Cryptography on Indian Languages

Presented by

Dindayal Mahto MT/IS/1009/10

(Internal Guide) Prof. Sandip Dutta (Assco. Prof.) Dept of IT BIT Mesra, Ranchi

(External Guide) Prof.(Dr.) D.A. Khan (Assco. Prof.) Dept of CA NIT Jamshedpur

Objectives
1. Introduction 2. Proposed Methodology 3. Results and Discussions 4. Conclusions 5. Future Scope of Work 6. References
2

1.1. Why Cryptography on Indian Languages

Hindi is the third most spoken language in the world.-HLI, New Delhi It is spoken by more than 800 million people all over the world & It is the official language of India too.-HLI, New Delhi It is the non-English speaking world that has come to dominate Internet use with over 50% of the 147 million Internet users, being nonEnglish speakers.-Broin 2000
3

1.2. Problem Definition

The use of a multilingual approach in cryptography is not prevalent. A noble Algorithm of doing Cryptography on Hindi Text.

Literature Reviews
1. A Multilanguage Encryption Technique by G.P. Kr et al, 2010 Seventh International Conference on Information Technology, 978-0-7695-3984-3/10 2010 IEEE This algorithm that focuses on encryption of plain text over a range of languages supported by Unicode.

Literature Reviews (Contd.)

Uses a direct mapping technique: I. Unicode of each character is obtained II. This value is divided by mapping Constant M III. Remainder for Encryption IV. Quotients for Decryption But, Problems are like: I. Mapping of characters II. Enhancing the replacement strategy
6

Literature Reviews (Contd.)

2. A Method for the Construction of MinimumRedundancy Codes by David A Huffman, Associate, IRE, Massachusetts Institute of Technology, Cambridge, Mass. A minimum-redundancy code is constructed in such a way that the average number of coding digits per message is minimized. It is used for compressing the text data.
Dindayal Mahto, MT/IS/1009/10 7

Literature Reviews (Contd.)

3. Lossless Text Compression using Dictionaries, Umesh S. Bhadade, G.H. Raisoni Institute of Engineering & Management, Jalgaon (MS) India 425001, International Journal of Computer Applications (0975 8887), Vol 13 No.8, Jan.2011 It proposed a pre-compression technique that can be applied to text files. The output of it can be further applied to standard compression techniques.
Dindayal Mahto, MT/IS/1009/10 8

Literature Reviews (Contd.)

4. A New Approach for Text Steganography using Hindi Numerical Code, Megha Pathak, International Journal of Computer Applications (0975 8887), Volume 1 No. 8 , 2010 Uses letters and its numerical code. A numerical code to Hindi letters which is built on the basis of 4-bit binary value. Later it replaced each 4-bit code with a different word which starts with the respective letters that are assigned in the scheme.
9

Literature Reviews (Contd.)

5. Efficient Mapping Methods for ECC, O. Srinivasa Rao et. al., International Journal of Engineering Science and Technology, Vol. 2(8), 2010, 3651-3656 Mapping methods of the alphanumeric characters on to the x-y co-ordinate of the Elliptic curve defined over a finite field Zp. I. Static (One-to-One) Mapping Method II. Dynamic (One-to-N) Mapping Method
10

1.3. Aims of the Thesis

Proposal of new Algorithm for doing Cryptography on Hindi text. We have implemented the work using two algorithms. 1.3.1. 1.3.2. Use of Elliptic Curve Cryptosystem Use of Huffman Coding Algorithm

11

1.4. Why ECC?

The popular public-key cryptography for encryption and digital signatures is RSA. But the Key Size for secure RSA, has increased over recent years. This has put a heavier processing load on applications using RSA. For solving the burden of RSA, an alternative system emerged that is ECC.
12

1.4. Why ECC? (Contd)

ECC Key Sizes RSA Key Sizes Key Size Ratio (Bits) 163 256 384 512 (Bits) 1024 3072 3072 15360 (Bits) 1:6 1:12 1:20 1:30
13

Table1: Key Comparison of RSA and ECC

1.4.1. What is Elliptic Curve?

An elliptic curve can be defined by an equation in two variables with coefficients. The general form of an elliptic curve is: y2 + b1xy + b2y = x3 + a1x2 + a2x + a3 Where x and y are the variables, while a1, a2, a3, b1 and b2 are the coefficients.
14

1.4.1. What is Elliptic Curve? (Contd.)

There are three kinds of Elliptic Curve, which are as follows: 1. Elliptic Curve over Real Number y2 = x3 + ax + b
Here, x, y, a & b take real values. If 4a3 + 27b2 0, the equa on represents a nonsingular curve (3 roots), otherwise, singular EC (not 3 roots).

15

1.4.1. What is Elliptic Curve? (Contd.)

2. Elliptic Curve over Finite Field GF(p) y2 mod p = (x3 + ax + b) mod p Here, the EC is denoted as Ep(a, b).
Where x, y, a, & b are bound to be elements of finite field GF(p) and 0 <= x <= p.

