A Novel DNA Encryption System Using Cellular Automata

International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 DOI : 10.5121/ijsptm.2015.4404 39 A N OVEL D NA E NCRYPTION S YSTEM U SING C ELLULAR A UTOMATA G.Shanmugasundaram 1 , P.Thiyagarajan 2 , S.Pavithra 1 1 Department of Information Technology, Sri Manakula Vinayagar Engineering College, Madagadipet, Puducherry. 2 Department of Computer Science and Engineering, Mahindra Ecole Centrale, College of Engineering, Hyderabad - 43. A BSTRACT DNA Cryptography is a new born cryptographic field emerged with the research of DNA Computing in which DNA is used as an Information carrier. Cellular automata is dynamic in nature so it provide dynamic behavior in the system which may increase the security in the system. DNA cryptography is provide a secure way to encrypt the text and automata changes the state of the system based on the present state, it will occur in discrete time. These qualities are most impressive in these techonology which help us to provide a highly secured security system for the users. Most of the encryption techniques based on the cellular automata have limitations. To overcome this lacuna, we propose a novel DNA cryptography algorithm with cellular automata to achieve randomness, parallelism, uniformity, reversibility and stable. An algorithm implemented and its results obtained are depicts here, and a result analysis is done with other algorithms. K EYWORDS Cellular automata, DNA, Thymine, Uracil. 1. INTRODUCTION Nowadays, Internet has become parts and parts of human life and it has emerged closely with human such that it is elevated as one of the basic needs of human life. The facility of using the Internet in mobile phones has increased its use among common man. Many sectors of business such as bank, travel, shopping etc., make use of Internet to reach many people and flourish their business by introducing their application (apps). Though there are many advantages of Internet, it has adverse effect such as security. Sensitive information such as financial transactions, confidential government information, and medical records is transmitted through Internet. The protection of this sensitive information in the Internet is done by information security systems such as cryptography [1], steganography [2] [3] and combination of cryptography and steganography [4]. Security in Internet is provided by the information security systems. With the advancement in computational infrastructure and computer processing capacity, standard information security systems are no longer secure 1 . As traditional cryptographic methods are based on mathematical and theoretical models, with the recent advancements in computational techniques, cryptographic methods are vulnerable to attacks. A novel technique for securing data was introduced using the biological structure of DNA called DNA Computing. It was invented by Leonard Max Adleman in the year 1994, for solving the complex problems such as directed Hamilton path problem, NP-complete problem and travelling salesman problem. 1 http://www.symantec.com/connect/blogs/2015-internet-security-threat-report-attackers-are-bigger-bolder-and-faster International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 40 Fig.1. Cell Definition The DNA computing is also used in cryptography research. DNA is used in cryptography for storing and transmitting the information, as well as for computation. This paper, try to utilize the power of DNA to hide message [5]. The massive parallelism and extraordinary information density inherent in this molecule are exploited for cryptographic purposes. The current main difficulties of DNA cryptography are the absence of theoretical basis, the high tech lab requirements and computation limitations [6]. In this paper DNA cryptography is used along with cellular automata. A brief introduction of cellular automation is described in the below paragraph. Cellular automation is a system made up of many discrete cells, each of which may be in one of a finite number of states [16]. A cell or automaton may change state only at fixed, regular intervals, and only in accordance with fixed rules [20] that depend on cells own values and the values of neighbors within certain proximity. The formal definition of Cellular Automation (CA) is one or two dimensional grid [10] of identical automata cells. Each cell processes information and proceeds in its action depending on its neighbors. Each cell (automaton) is defined by • Set of States S = {S 1 , S 2 , S 3 ,S 4 …, S N } • Transition Rules T • Therefore A ~ (S,T,R) (R: neighboring automata) • T: (S t , I t ) → S t+1 The rest of the paper is organized as follows: Section 2 of this paper deals with related work in DNA cryptography. The proposed work and experiment methodology are discussed in section 3. Experiment results of the proposed work are presented in section 4. Section 5 concludes the paper. 2. L ITERATURE S URVEY This section describes some of the DNA computing and cellular automata methods reported in literature and it also describes the motivation to combine DNA computing and cellular automata. According to Guangzhao Cui et al [8], DNA computing is a new computational paradigm by harnessing the potential massive parallelism, high density information of bio-molecules and low power consumption, which brings potential challenges and opportunities to traditional cryptography. Guangzhao Cui et al [8] in his paper discussed about DNA computing and its application in information security field and reviews the principle of DNA computing. It also elaborates about schemes with secret key searching and introduces the application of DNA computing in encryption, steganography and authentication [8]. S.Jeevidha et al [21] proposed an analysis on DNA based cryptography to transfer the data securely and also discusses DNA technology used in the cryptographic methods. DNA technology includes all biological process like polymerase chain reaction and electrophoresis. