Detecting and Predicting Malicious Nodes in Mobile Ad-Hoc Networks using a Secure Technique

Abstract

Mobile Ad Hoc Networking (MANET) is a rapidly growing area of interest in the realm of communication frameworks. Because the MANET lacks a basis, it exhibits the dynamic character of a self-assertive network architecture. Security concerns are critical in these networks. Nodes in MANETs may launch a variety of attacks or become conspicuously self-centred to maintain their advantage. These nodes may be considered malicious. Identification of such malicious nodes is critical for the successful operation of MANETs. A collection of networks has been presented, but each one has its own set of constraints. The scope of this proposal is to conduct research on black hole, wormhole, collaborative malevolent, and flooding attacks, and to establish a network of counteractive measures using responsive directing conventions. For execution analysis and replication, an AODV, NS-2 organised test network is used. To prevent black hole, wormhole, malevolent, and flooding attacks, a countermeasure is employed that calculates the Trust value based on the route request, route response, and information packet. Following the count, place stock in values ranging from 0 to 1. If the trust esteem is more than 0.5, the node is solid and permits access to the network as a whole. The suggested convention, secure Ad hoc On-demand Distance Vector (SAODV), is evaluated in terms of network performance. When compared to the standard AODV convention, the results reveal a notable difference in execution. By increasing the duration of a dip in throughput, SAODV achieves a throughput superior to that of the joint malicious assault AODV and the current protocol. SAODV's packet delivery ratio is superior to that of the joint malicious attack AODV and the established AODV protocol. SAODV's end-to-end delay is superior to the joint malicious attack AODV and the current AODV protocol.