Position-based routing protocol using Kalman filter as a prediction module for vehicular ad hoc networks

Abstract

A vehicle’s Global Positioning System (GPS) location is used as the Location_ID in position-based routing protocols. The accuracy of GPS is a major concern in vehicular ad hoc networks (VANETs), as it is influenced by environmental and technical factors. Therefore, a vehicle’s actual location is different from its GPS location, with a margin of error of plus or minus 5–100 m. To reduce the effect of the margin of error on routing, this study evaluated the position-based routing protocol by using the Kalman filter (KF) and extended Kalman filter (EKF). Each of these was utilized individually as a prediction module for the purpose of improving the average delay (AD), packet delivery ratio (PDR), and throughput by minimizing location error. The KF and EKF prediction modules were implemented using C++ programming and the Eigen library, which were further incorporated into an NS-3.23 simulator. The proposed routing protocol performance was compared with cross-layer, weighted, and position-based routing (CLWPR) protocol in terms of PDR, AD, and throughput.