CLASSIFICATION APPROACH-BASED SYBIL NODE DETECTION IN MOBILE AD HOC NETWORKS

Mobile Ad Hoc Network (MANET) is an auto-configuring network that is designed spontaneously by a mix of mobile nodes without the intervention of a centralised administration or fixed infrastructure, thanks to technological advancements. In MANET, Intrusion Detection Systems (IDS) must system blocks of packets with a variety of attributes that prevent anomalies from being detected. Sampling and Feature Selection can be used to reduce computing time and hence reduce the time it takes to detect intrusions. Those that create assaults on network, data link, application layer, and physical layer functioning are selfish and nasty. In this paper, three feature selection strategies and a suggested Artificial Neural Network (ANN) classification model are proposed to improve the classification of Sybil nodes in the MANET. The accuracy, precision, and recall of the Sybil node identification architecture, as measured by numerous evaluation measures. Other classification techniques, such as Support Vector Machine, are used to evaluate the proposed ANN architecture (SVM).