Structure-free message aggregation and routing in traffic information system (SMART)

Abstract

In a VANET (Vehicular Ad-Hoc Network) environment, it is highly probable that multiple nodes detect and report common events. Hence, elimination of redundant messages can help in efficient bandwidth usage. Data aggregation has also been identified as another potential tool to address broadcast storm in VANETs. Though many aggregation and in-network filtering schemes have been proposed, few studies have been done to combine the two approaches in a structured fashion in order to reap the full benefits.

In this work, we have shown how our proposed SMART protocol utilises the data aggregation tool in conjunction with message fragmentation, to address the prevailing challenge of vehicular networks—efficient bandwidth usage. The data is broken down into fragments and subsequently analysed, thus resulting in the elimination of redundant data than simply suppressing the transmission of duplicate data. The data received at every node is compared, analysed and stored into a well-organised data structure for faster and efficient look-up, thus transforming the network into a multi-level filtering system. Our SMART protocol is an efficient and scalable data aggregation scheme that achieves information dissemination without incurring any overhead of constructing a structure. As a proof of concept, we have shown that the SMART protocol improves efficiency in VANET communication by restricting the transmission of 75–80% redundant data.

Introduction

As a continually evolving branch of study in telecommunications, Vehicular Ad hoc Networks have attracted immense interest both in the application industry and in academic circles. Vehicular Ad Hoc Networks (VANETs) or Intelligent Vehicular Ad Hoc Networks (InVANET) are forms of MANETs used for communication amongst vehicles and between vehicles and roadside equipment.

VANETs however need to confront the unique challenge of unnecessary bandwidth consumption due to duplicate messages. Due to the nature of the VANET environment, it is highly probable that multiple nodes could detect a common phenomenon, thus transmitting redundant data. This may also lead to what is known as ‘Broadcast storm’. Hence, in-network filtering can help in efficient bandwidth usage. Data aggregation has also been identified as another potential tool to address broadcast storm in VANETs. Though many aggregation and in-network filtering schemes have been proposed, few studies have been done to combine the two approaches in a structured fashion in order to reap the full benefits.

In this paper, we propose SMART (Structure-free Message Aggregation and Routing scheme for Traffic information systems). The SMART protocol utilises the tool of data aggregation in conjunction with message fragmentation, to address the prevailing challenge of vehicular networks—the elimination of redundant data. SMART's context-adaptive, in-network filtering approach helps in pre-processing data prior to transmitting it to the next node, thus leading to efficient bandwidth usage. The data received from different source nodes are compared, analysed and stored into a well-organised data structure for faster and efficient look-up and comparison.