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 41 H.Z.Hsu [7] in his method exploits the interesting properties of DNA sequence to encrypt data. The author presents three methods namely, the insertion method, the complementary pair method and the substitution method. In [8] the paper analyse some schemes with secret key searching and introduce the application of DNA computing in encryption, steganography and authentication. Anupriya et.al [9] in her paper proposed an algorithm which is based on reference sequence known only to sender and receiver. Since there are roughly 55 million publicly available DNA sequence it is virtually impossible to crack the sequence. A Study of Data Encryption Standard (DES) Algorithm with Cellular Automata was proposed by Rama R et al [11]. DES algorithm was accepted as the standard encryption algorithm in 1976. Since then the development and usage of DES is widely recognized. The concept of Cellular Automata (CA) introduced by John von Neumann (1950s) is the first model which exhibited the capability of producing complex and random behavior. Cellular Automata is a parallel processing machine and it involves context dependent behavior of all the value at time ‘t’ [17]. The literature has witnessed the usage of Cellular Automata rules which are simple in nature for the operation of DES and Advanced Encryption Standard (AES) algorithms. Cellular automata is applied in variety of fields such as smart devices to provide security [14], pattern classification [15], traffic modelling etc. Rama R et al [11] extend the usage of Cellular Automata (CA) for the key generation in DES. The proposed CA-based key generation methodology uses the elementary Cellular Automata rule 30 which possess more randomness. The initial seed involved is 15 bit multi-seed and presented the implemented DES with key generated using Cellular automata in Java. Due to cellular automata rule 30 randomness is highly provided in this method. The demerit of this method is the key size is minimal in this approach. Pratibha Sharma et al [12] proposed Text Security Using 2D Cellular Automata Rules Encryption method which is the most common method for hiding text from unauthorized access. Encryption provides only one level of security during transmission over the channel. The main aim would be to provide 2 levels of security. First level comprises of hiding text to be sent behind some image using password and the second level comprises of encryption using 2D Cellular rules. If one level security is broken then the other level would provide security thereby ensuring more security to the transmitted message. Encryption would be done using 2 dimensional rules of Cellular Automata. Moore neighborhood model have been used as Cellular Automata model. In Moore neighborhood model 9 cells are considered at a time including the cell itself. The value of current cell (i.e. central cell) depends on its 8 neighbors. The merits of this method are: two level securities are provided in this method and the other is due to chip in automata machine which enhances the speed of encryption and decryption. Triple Stage DNA Cryptography Using Sequential Machine was proposed by Rupali Soni et al [13], to secure data there are some traditional cryptographic techniques such as substitution, transposition, and various algorithms such as RSA, DES, AES, IDEA etc. and protocols like SSL are used. This work in [13] represents a new approach to secure the data which is based on DNA logic, cryptographic scheme and automata theory. DNA Cryptography technique was proposed by L. M. Adleman in 1994 for solving Hamiltonian path problem and automata theory was firstly initiated in 20th century. In the proposed work [13] these two techniques are combined for creating powerful security algorithm. This algorithm is more reliable and powerful because of the use of secret key, auto generated Moore machine and use of password [13]. The merits of the method are security is provided in triple stage in this method and it is resist to many known attacks. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 2.1. Merging of DNA Cryptography and Cellular Automata DNA cryptography and cellular automata used in the cryptography field, but both are not combined to provide security. As both domains are yielding better solutions to the security issues, if these fields are combined, security provided by them will be enha this paper combines both fields in an efficient way for developing highly secured system. As cellular automata are dynamic in nature it contributes for dynamic behavior in the system which increases the overall security. DNA cryp These qualities are most impressive which motivates to provide a highly secured security system for the users. The ultimate aim of this proposed work is to provide high level security which may make the h ackers less possible to crack the system. 3. P ROPOSED S YSTEM The objective of the work is to develop a prominent secure system that satisfies the needs of the user. The DNA cryptography and cellular automata will lead to provide effective cryptographic algorithms. Each field has their own characteristics to establish if they are combined to form a cryptosystem it will lead to achieve properties of both fields. technique of both DNA cryptography and cellular automata and it is advanced encryption algorithm using DNA substitutaion combined with Elementary Cellular automata rule (ECA51)” [18] [19]. 