Duplicate suppression for efficient floating car data collection in heterogeneous LTE-DSRC vehicular networks

Abstract

Collecting data from a large number of agents scattered over a region of interest is becoming an increasingly appealing paradigm to feed big data archives that lay the ground for a vast array of applications. Vehicular Floating Car Data (FCD) collection is a major representative of this paradigm. Massive data collection from floating vehicles is the key to Intelligent Transportation Systems. We address the design and performance evaluation of a data collection protocol for the use case of periodic data collection. We target robustness, optimizing the amount of data and the value of the collection period, keeping in mind the goals of autonomous node operation and minimal coordination effort. From a system point of view, we believe that best solutions should jointly exploit the Long Term Evolution (LTE) cellular access network and the Dedicated Short-Range Communication (DSRC) based Vehicular Ad Hoc Network (VANET). Through a detailed comparative analysis, we show that such a hybrid approach offers superior performance, especially as for offloading the cellular radio access. A lightweight signaling procedure is designed, based on the DSRC VANET, which is able to avoid most of the duplicated data records, even if a distributed operation approach is pursued. The impact of the proposed protocol on the VANET load is evaluated and proved to be quite small, so that it does not interfere with other VANET-specific messages.