Implementation of Machine Learning Classifier for DTN Routing

Abstract

This paper presents, better routing method in Delay Tolerant Network using Machine learning. Delay Tolerant Network is a wireless network, in which nodes are changing its positions dynamically in an unexpected way due to that Round trip time and error rates are very high. Examples are Disaster area, under the sea, Space communication, etc. In the proposed method neighbouring nodes are predicted by machine learning classifiers. These nodes use message history delivery information to deliver the message on destination. With the help of Bundle protocol implementation IBR-DTN [3], collects network traffic status and real-world location trace. These information uses to emulate DTN environment by Common Open Research Emulator (CORE) [2]. The new application is used to predict the results, preparation for the network history data, analysis and classification-based routing.