Study on Predictive Maintenance based Approaches for Designing Routing Protocols in Wireless Sensor Network

Abstract

Wireless sensor network (WSN) extends its applications to a wide field due to its cost efficiency, ease of installation, high reliability and improved throughput of the networks. There are various factors that may cause the network configuration to change dynamically and needs to be redesigned frequently. Maintaining large number of nodes without decreasing the scalability and efficiency of the network can be achieved without emphatic redesigning by machine learning (ML) making the computation more effective. Deep learning also falls under of machine learning category that helps to deal with formation of test data sets to be used in several applications including WSN. In this paper we shall discuss several machine learning (ML) and deep learning (DL)strategies for utilizing them in formation of routing protocols for better node synchronization, network congestion control, lifetime enhancement of the node intermediate network and for overall improved system performance.

Keywords—WSN, Machine Learning, Deep Learning