VeSCA: Vehicular stable cluster-based data aggregation

Abstract

In-network data aggregation is a promising technique to reduce data transmission that contributes the effective usage of bandwidth and co-existence of different applications in vehicular ad-hoc networks (VANET). Early aggregation schemes in VANET are grouped into two categories; structure free and structure based data aggregation. In structure-free data aggregation, vehicles apply pre-defined delay value before forwarding a data packet to the next hop. On the contrary, structure-based data aggregation uses a hierarchical structure, based on either road information or vehicles, to perform data aggregation. To provide efficient and scalable VANET communication, data aggregation is essential for reducing per vehicle bandwidth requirements. In this paper, we propose a multi-hop structure based data aggregation method namely VeSCA where mobile nodes are grouped based on relative mobility with minimum-overhead cluster construction and cluster members apply data aggregation before forwarding data packet to the parent node. Using various key metrics including data aggregation ratio, delay and aggregated data delivery ratio, we demonstrate superior performance VeSCA compared both previous cluster based data aggregation and alternative aggregation mechanism via extensive simulations in ns-3 with the vehicle mobility input from the Simulation of Urban Mobility (SUMO). VeSCA achieves over 70% aggregated data packet delivery ratio with aggregation ratio 40%.