Neural cryptography is a new way to create shared secret key. It is based on synchronization of Tree Parity Machines (TPM) by mutual learning. Two neural networks trained on their mutual output bits synchronize to a state with identical time dependent weights. This has been used for creation of a secure cryptographic secret key using a public channel. In this paper a genetic approach has been used in the field of neural cryptography for synchronizing tree parity machines by mutual learning process. Here a best fit weight vector is found using a genetic algorithm and then the training process is done for the feed forward network. The proposed approach improves the process of synchronization.
Dr. S. Santhanalakshmi, Sudarshan, T. S. B., and Patra, G. K., “Neural Synchronization by Mutual Learning Using Genetic Approach for Secure Key Generation”, in International Conference on Security in Computer Networks Distributed Systems (SNDS-2012), IITM-K Techno park at Trivandrum, 2012, vol. 335, pp. 422-431.