3.1. Working of proposed system Initially the plain text is taken from the user either in file or direct input. convert the data in the file to DNA sequence until the end of file. Then convert the DNA sequence into binary file and divide the binary sequences into parts and perform xor operation on the binary parts with the help of key on of xor operation i.e binary sequence into DNA sequence and send to automata machine. Here with the help of automata machine the DNA sequence is convert into DNA code words which is in the form o f text and number and in unrecognized form. Thus along with the result output embedded the key value and its generation used as cipher text to the receiver. The algorithm flow is shown in the architecture diagram shown in fig 2. Fig.2. Pictorial illust International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 Merging of DNA Cryptography and Cellular Automata DNA cryptography and cellular automata used in the cryptography field, but both are not combined to provide security. As both domains are yielding better solutions to the security issues, if these fields are combined, security provided by them will be enha nced. The proposed work in this paper combines both fields in an efficient way for developing highly secured system. As dynamic in nature it contributes for dynamic behavior in the system which increases the overall security. DNA cryp tography provides a secure way to encrypt the data. These qualities are most impressive which motivates to provide a highly secured security system for the users. The ultimate aim of this proposed work is to provide high level security which may ackers less possible to crack the system. The objective of the work is to develop a prominent secure system that satisfies the needs of the user. The DNA cryptography and cellular automata will lead to provide effective cryptographic algorithms. Each field has their own characteristics to establish if they are combined to form a cryptosystem it will lead to achieve properties of both fields. This algorithm comprises the technique of both DNA cryptography and cellular automata and it is named as “A hybrid advanced encryption algorithm using DNA substitutaion combined with Elementary Cellular 51)” [18] [19]. Working of proposed system Initially the plain text is taken from the user either in file or direct input. The input file is intially convert the data in the file to DNA sequence until the end of file. Then convert the DNA sequence into binary file and divide the binary sequences into parts and perform xor operation on the binary parts with the help of key on the n parts after xor operation. Convert the resultant value of xor operation i.e binary sequence into DNA sequence and send to automata machine. Here with the help of automata machine the DNA sequence is convert into DNA code words which is f text and number and in unrecognized form. Thus along with the result output embedded the key value and its generation used as cipher text to the receiver. The algorithm flow is shown in the architecture diagram shown in fig 2. Fig.2. Pictorial illust ration of DNA based Cellular Automata System International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 42 DNA cryptography and cellular automata used in the cryptography field, but both are not combined to provide security. As both domains are yielding better solutions to the security issues, nced. The proposed work in this paper combines both fields in an efficient way for developing highly secured system. As dynamic in nature it contributes for dynamic behavior in the system which tography provides a secure way to encrypt the data. These qualities are most impressive which motivates to provide a highly secured security system for the users. The ultimate aim of this proposed work is to provide high level security which may The objective of the work is to develop a prominent secure system that satisfies the needs of the user. The DNA cryptography and cellular automata will lead to provide effective cryptographic algorithms. Each field has their own characteristics to establish if they are combined to form a This algorithm comprises the named as “A hybrid advanced encryption algorithm using DNA substitutaion combined with Elementary Cellular The input file is intially convert the data in the file to DNA sequence until the end of file. Then convert the DNA sequence into binary file and divide the binary sequences into parts and perform xor operation on the n parts after xor operation. Convert the resultant value of xor operation i.e binary sequence into DNA sequence and send to automata machine. Here with the help of automata machine the DNA sequence is convert into DNA code words which is f text and number and in unrecognized form. Thus along with the result output embedded the key value and its generation used as cipher text to the receiver. The algorithm flow International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 43 3.2. Proposed Algorithm The proposed algorithm is detailed below in the form of pseudocode. Input: F[file of any type], A[Automata for Plain text to AGTC data generation], c1[code book1], c2[code book2], c3[code book3], c4[code book4] Output: File F with the AGTC coded DNA sequence Algorithm : Encryption Step 1: The File F is converted into DNA sequence: F→ F[DNA(U)] ; // AGTC sequence file. Step 2: In the File F[DNA] convert the Uracil into Thymine acid F[DNA(U)] → F[DNA(T)]; Step 3:The DNA