16

1.4.1. What is Elliptic Curve? (Contd.)

3. Elliptic Curve over finite field GF(2n) y2 +xy= x3 + ax2 + b
Here, the EC is denoted as E2n(a, b). Where x, y, a, & b are bound to be elements of finite field GF(2n).

17

1.4.2. What is ECC?

Elliptic curve cryptography (ECC) is an approach to public-key cryptography based on the algebraic structure of elliptic curves over finite fields. The use of elliptic curves in cryptography was suggested independently by Neal Koblitz and Victor S. Miller in 1985.
18

1.4.3. ECC with Diffie-Hellman key exchange

Consider that Alice and Bob are two users who wish to communicate and, exchange the secret key using ECC. The exchange of key between Alice and Bob proceeds as follows: 1.Choose a large prime p, such that p is either a prime or in the form 2n.
19

1.4.3. ECC with Diffie-Hellman key exchange (Contd.)

2.Choose the elliptic curve coefficients a and b for the cubic equations of the form y2 mod p = (x3 + ax + b) mod p or y2 +xy= x3 + ax2 + b. This defines Ep(a, b), the elliptic group of points. 3.Choose a base point G=(x1, y1) in Ep(a, b), whose order is very large value, m. 4.Alice chooses an integer XA < m, which becomes his or her private key.
20

1.4.3. ECC with Diffie-Hellman key exchange (Contd.)

Then, Alice calculates his or her public key YA, as shown here: YA = XA * G The public key YA is a point in Ep(a, b). 5. Bob chooses an integer XB < m, which becomes his or her private key. Then, Bob calculates his or her public key YB, as shown here: YB = XB * G The public key YB is a point in Ep(a, b).
21

1.4.3. ECC with Diffie-Hellman key exchange (Contd.)

6. Alice calculates the secret key K using his or her private key XA and the public key of Bob (that is, YB), as: K = XA * YB 7. Similarly, Bob calculates the secret key K using his or her private key XB and public key of Alice (that is, YA), as shown here: K = XB * YA
22

1.4.3. ECC with Diffie-Hellman key exchange (Contd.)

Figure1: Asymmetric Key Exchange Method

23

1.5. ECC Encryption

When Alice has to send a message (say, Pm) to Bob, Alice first chooses a random integer r. Then, Alice encrypts the message using Bobs public key YB and the base point G to produce the cipher-text Cm, containing the pair of points as shown here: Cm = {r * G, Pm + r * YB}

24

1.6. ECC Decryption

On receiving the cipher-text Cm, Bob decrypts the cipher-text to obtain the original plain-text Pm. For this, it multiplies the first point in Cm (that is, r * G) with its private key XB, and then subtracts it from the second point (that is, Pm + r * YB), as shown here: (Pm + r * YB ) (XB * (r * G)) => Pm + (r * (XB * G)) (XB * (r * G)) => Pm
25

1.7. ECC Mapping Methodology

Elliptic Curves over finite field GF(p): For elliptic curves over finite field GF(p), we use a cubic equation in which the variables and coefficients all take on values in the set of integers from 0 through p-1, for some prime number p, and in which calculations are performed modulo p. y2 mod p = (x3 + ax + b)mod p
26

1.7. ECC Mapping Methodology (Contd.)

For the given a=1,b=1,and p=79,the points of the elliptic curve are
(0,1),(0,78),(2,13)(2,66)(3,30)(3,49)(5,17)(5,62)(6,12)(6,67)(11,0)(14,28) (14,51)(15,28)(15,51)(16,20)(16,59)(18,20)(18,59)(20,11)(20,68)(21,35) (21,44)(23,5)(23,74)(25,3)(25,76)(26,12)(26,67)(27,35)(27,44)(28,16) (28,63)(29,18)(29,61)(30,31)(30,48)(31,35)(31,44)(43,36)(43,43)(44,38) (44,31)(46,22)(46,57)(51,29)(51,50)(53,20)(53,59)(56,0)(59,33)(59,46) (60,16)(60,63)(61,39)(61,40)(64,12)(64,67)(65,12)(65,67)(70,12)(70,67) (71,21)(71,58)(78,0)

The above points can be mapped to the alphanumerical characters in two ways
27

1.7. ECC Mapping Methodology (Contd.)

The above points can be mapped to the alphanumerical characters in two ways 1. Static (one-to-one) Mapping Method 2. Dynamic (one-to-n) Mapping Method

28

1.7. ECC Mapping Methodology (Contd.)

1. Static (one-to-one) Mapping Method: From the cubic equation of Elliptic curve, for each given value of x, there are two values for y. One of these values of y and corresponding x will be used to map the any alphanumeric character. Like this, all numeric characters are mapped on to the different the x coordinates and their corresponding y coordinates of the given curve.
29

1.7. ECC Mapping Methodology (Contd.)

2. Dynamic (one-to-n) Mapping Method: In this method, alphanumerical characters are mapped on to the points of the Elliptic curve dynamically. For every transmission of the message from source to destination, the alpha-numeric characters mapping mechanism changed dynamically.

