Computer Science

A Distributed System for Reducing Uploaded Data Redundancy in Vehicular Networks

Document Type

Conference Paper

Abstract

Uploading vehicle sensor data to support autonomous driving is necessary to understand the current situation to make the best decision. In this paper, we develop a system that relies on a peer-to-peer mechanism to obtain information in a vehicular network. Consider a scenario in which each vehicle equipped with a camera and communication capability is responsible to upload new snapshots to a datacenter, and the datacenter combines the snapshots to create a map. In a naïve sensor data upload scheme, each vehicle uploads its snapshot periodically and the datacenter will find out new information and integrate it to a map. This method might work if only a small number of vehicles are uploading at the same time. However, this naïve method is not scalable when dozens of vehicles need to upload, as the communication bandwidth will be a bottleneck. To address the challenge, we propose a novel distributed system to reduce data redundancy. As a result, the bandwidth consumption between vehicles and the datacenter is reduced as well. The key idea is to use location information (e.g., GPS coordinates) to simplify the design and coordination among peers (vehicles), and rely on computer vision algorithms to remove redundant data and identify important information to be uploaded. In this paper, we outline the design of our system and verify the efficacy of our system through a simulation study.

Publication Title

2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019

Publication Date

2019

First Page

998

Last Page

1003

ISBN

9781538691519

DOI

10.1109/PERCOMW.2019.8730671

Keywords

bandwidth consumption, data redundancy, distributed system, upload bandwidth

APA Citation

Wang, Z., Zhao, Y., Tseng, L., Higuchi, T., & Altintas, O. (2019, March). A distributed system for reducing uploaded data redundancy in vehicular networks. In 2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) (pp. 998-1003). IEEE.

Link to Full Text

Share

COinS