Performance Evaluation of Data Scheduling in IEEE802.16e in VANET Traffic Context

Abstract

VANET (Vehicular Ad hoc Network) have become an exciting research going in the era of smart cities as a means for increasing traffic safety and comfort. Nevertheless, VANETs possess some particular characteristics such as very high mobility, encountering rapid changes of topology, frequent disconnected network and variable network density. Gathered VANET traffic will be transferred from road side unit to base station which is responsible of routing traffic to the data centre using long-range support like WiMax( Worldwide Interoperability for Microwave Access) in smart city solutions. As VANET have a special network behaviours and stringent traffic physiognomy, this work aims to determinate the most suitable WiMax scheduler to support VANET traffic characteristics. For this purpose, we used in this work a VANET traffic model as input to our NS3 simulation scenarios. In this paper, we evaluate three basic scheduling algorithms (simple scheduler, simple priority, based real time polling service scheduler and migration based uplink scheduler), mainly at the base station level in WiMax. We analyse their performance in the case of VANET traffic context in order to find the best suitable scheduler in terms of throughput, latency, and reliability. Results demonstrate that the simple priority based first come first served scheduler(simple scheduler) and a based real time polling service scheduler (scheduler RTPS) outperform the migration based uplink scheduler (scheduler MBQOS) when the number of nodes less than or equal to 20. On the other hand, the scheduler MBQOS is more effective for higher network density and favors the real-time traffic in terms of reliability and delay.