An Efficient Intrusion Detection Model Using Adapting Boosting With Uninterrupted Bayesian Time Mobile Network

Mobile Ad-hoc network (MANET) is a wireless system without any infrastructure that consists of mobile nodes such as mobile phones, laptops and Personal Digital Assistants (PDA). The network system with mobile nodes becomes more difficult when an attack is said to occur, resulting in unsecure network path. As a result, a detection system needs to be developed to overcome the intrusion issue. Many secure payment and trust management schemes were introduced with the objective of minimizing the intrusion reducing the computation overhead and improving the reliability. However, an intelligent intrusion detection mechanism is necessary to address the security issues and combat against collusion attack. In this paper, to perform an intelligent intrusion detection model, Adaptive Boosting with Uninterrupted Bayesian Time Mobile (AB-UBTM) Networks is developed. Adaptive boosting combines the spanning tree classifier with Uninterrupted Bayesian Timed Mobile nodes in ad hoc network, which in turn increase resource utilization factor. Uninterrupted Bayesian Time Mobile algorithm initializes the weight, depending on the node's class state in the spanning tree and improves the trust accuracy across the ad hoc network.