Paper Title :Formation Of Information Security System Using Artificial Neural Networks And Genetic Algorithms Applying Elgamal Public Key Encryption Algorithm

Author :Harshada Sawant, Revati Salvekar, Dipti Patil, Aishwarya Pandey

Article Citation :Harshada Sawant ,Revati Salvekar ,Dipti Patil ,Aishwarya Pandey , (2015 ) " Formation Of Information Security System Using Artificial Neural Networks And Genetic Algorithms Applying Elgamal Public Key Encryption Algorithm " , International Journal of Advance Computational Engineering and Networking (IJACEN) , pp. 144-147, Volume-3, Issue-6

Abstract : Cryptography is that the science of writing secretly code associate degree is an ancient art. By creating use of computing (AI), Human Intelligence are often simulated by a machine, Neural Networks is one of the sub field of AI. Artificial Neural Networks (ANN) consists of nerve cells and weights assigned to lay neuron connections helps in storing the noninheritable data. This paper makes use of Hebbian learning rule to coach the ANN of each sender and receiver machines. within the field of Public Key Cryptography (PKC), Pseudo Random Number Generator (PRNG) area unit wide want to generate distinctive keys and random numbers utilized in ANN that area unit found to possess many varieties of possible attacks. It's essential for a key to possess randomness for key strength and security. This paper proposes the concept of key generation for PKC by combination of ANN and Genetic Algorithm (GA).It absolutely was detected that use of ANN together with GA has not thus far been explored. GA approach is usually applied for obtaining improvement and solutions in search issues. GA correlates to the character to an outsized extent manufacturing population of numbers wherever variety possessing higher fitness worth is replicated a lot of. Thus, creating GA a really sensible competitor for PRNGs. Sensible Fitness perform helps in exploring search area of random numbers in additional economical manner. GA PRNGs result samples satisfies frequency take a look at and gap take a look at. therefore the numbers generated when every iteration by GA PRNG area unit statistically verified to be random and nonrepeating, having no previous relation of next variety from the previous ones, acting as a vital initialization parameter for neural algorithmic rule overcomes the problem of acknowledging the random variety generated by traditional PRNG. Elgamal Public Key Encryption Algorithm was used for encryption and decryption of user’s data.Our algorithmic rule was determined to grant quick and improved performance results having sensible and possible implementation. Keywords-Artificial Neural Networks, Elgamal Algorithm, Genetic Algorithm, Hebbian Learning Rule

Type : Research paper

Published : Volume-3, Issue-6

	|		PDF	|	Viewed - 76	|	Published on 2015-06-17