sequence is converted into binary sequence, The DNA output file of F[DNA(T)]→Binary[F[DNA(T)]]; // Binary file Step 4: Split the Binary Sequence into equivalent parts: Binary[F[DNA(T)]]→ {Binary[F1[DNA(T)]], → Binary[F2[DNA(T)]], → Binary[F3[DNA(T)]], → Binary[F4[DNA(T)]]} Step 5: Perform XOR operation with help of keys K1, K2, K3, K4 for all Binary Sequences F1, F2, F3, F4. // key generated and used. X1=(Binary[F1[DNA(T)]]) (XOR)K1 X2=(Binary[F2[DNA(T)]]) (XOR)K2 X3=(Binary[F3[DNA(T)]]) (XOR) K3 X4=(Binary[F4[DNA(T)]]) (XOR) K4 Step 6: Combine the all output of the binary sequence into a single sequence. X= X1+X2+X3+X4; Step 7: Apply the rule 51 for the resultant binary sequence X. Now the state of x in time t is X(t). RULE 51 (00110011) → s(t+1) = NOT(s_currentCell(t)) Next State of X(t) is X(t+1). X(t+1) = Rule 51(X(t)); After applying rule, resultant will be binary sequence. // Initial state t is transformed into t+1 next state. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 3.3. Implementation of Algorithm The algorithm is implemented in three the user is converted into DNA sequence by the codebook1 and codebook 2 given below in table 1 and table 2 respectively. This code book 1 and 2 converts the plain text into DNA Sequence for per operation of the algorithm. Code Book 2 Then convert DNA sequence contains uracil into thymine component, the now converted DNA sequence is converted into binary sequence by based on the codebook 3 given below. By using the codebook 3 the DNA sequence is converted into binary sequence, for performing step 4,5and 6. Where it converts into binary sequence and perform xor operation with key. Code Book 3 Apply rule 51 of cellular automata [20] for the binary sequence, and convert the bi into DNA code word using the table 3 and Send the AGTC data to Automata machine and with the help of the code book 4 and automata machine convert the DNA sequence into DNA code word. Table 1: Code Book 1 Step 8: Convert the Binary sequence into DNA sequence X(t+1) → X[DNA(t+1)]; S= X[DNA(t+1)] Step 9: In the Sequence S convert the Thymine(T) into S→ S(T); Step 10:Send the S(T) → Automata Machine A[S(T)]; // DNA code word is generated Step 11: Repeat the above steps until reach the end of the file; Step 12: The Encrypted Data is the output of the automata machine A[S(T)]→ Encrypted Se quence; Step 13: Append the Key in the output file at the end Encrypted File[F] → Encrypted Sequence + key; This output file Contains the DNA code sequence. End Algorithm International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 Implementation of Algorithm The algorithm is implemented in three stages, initially the plain text which is taken as input from the user is converted into DNA sequence by the codebook1 and codebook 2 given below in table This code book 1 and 2 converts the plain text into DNA Sequence for per forming the step 1 Then convert DNA sequence contains uracil into thymine component, the now converted DNA sequence is converted into binary sequence by based on the codebook 3 given below. 3 the DNA sequence is converted into binary sequence, for performing step 4,5and 6. Where it converts into binary sequence and perform xor operation with key. Apply rule 51 of cellular automata [20] for the binary sequence, and convert the bi nary sequence into DNA code word using the table 3 and Send the AGTC data to Automata machine and with the help of the code book 4 and automata machine convert the DNA sequence into DNA code Table 1: Code Book 1 - Plain text to DNA Sequence Convert the Binary sequence into DNA sequence → X[DNA(t+1)]; In the Sequence S convert the Thymine(T) into Uracil(U) acid. → Automata Machine A[S(T)]; // DNA code word is Repeat the above steps until reach the end of the file; The Encrypted Data is the output of the automata machine quence; Append the Key in the output file at the end → Encrypted Sequence + key; This output file Contains the DNA code sequence. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 44 stages, initially the plain text which is taken as input from the user is converted into DNA sequence by the codebook1 and codebook 2 given below in table forming the step 1 Then convert DNA sequence contains uracil into thymine component, the now converted DNA sequence is converted into binary sequence by based on the codebook 3 given below. 3 the DNA sequence is converted into binary sequence, for performing step 4,5and 6. Where it converts into binary sequence and perform xor operation with key. nary sequence into DNA code word using the table 3 and Send the AGTC data to Automata machine and with the help of the code book 4 and automata machine convert the DNA sequence into DNA code International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 Table 2: Code Book 2 Table 3: Code Book 3 International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 2: Code Book 2 - Numbers and Special Characters to DNA Sequence Table 3: Code Book 3 - DNA Sequence to Binary Sequence International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 45 International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 Table 4: Code Book 4 Finally append the key with the DNA code word and send to receiver i.e. encrypted file. 4. R ESULTS A ND D ISCUSSION The proposed algorithm is implemented using Java and MySql as back end. The sample screen shot of the implementation is shown in Fig.3. decrypted values are stored. The following are the computational parameters which have been considered for evaluation for the proposed algorithm, i) Encryption and decryption time consumed by the data (bits) ii) Space utilization (MB) iii) Throughput