Neural Synchronization Using Genetic Algorithm for Secure Key Establishment

Abstract

Two neural networks that are trained on their mutual output synchronize to an identical time dependant weight vector. This novel phenomenon can be used for creation of a secure cryptographic secret-key using a public channel. Neural cryptography is a way to create shared secret key. Key generation in Tree Parity Machine neural network is done by mutual learning. Neural networks here receive common inputs and exchange their outputs. Adjusting discrete weights according to a suitable learning rule then leads to full synchronization in a finite number of steps and these identical weights are the secret key needed for encryption. A faster synchronization of the neural network has been achieved by generating the optimal weights for the sender and receiver from a genetic process. Here the best fit weight vector is found using a genetic algorithm. In this paper the performance of the genetic algorithm has been analysed by varying the number of hidden and input neurons.

Keywords

• Neural synchronization
• neural cryptography
• key exchange
• genetic algorithm
• tree parity machine. ___________________________________________________