Effective Classification and Handling of Incoming Data Packets in Mobile Ad Hoc Networks (MANETs) Using Random Forest Ensemble Technique (RF/ET)

Abstract

Mobile ad hoc network is a network type, where wireless network comprises independently moving nodes that cooperatively contribute towards making a network operate successfully. These nodes are tiny electronic devices operating on battery power and have limited operational resources, such as memory, buffer capacity and processing units. These nodes can act as client, server or router, depending upon the network requirement because of the absence of a centralized server control. MANET faces lots of challenges while maintaining a secured and resilient communication due to its limited resources and operational capacity. One of the serious challenges is handling and classifying tons of incoming data at a single point of time for efficient processing. As tons of data enter MANETs network at a single point of time, it is necessary to deploy a new mechanism not only for classification but also for further processing. The heavy and extensive data become unmanageable on the part of mobile nodes while implementing techniques like data analysis to trace anomaly or extracting the required information from the data pool incoming into MANETs. This unmanageable data calls for classification technique so as to segregate data into specific group that can be further utilized for implementing or formulating a specialized mechanism that tends to solve many other research problems. The objective of this paper is to classify the incoming data into MANETs and reduce the data set using the RF/ ET technique. The results showed that the proposed model attained an accuracy level of 86% in handling data packets in MANETs.