Table 5: Comparing the data based on the Types of Algorithm Computation time (s) of data (bits) 8 16 32 RSA 50 52 54 AES 48 49 50 DES 49 49 51 Proposed Method 50 50 51 The encryption time is nothing but the time taken by the data to be encrypted and then to be stored in the database similarly the decryption time is the time taken to obtain the plain text from International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 Table 4: Code Book 4 - DNA code word Finally append the key with the DNA code word and send to receiver i.e. encrypted ISCUSSION The proposed algorithm is implemented using Java and MySql as back end. The sample screen shot of the implementation is shown in Fig.3. The database is created and the encrypted and The following are the computational parameters which have been considered for evaluation for the proposed algorithm, Encryption and decryption time consumed by the data (bits) Table 5: Comparing the data based on the encryption time and storage Computation time (s) of data (bits) Average time (ms) Throughput (ms) Storage (MB) 64 128 256 56 58 60 55 9.163 82.26 50 53 56 51 9.881 82.15 51 52 55 51.16 9.853 82.35 51 52 53 50.26 10.02 82.03 The encryption time is nothing but the time taken by the data to be encrypted and then to be stored in the database similarly the decryption time is the time taken to obtain the plain text from International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 46 Finally append the key with the DNA code word and send to receiver i.e. encrypted text data or The proposed algorithm is implemented using Java and MySql as back end. The sample screen The database is created and the encrypted and The following are the computational parameters which have been Storage (MB) 82.26 82.15 82.35 82.03 The encryption time is nothing but the time taken by the data to be encrypted and then to be stored in the database similarly the decryption time is the time taken to obtain the plain text from International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 the cipher text. The standard algorithms which are taken for t DES algorithms. The parameters which are taken for comparison are average encryption and decryption time, throughput and storage. The comparison of the well proposed algorithm is given in table 5. As the bits individually, so the values of the database was assigned as zero initially and after aggregating all the set of the bits the final value was observed. To calculate the computation time of dat a different block size in bits are taken into account, they are 8,16,32,64,128 and 256 bits. The throughput can be obtained by dividing the average time with the total size of the data bit and it is expressed by the formulae: Where, η is throughput, D is data and T is the average time consumed by the data. Encryption The Plain text given by the user is encrypted using this algorithm effectively by applying all the steps discussed in the algorithm and rule 51 is also applied to generate the encrypted data. Result Analysis The result based on the throughput obtained for the different algorithms during the encryption process in shown in fig 4. It denotes that the DNA based Cellular Automata cryptographic algorithm is having better performance than the other cryptographic algorithm. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 the cipher text. The standard algorithms which are taken for t he comparison are RSA, AES and DES algorithms. The parameters which are taken for comparison are average encryption and decryption time, throughput and storage. The comparison of the well - known algorithms with the proposed algorithm is given in table 5. As the storage size shows slight deviation on considering the bits individually, so the values of the database was assigned as zero initially and after aggregating all the set of the bits the final value was observed. To calculate the computation time a different block size in bits are taken into account, they are 8,16,32,64,128 and 256 bits. The throughput can be obtained by dividing the average time with the total size of the data bit and T D n i n ∑ = = 0 η is throughput, D is data and T is the average time consumed by the data. The Plain text given by the user is encrypted using this algorithm effectively by applying all the in the algorithm and rule 51 is also applied to generate the encrypted data. Fig.3. Encryption output The result based on the throughput obtained for the different algorithms during the encryption denotes that the DNA based Cellular Automata cryptographic algorithm is having better performance than the other cryptographic algorithm. International Journal of Security, Privacy and Trust Management (IJSPTM) Vol 4, No 3/4, November 2015 47 he comparison are RSA, AES and DES algorithms. The parameters which are taken for comparison are average encryption and known algorithms with the the storage size shows slight deviation on considering the bits individually, so the values of the database was assigned as zero initially and after aggregating all the set of the bits the final value was observed. To calculate the computation time a different block size in bits are taken into account, they are 8,16,32,64,128 and 256 bits. The throughput can be obtained by dividing the average time with the total size of the data bit and The Plain text given by the user is encrypted using this algorithm effectively by applying all the in the algorithm and rule 51 is also applied to generate the encrypted data. The result based on the throughput obtained for the different algorithms during the encryption denotes that the DNA based Cellular Automata cryptographic