EpiDOL: epidemic density adaptive data dissemination exploiting opposite lane in VANETs

Abstract

Vehicular ad-hoc networks (VANETs) aim to increase the safety of passengers by making information available beyond the driver’s knowledge. The challenging properties of VANETs such as their dynamic behavior and intermittently connected feature need to be considered when designing a reliable communication protocol in a VANET. In this study, we propose an epidemic and density adaptive protocol for data dissemination in vehicular networks, namely EpiDOL, which utilizes the opposite lane capacity with novel probability functions. We evaluate the performance in terms of end-to-end delay, throughput, overhead and usage ratio of the opposite lane under different vehicular traffic densities via realistic simulations based on SUMO traces in ns-3 simulator. We found out that EpiDOL achieves more than 90% throughput in low densities, and without any additional load to the network 75% throughput in high densities. In terms of throughput EpiDOL outperforms the Edge-Aware and DV-CAST protocols 10% and 40% respectively.