A Simple and Cost-Effective Anomaly Detection Paradigm on the Basis of Computational Intelligence for Mobile Ad-Hoc Networks from a Security Viewpoint

Abstract

A Mobile Adhoc Network (MANET) comprises of wide range of applications, since past one decade it has offered a tremendous breakthrough in the field of wireless communication. Therefore, its potential cost-effective routing and resource scheduling operations have not only enhanced the communicational benefits but also made it a hot topic of interest among different scientific and research communities. However, despite having so many diverse potential aspects, it lacks efficiency in terms security owing to the dynamic and non-centralized way of topology formation while performing communication. It also often found undergoing through an increasingly diverse application requirement which made it resource constrained. Thereby, in this type of network formation, the probability of threat occurrence for the purpose of disrupting communication process is more which also affects the data reliability to a huge extent. This study inclines towards addressing the security issue which is still an open research problem in the domain of MANET and further introduces a cost-effective computational intelligence driven mechanism which can enhance the resistivity of network model by discretizing all form of intrusions. Identifying network intrusion within a specified time increases the network reliability factor which is quite essential in MANET data computation and analysis. The proposed formulated analytical design is a computational process and experimented in a numerical computing platform to validate the performance in terms of throughput and packet delivery ratio.