An adaptive vulture based deep belief mechanism for searching user's resources in mobile P2P environment.

Mobile devices are pervasive in today's world, and the ability to create ad-hoc networks among these devices enables increased connectivity. Peer-to-peer (P2P) Mobile Ad-hoc Network (P2P MANET) offer flexibility and scalability. Devices can join or leave the network seamlessly, and the network can adapt to changing conditions. This makes them suitable for scenarios with varying numbers of devices or where the network topology changes frequently. Moreover, the P2P network is a robust, fault-tolerant, and distributed network for sharing the resources like CPU, memory, files, etc. However, solid mobility, malicious behavior and short wireless transmission range more risk to searching the mobile P2P network resources. In this paper, a novel Vulture-based Deep Belief Model (VbDBM) is designed to enhance the performance of resource search in a mobile P2P network. Also, it will neglect the attacks and errors are presented during the transmission stage. In addition, the VbDBM framework is implemented in the NS-2 tool, and there are nine nodes created for resource search. Initially, the developed technique analyzes the user needs and identifies the neighboring peer of the user location with the help of vulture fitness. At next, the proposed technique provides the resource to the volunteer node (neighboring peer). Finally, the volunteer node offers the resource to the user. The attained performance metrics of developed VbDBM are compared with other existing techniques in terms of search rate, simulation time, data transfer rate, number of Data, and throughput. From the comparison proposed algorithm has improves 350 kbps throughput and 28% search rate respectively. [ABSTRACT FROM AUTHOR]