Image Uploading for Safe Driving Applications in Vehicular Networks Based on Mobile Edge Computing Technologies

Abstract

This paper proposes an image uploading system for vehicle safety monitoring applications based on mobile edge computing technologies. In this system, numerous camera nodes are placed within vehicles to collect vehicle journey data, and when emergency events such as accidents are detected, a camera with a view in the direction of the accident can upload more data than those in other directions. In addition, other vehicles within the area of the accident will also upload image data about the event, meaning that a greater amount of useful data on the accident will be collected for analysis and assistance to drivers. We propose to use mobile edge computing because of its many advantages, such as fast service response, low end-to-end delay and suitability for large data storage systems. We also propose solutions to minimize the uploading of redundant, repetitive and similar image data, a problem that has been encountered in previous systems. The results of both experiments and simulations showed that our system achieves better performance than those proposed in related studies.