A Map-based Sensor data Delivery Protocol for vehicular networks

摘要

Nowadays our vehicles incorporate multiple sensors that collect information not only about speed or fuel consumption, but also about weather (rain sensors), road state (ESP sensors), etc. Achieving a distributed traffic monitoring system that combines all this information gathered by different vehicles is a challenging task. In particular, we consider that vehicular ad-hoc networks (VANETs) are particularly suitable for this endeavor. VANETs present some characteristics such as a high mobility, a variable node density or frequent radio obstacles that are prone to generate network partitions. Under these conditions, a complete multi-hop path between the sender and the receiver of a message is rarely available. Previous studies have proposed using DTN protocols to address these problems. In this paper we propose a collecting sensor data routing protocol, called Map-based Sensor-data Delivery Protocol (MSDP), which combines information about the road map and the nodes' future routes to improve data delivery. Through simulations based on accurate mobility and propagation models, and using the well-known Epidemic protocol as reference, we demonstrate that our proposal is able to significantly reduce channel usage, while maintaining or even improving the packet delivery ratio.