Route classification scheme based on covering rough set approach in mobile ad hoc network (CRS-MANET)
International Journal of Intelligent Unmanned Systems
ISSN: 2049-6427
Article publication date: 22 October 2019
Issue publication date: 4 May 2020
Abstract
Purpose
The purpose of this paper is to obtain correctly classified routes based on their parameters.
Design/methodology/approach
In this paper, a covering rough set (CRS) approach is proposed for route classification in wireless ad hoc networks. In a wireless network, mobile nodes are deployed randomly in a simulation region. This work addresses the problem of route classification.
Findings
The network parameters such as bandwidth, delay, packet byte rate and packet loss rate changes due to the frequent mobility of nodes lead to uncertainty in wireless networks. This type of uncertainty can be very well handled using a rough set concept. An ultimate aim of classification is to correctly predict the decision class for each instance in the data.
Originality/value
The traditional classification algorithms, named K-nearest neighbor, J48, general rough set theory, naive Bayes, JRIP and multilayer perceptron, are used in this work for comparison and for the proposed CRS based on route classification approach revealing better accuracy than traditional classification algorithms.
Keywords
Acknowledgements
The first author would like to thank the UGC-RGNF (Rajiv Gandhi National Fellowship) for financial support to do this work. The Grant No. is F1-17.1/2014-15/RGNF-2014-15-SC-TAM-84052 (SA-III/Website). Authors would like to take this opportunity to thank anonymous reviewers for the effort and expertise contributed by them.
Citation
T., S., H., H.I. and S., S.K. (2020), "Route classification scheme based on covering rough set approach in mobile ad hoc network (CRS-MANET)", International Journal of Intelligent Unmanned Systems, Vol. 8 No. 2, pp. 85-96. https://doi.org/10.1108/IJIUS-08-2019-0046
Publisher
:Emerald Publishing Limited
Copyright © 2019, Emerald Publishing Limited