An Enhanced technique of Self-Correcting Localization Algorithm for Vehicular Node Position Accuracy in the Distributed VANET

Abstract

Applications for intelligent transportation utilize the Global Positioning Systems (GPS) signals on the Vehicular Ad-hoc Networks (VANETs) to improve road safety, traffic management and transportation system efficiency. However, Vehicular Nodes (VNs) using those applications may face the deterioration or complete loss of GPS signals due to many reasons such as changes in node velocity and/or positioning, dense foliage, compact high buildings and/or distance between vehicular nodes. Although several solutions have been developed to solve this issue through internal communication with surrounding vehicle nodes (i.e., beacons), the VNs remain to suffer the poor positioning accuracy or errors in localization estimation. In this paper, an Extended Self-Correcting Localization (ESCL-VNET) algorithm is developed to enhance the positioning accuracy and improve the localization estimation of a given VN over the distributed VANET. The technique integrates the use of Received Signal Strength Indication (RSSI) technique to enable the VNs to estimate their locations and the use of Signal to Interference Noise Ratio (SINR) values obtained through the Dedicated Short-Range Communications (DSRC) messaging by the other VNs to weight the localizations using the Weighted Centroid Localization (WCL) process. A simulation program was developed to conduct performance evaluation of the ESCL-VNET algorithm and to assess the given results. The simulation shows that the new ESCL-VNET algorithm can generate a more accurate position estimation or localization than in the case of the standard SCL-VNET. The ESCL-VNET algorithm may contribute to the development of better and more efficient localization applications as part of robust Intelligent Transportation System (ITS).