Cryptography based on Artificial Neural Network
Shital Daulat Jagtap
ME (VLSI & ES) Dept. of E&TC
Junnar, Pune, India
Mr. P. Balaramudu
ME (VLSI & ES) Dept. of E&TC
Junnar, Pune, India
Abstract—Cryptography is the technique of changing data or information into unreadable for unauthorized persons. In cryptography process information transfer from sender to receiver in a manner that prevents from unauthorized third person. There are so many cryptography methods are available which based on number theory. The cryptography based on number system has some disadvantages such as large computational power, complexity and time consumption. To overcome all the disadvantages of number theory based cryptography, artificial neural network based cryptography used.
The ANN have many characteristics such as learning, generation, less data requirement, fast computation, ease of implementation and software and hardware availability. It is very useful for many applications.
Keywords—Cryptography, Artificial Neural Network, Decryption, Encryption, Key generation, Chaotic map.
Cryptography provides the information security for some useful applications such as in encryption, message digests and digital signature. A neural network is a machine which is designed for modeling the way in which the brain performs a particular task. Cryptography is defined as the exchange of data into mix code. Cryptography is also used in many applications such as computer passwords, ATM cards and electronic commerce.
Cryptography has two types of encryption data:
1. Symmetrical encryption
2. Asymmetrical encryption
1. Symmetrical encryption: Symmetrical encryption use the same key for encryption and decryption process and it defines secret-keys, shared keys and private keys.
2. Asymmetric encryption: Asymmetric cryptography uses different key for encryption and decryption process. It has pair of keys one for encryption and one for decryption.
II. BIOLOGICAL MODEL
The human nervous system can be broken down into three stages that may be represented as follows:
Fig.1. Block Diagram of a Human Nervous System.
The receptors collect information from the environment. The effectors generate interactions with the environment e.g. activate muscles. The flow of information/activation is represented by arrows. There is a hierarchy of interwoven levels of organization: Molecules and Ions Synapses, Neuronal microcircuits, Dendritic trees Neurons, Local circuits ,Inter-regional circuits, Central nervous system
III. ARTIFICIAL NEURAL NETWORK
Artificial Neural Network is an information processing and modeling system which mimics the learning ability of biological systems in understanding unknown process or its behavior.
ANN is configured for a specific application, such as pattern recognition or data classification, through a learning process. Learning in biological systems involves adjustments to the synaptic connections that exist between the neurons. This is true of ANNs as well.
ANNS have developed as generalizations of mathematical models of human cognition or neural biology. Based on the assumptions that :
1. Information procures at many simple elements called neuron.
2. Signals are passed between neurons over connection links.
3. Each connection link has associated weight. Which in a typical neural net multiplies the signal transmitted?
4. Each neuron applies an activation function usually nonlinear to its net input (sum of weighted input signals) to determine its output signal.
An Artificial Neural Network is a network of many very simple processors (units), each possibly having a (small amount of) local memory. The units are connected by unidirectional communication channels which carry numeric data. The units operate only on their local data and on the inputs they receive via the connections. The design motivation is what distinguishes neural networks from other mathematical techniques: A neural network is a processing device, either an algorithm, or actual hardware, whose design was motivated by the design and functioning of human brains and components .
There are many different types of Neural Networks, each of which has different strengths particular to their applications. The abilities of different networks can be related to their structure, dynamics and learning methods.
A. Cryptography Using Artificial Neural Network
Fig.2. The block diagram of trained ANN model.
The block diagram of the proposed ANN model is given in Figure 2 . As shown in the figure, three initial conditions and time variable were applied to the inputs and three chaotic dynamics ˆx, ˆy and ˆz were obtained from the outputs of the ANN. For the training and test phases of the ANN, approximately 1800 input-output data pairs which belong to 24 different initial condition sets were obtained from Equation (4). A quarter of those 1800 data pairs were sorted to use in the test phase and the rest of data were used in the training phase.
Cryptography is the most important field of computer security providing secure services. It is the process of transferring private data through open network communication. Earlier cryptography was considered the domain of military and governments only. Everywhere the use computers and the advent of internet has made it an integral part of our daily lives. Today cryptography is at the heart of many secure applications such as online banking, online shopping, online government services such personal income taxes, cellular phones, and wireless LANS (Local Area Networks) etc.
V. Requirement Of Cryptography
Cryptography is generally used in practice to provide four services: privacy, authentication, data integrity and non- repudiation. The goal of privacy is to ensure that communication between two parties remain secret. This often means that the contents of communication are secret; however in certain situations the of fact communication took place and must be a secret as well. Encryption is generally used to provide privacy in modern communication. Authentication of one or both parties during a communication is required to ensure that information is exchanged with the legitimate party. Passwords are common examples of one-way authentication in which users authenticate themselves to gain access to system.
VI. SYMMETRIC KEY ENCRYPTION
In symmetric key encryption a secret key is shared between the sender and receiver. The word "symmetric" refers to the fact that both sender and receiver use the same key to encrypt and decrypt the information.