30

1.7. ECC Mapping Methodology (Contd.)

It is thus concluded that dynamic mapping method can strengthens the elliptic curve cryptosystem.

31

1.8. Security of ECC?

Alice encrypts the message Pm with r * YB (r is only known to Alice) and r * G; therefore, the attacker needs the value of r, G and r * G to decrypt the message, which is not so easy.

32

1.7. Why Compression?

It is required because our Algorithm uses Hindi text, which is represented by Devanagari script, and that takes 16 bits to represent each Hindi alphabet in Unicode format. So, in order to reduce network size and time, we have applied the concept of Data Compression technique.
33

1.7. What is Compression?

Compression is the art of representing the information in a compact form rather than its original or uncompressed form.

34

1.8. Huffman Coding

Huffman Encoding Algorithms use the probability distribution of the alphabet of the source to develop the code-words for symbols. Shorter code-words for higher probabilities and longer code-words for smaller probabilities are assigned.

35

1.8. Huffman Coding (Contd.)

Suppose that we have a file of 100 characters. Each character is one of the alphabets from a to h.

Figure 2.: Huffman Tree

36

1.8. Huffman Coding (Contd.)

a Frequency 45 Variable Length 0 Code b 13 101 c 12 100 d 16 111 e 9 1101 f 5 1100 g 0 h 0 -

Table 2: Huffman Probability Distribution Average Length:

Space required to store the original message = 100*8=800 bits Space required to store decoded messages =(45*1+13*3+12*3+16*3+9*4+5*4)=224 Compression Ratio=28%

37

2. Proposed Methodology
Working behaviors of new Algorithm have been discussed in following way:
For Sender
Step 1: Use of Huffman Encoding, Step 2: ECC Encryption on the compressed data,

For Receiver:
Step 3: ECC Decryption Step 4: Huffman Decoding

38

2.1. Encryption
Steps for Encryption 1. This new system will take input through text file, which contains plain Hindi message. 2. Input messages are compressed through compression algorithm. 3. Use of ECC to encrypt the messages 4. Output is generated

39

2.1. Flowchart: Encryption (Contd.)

Start

Take Input from .txt file as Plain Text Processes based on Huffman compression technique Processes based on ECC technique Cipher Text is generated Stop
40

2.2. Decryption
Steps for Decryption 1. In this step, it will take input through text file, which contains cipher messages. 2. Use of ECC to decrypt the messages. 3. then messages are uncompressed through compression algorithm. 4. Output is generated.

41

2.2. Flowchart: Decryption

Start

Take Input from .txt file as Cipher Text Processes based on ECC Decryption technique Processes based on Huffman decompression technique Plain Text is generated Stop
42

3. Results & Discussions

Our new Algorithm has been implemented in MATLAB 2008.

Dindayal Mahto, MT/IS/1009/10

43

3. Results & Discussions (contd.)

Figure 3: Main Screen of our Algorithm

44

3. Results & Discussions (contd.)

Figure 4: Choosing Hindi text file

45

3. Results & Discussions (contd.)

Figure 6: After Encryption

46

3. Results & Discussions (contd.)

Figure 4: After Decryption

47

4. Conclusion
This work uses the compression algorithm and asymmetric key for achieving: 1. Space problem 2. Speed issue 3. Secrecy problem

Dindayal Mahto, MT/IS/1009/10

48

5. Future Scope of Work

It is working only on stand-alone PC. It to be developed for Multiuser Systems. With more efficient Compression Algorithm.

Dindayal Mahto, MT/IS/1009/10

49

References
[1] A Multilanguage Encryption Technique, NIT, Durgapur, P. Choudhury, G.P. Kumar, A.K. Murmu, B. Parajuli, 2010 Seventh International Conference on Information Technology, 978-07695-3984-3/10 2010 IEEE DOI 10.1109/ITNG.2010.105 [2] A Method for the Construction of Minimum-Redundancy Codes by David A Huffman, Associate, IRE, Massachusetts Institute of Technology, Cambridge, Mass. [3] Lossless Text Compression using Dictionaries, Umesh S. Bhadade, G.H. Raisoni Institute of Engineering & Management, Jalgaon (MS) India 425001, International Journal of Computer Applications (0975 8887), Volume 13 No.8, January 2011
Dindayal Mahto, MT/IS/1009/10 50

References
[4] Pu, I.M., 2006, Fundamental Data Compression, Elsevier, Britain [5] Efficient Mapping Methods for Elliptic Curve Cryptosystems, O. Srinivasa Rao et. al. / International Journal of Engineering Science and Technology Vol. 2(8), 2010, 3651-3656 [6] Securing Messages in Wireless Channel by using New Method of Compression, A.C. Shakir, J. Min, R.J. of Applied Sc., Eng & Tech, 3(7), 625-632, 2011, ISSN: 2040-7467

Dindayal Mahto, MT/IS/1009/10

51

Thank You

Dindayal Mahto, MT/IS/1009/10

52