OWL: Optimized weighted localization for vehicular ad hoc networks

Vehicular Ad Hoc Networks (VANETs) allows the exchange of messages between neighboring vehicles or roadside units (RSU). The performance of VANET applications is subject to the ability of determining the location of the vehicles at anytime and anywhere within the network, and thus demands real-time, precise position of vehicles. Accordingly, a number of methods and protocols have been proposed to fulfill this requirement, however, the accuracy of the obtained location is not sufficient for VANET safety applications. We propose an enhancement to our previous localization method, weighted localization using distance information (WLD), that uses signal to interference-noise ratio (SINR) obtained from the exchange messages, and distance between the neighboring vehicles to assist vehicles in estimating their positions. Our proposed optimized weighted localization (OWL) uses the heading information shared by the neighboring vehicles in addition to the SINR and distance information. Our proposed method is an extension to centroid localization (CL) with a weight assigned to each of the neighboring vehicles´ coordinates, based on SINR values, distance and heading. We implement a simulation program to evaluate the proposed method against CL, WLD and relative span weighted localization (RWL). The results show the proposed method to have better performance and consistently less average location errors in varying densities.