 0.0386 0.0381 2637 88
LINEAR.CPP A4.CPP 9540 0.1030 0.1025 13517 83
BUBBLE.CPP A5.CPP 9558 0.1031 0.1027 14169
83
6.2RESULT FOR FREQUENCY DISTRIBUTION TESTS
The frequency of each of the 255 characters in the source file and the same in the encrypted file
were calculated and compared to assess the efficiency of the proposed technique.
Figure 3.Segment of Frequency Distribution Chart for .SYS File

70
6.3 KEY STORAGE COMPARISON AND ANALYSIS WITH EXISTING METHODS
Table 7 shows the comparison results among ACICT, AES, RC4 and Vernam Cipher.
Table 6
Comparison of key storage in Proposed ACICT, AES, RC4 and Vernam Cipher
Length
of
Plain
text
Key Storage
( ACICT )
Key
Storage
(AES)
Key
Storage
(RC4)
In ACICT only 15 bits key stream need to be store for plain text size 64, 120 500 and for the plain
text of length 1000 only 20 bits key stream need to be store. In ACICT If the key stream size is
less than the plain text to be encrypt then the key stream can be expand by summing up a fixed
value with the existing key stream for the plain text grater than the key stream in terms of length.
Figure 4: Memory heap
The figure 4 shows the memory heap usage for at the run time.
Key
Storage
(Vernam
Cipher)
64 15 128 52 60
120 15 128 106 120
500 15 128 437 500
1000 20 128 913 1000

71
Figure 5: Memory Gantt Chart
Figure 5 shows Memory Gantt Chart during execution.
Figure 6: Number of Threads at runtime
Figure 6 shows the various numbers of threads active during runtime.
7. CONCLUSION
In this paper an Ant Colony Intelligence algorithm to generate the key stream for encryption
based on the distribution of characters in the plain text has been proposed. In this method the
pheromone deposition of the ant agent evaporates when it moves to the next trail and therefore
the ant agent needs to update the pheromone deposition representing the key stream. The energy
value denoting its attractiveness towards the solution is found by counting the number of
characters in the key stream occurring in the plain text. In future other soft computing tools can
be applied in cryptographic domain.

72
ACKNOWLEDGEMENTS
The author expresses deep sense of gratitude to the DST, Govt. of India, for financial assistance
through INSPIRE Fellowship leading for a PhD work under which this work has been carried out.
References
[1] Atul Kahate, Cryptography and Network Security, 2003, Tata McGraw-Hill publishing Company Limited, Eighth
reprint 2006.
[2] R. Mislovaty, Y. Perchenok, I. Kanter, and W. Kinzel. Secure key-exchange protocol with an absence of injective
functions. Phys. Rev. E, 66:066102, 2002.
[3] A. Ruttor, W. Kinzel, R. Naeh, and I. Kanter. Genetic attack on neural cryptography. Phys. Rev. E, 73(3):036121,
2006.
[4] Wolfgang Kinzel and ldo Kanter, Neural cryptography proceedings of the 9th international conference on Neural
Information processing(ICONIP 03).
[5] Charles Pfleeger, Shari Lawrence Pfleeger, Security in computing, Third Edition 2003, pp 48, Prentice Hall of India
Pvt Ltd, New Delhi.
[6] Biham, E. and Seberry, J.”Py (Roo): A Fast and Secure Stream Cipher”. EUROCRYPT'05 Rump Session, at the
Symmetric Key Encryption Workshop (SKEW 2005), 26-27 May 2005.
[7] Chung-Ping Wu, C.C. Jay Kuo, “Design of Integrated Multimedia Compression and Encryption Systems”, IEEE
Transactions on Multimedia, Volume 7, Issue 5, Oct. 2005 Page(s): 828 – 839.
[8] HongGeun Kim, JungKyu Han and Seongje Cho.”An efficient implementation of RC4 cipher for encrypting
multimedia files on mobile devices”. SAC '07 Proceedings of the ACM symposium on Applied computing, 2007,
pp 1171--1175, NewYork, USA.
[9] Mantin and A. Shamir, “Weaknesses in the key scheduling algorithm of RC4”, Lecture Notes in Computer Science,
Vol. 2259, Revised Papers from the 8th Annual International Workshop on Selected Areas in Cryptography, pp: 1 -
24, 2007.
[10]Sarkar Arindam, Mandal J. K., “Artificial Neural Network Guided Secured Communication Techniques: A
Practical Approach”, Paperback: 128 pages, Publisher: LAP LAMBERT Academic Publishing (June 4, 2012),
Language: English, ISBN-10: 3659119911, ISBN-13: 978-3659119910.
Authors
Arindam Sarkar is INSPIRE FELLOW(DST, Govt. of India), MCA (VISVA BHARATI,
Santiniketan, University First Class First Rank Holder), M.Tech (CSE, K.U, University
First Class First Rank Holder). Total number of publications 25.
Jyotsna Kumar Mandal is M. Tech.(Computer Science, University of Calcutta),
Ph.D.(Engg., Jadavpur University) in the field of Data Compression and Error Correction
Techniques, Professor in Computer Science and Engineering, University of Kalyani, India.
Life Member of Computer Society of India since 1992 and life member of cryptology
Research Society of India. Dean Faculty of Engineering, Technology Management,
working in the field of Network Security, Steganography, Remote Sensing GIS Application, Image
Processing. 25 years of teaching and research experiences. Eight Scholars awarded Ph.D. and 8 are
pursuing. Total number of publications 267.

Computational science guided soft

Recommended

More Related Content

What's hot (19)

Viewers also liked (13)

Similar to Computational science guided soft (20)

Recently uploaded (20)

Computational science guided soft