Gravitational Search-Based Efficient Multilayer Artificial Neural Coordination

In this paper, a Gravitational Search-based neural weight optimization technique for faster neural synchronization is proposed. To share the key over a public channel, two neural networks are coordinated by mutual learning. The primary problem of neural synchronization in the absence of a weight vector from another party is deciding how to determine the synchronization of two communication parties. The existing synchronization evaluation strategies also have a latency problem that affects the protection and privacy of neural synchronization. A significant technique for evaluating synchronization is introduced in this paper to evaluate the complete coordination of two neural networks more effectively and rapidly. The frequency of the two networks having the same output in previous iterations is used to measure the degree of synchronization. When a certain threshold is reached, the hash is used to decide if both networks are fully synchronized. This proposed methodology uses Gravitational Search optimized weight vectors to achieve total synchronization between two communicating parties. Unlike existing approaches, the proposed solution helps two communication parties to detect complete synchronization faster. The successful geometric probability is decreased as a consequence. As a result, the proposed method strengthens the security of the neural key exchange protocol. Different parametric experiments have been performed on the proposed methodology. In terms of the cited findings in the study, simulations of the procedure indicate efficacy.