Localization of vehicular ad-hoc networks with RSS based distance estimation

Abstract

Location information of a vehicle provides numerous applications such as, emergency calling, navigation, vehicle tracking and other location based services. Vehicles localization in vehicular adhoc networks (VANETs) in urban scenarios is a key issue for public safety applications. Motivated by localization of vehicles in urban areas for public safety, where the de facto standard solution global positioning system (GPS) does not provide the required localization accuracy, a local closed form solution is proposed for VANETs localization exploiting the communication with road side units (RSUs). In proposed technique the vehicle receives signals from the RSUs within its range, and computes the average receive signal strength (RSS) from each RSU. The average RSS measurements are fed to the proposed closed form localization algorithm which computes the the vehicle position. The proposed algorithm only take the RSS measurements from the closer RSUs with higher signal to noise ratio, which results in better location estimation. The performance of the proposed closed form solution is analyzed by deriving its Cramer Rao lower bound. Numerous simulations are performed to show that the proposed RSS based closed form solution outperforms the least square and weighted least square techniques.