Block Ciphers: A block cipher is symmetric key cryptographic primitive which takes as input an n-bit block of plain text and a secret key and outputs an n-bit block of cipher text using a fixed transformation. Figure 4 shows the general structure of a block cipher. The common block sizes are 64 bits, 128 bits and 256 bits. For a fixed key the block cipher defines a permutation on the n-bit input.
VII. LIMITATION OF CRYPTOSYSTEM
The limitations of this type of system are few, but potentially significant. This is effectively a secret-key system, with the key being the weights and architecture of the network. With the weights and the architecture, breaking the encryption becomes trivial. However, both the weights and the architecture are needed for encryption and decryption. Knowing only one or the other is not enough to break it.
VIII. ADVANTAGES OF CRYPTOSYSTEM
1. The advantages to this system are that it appears to be exceedingly difficult to break without knowledge of the methodology behind it.
2. It is tolerant to noise. Most messages cannot be altered by even one bit in a standard encryption scheme.
3. The system based on neural networks allows the encoded message to fluctuate and still be accurate.
4. Neural networks are ideal in recognizing diseases using scans since there is no need to provide a specific algorithm on how to identify the disease.
5. ANNs are used experimentally to implement electronic noses. Electronic noses have several potential applications in telemedicine. Telemedicine is the practice of medicine over long distances via a communication link.
6. There is a marketing application which has been integrated with a neural network system. The Airline Marketing Tactician (a trademark abbreviated as AMT) is a computer system made of various intelligent technologies including expert systems.
IX. RESULT AND CONCLUSION
A General n-state Sequential Machine
Sequential Machine was implemented as described in the last chapter. As an example, the serial adder was implemented using this machine. The following figures show different stages of the execution:
Figure 3: Entering the training data in the sequential machine
Fig.4. The plotted graph of the error function after the learning process
The data from the state table of the Serial Adder is entered into the program as shown in figure 3. The current state represents any previous carry that might be present whereas the next state represents the output carry. Thus, this sequential machine consists of 2 input, 1 output and 2 states. After the training data has been entered into the program, the back-propagation algorithm, to minimize the error function, executes. Figure 7 shows the plot of the error function against the number of iterations.
Most of the work is done to aim the cryptographic applications using artificial neural network. The work done in this paper can be used to obtained the data security from the other persons. Author Ms. Shital Daulat Jagtap thanks Prof.P. Bala Ramudu , Asst. Prof. And HOD of S.V.C.E.T., Rajuri, Pune for guiding throughout her work and also thanks to Principal Prof. Dr. Sanjay Bhaskar Zope and Vice-Principal Prof. Chanakya Kumar Jha of S.V.C.E.T., Rajuri, Pune for guiding throughout work.
 William Stallings, “Cryptography and Network Security: Principles and Practices”, second edition.
 Aloha Sinha, Kehar Singh, "A Technique for Image Encryption using Digital Signature", Optics Communications, Vol.2 No.8 (2203), 229-234.
 M. Zeghid, M. Machhout, L. Khriji, A. Baganne, R. Tourki, “A Modified AES Based Algorithm for Image Encryption”, World Academy of Science, Engineering and Technology 27 2007.
 K.Deergha Rao, Ch. Gangadhar, “Modified Chaotic Key-Based Algorithm for Image Encryption and its VLSI Realization”, IEEE, 15th International. Conference on Digital Signal Processing (DSP), 2007.
 Saroj Kumar Panigrahy, Bibhudendra Acharya, Debasish Jen, “Image Encryption Using Self-Invertible Key Matrix of Hill Cipher Algorithm”, 1st International Conference on Advances in Computing, Chikhli, India, 21-22 February 2008.
 Zhang Yun-peng, Liu Wei, Cao Shui-ping, Zhai Zheng-jun, Nie Xuan, Dai Wei-di, “Digital Image Encryption Algorithm Based on Chaos and Improved DES”, IEEE International Conference on Systems, Man and Cybernetics, 2009.
 Min Long, Li Tan, “A chaos-Based Data Encryption Algorithm for Image/Video”, IEEE, Second International Conference on Multimedia and Information Technology, 2010.
 HiralRathod, Mahendra Singh Sisodia, Sanjay Kumar Sharma, “Design and Implementation of Image Encryption Algorithm by using Block Based Symmetric Transformation Algorithm (Hyper Image Encryption Algorithm)” International Journal of Computer Technology and Electronics Engineering (IJCTEE), Vol.1, No.3 (2010/2011).
 Kuldeep Singh, Komalpreet Kaur, “Image Encryption using Chaotic Maps and DNA Addition Operation and Noise Effects on it”, International Journal of Computer Applications (0975 - 8887) Vol.23, No.6, June 2011.
 Qais H. Alsafasfeh, Aouda A. Arfoa, “Image Encryption Based on the General Approach for Multiple Chaotic Systems”, Journal of Signal and Information Processing, 2011.
...(download the rest of the essay above)