| Summary: | Vehicular Ad hoc Networks (VANETs) are specialized vehicle networks designed to provide diverse applications and services for vehicles in mobility. Nevertheless, VANETs face several challenges such as dynamic network structure, rapid changes in velocity of vehicles in various directions, data overload, and limited duration of connectivity, all of which can result in poor proficiency in delivering services to the vehicles. This hinders the progress of meeting the requirements of time-sensitive applications which could not tolerate much delay. Deploying a vehicular infrastructure such as the Roadside Unit (RSU) is crucial for improving the efficiency of data distribution in VANETs. The dynamic nature of the vehicle environment in VANET leads to constraints in data delivery, such as problems with high density and connectivity duration. Vehicles in high velocity can enter and exit the transmission range area of an RSU in Vehicle-to-Roadside (V2R) communication without obtaining all the data they requested. This sets added processing and load on the RSUs and leads to greater rates of dropped requests, delays, and reduced responsiveness. Furthermore, in extensive networks characterized by frequent and rapid changes in connectivity and dynamic network structure, such as urban highways, V2R communication experiences brief periods of connectivity when traveling at high speeds. Additionally, for V2R communication, vehicles does not receive any services when outside the Network Dwelling Range (NDR) of an RSU. This study introduces the Segmentation-Based Vehicle Location Detection (SVLD) mechanism that involves fragmenting the NDR of RSUs into overlapping segments and identifying vehicles within these segments. Experiments have been conducted in simulation using OMNeT++, SUMO and Veins with the goal of expanding the NDR and increase the duration of connectivity in V2R communication. Additionally, it aims to alleviate the burden on the RSU by facilitating a smooth transfer of workload to the next RSU. The results demonstrate a substantial expansion of the NDR using the RSU's NDR overlapping approach, as well as a reduction of 36.4% in the RSU's load.
|
|---|