Reverse CDN in Fog Computing: The lifecycle of video data in connected and autonomous vehicles

Abstract

This paper introduces the concept of the Reverse Content Distribution Network (rCDN) infrastructure synergized with Fog Computing. In many IoT use cases, video content and other streaming data flow from downstream devices (cameras) to upstream devices (e.g., IoT Gateway, Road Side Unit, network edge node) that can stream the content directly to the Cloud or can aggregate, store and process data in a distributed fashion. This builds a Content Distribution Network (CDN) that changes the traditional CDN model. This gives rise to rCDN, a collection of related video streams that flow from multiple content sources in a many-to-1 fashion, and where highly dynamic video data is likely aggregated, processed, analyzed, transformed, cached and migrated potentially multiple times en route to its final storage destination. In this paper, we present a solution that synergize rCDN and Fog Computing to serve connected and autonomous vehicles, where video content from vehicles cameras and street cameras are distributed across rCDN nodes. We also discuss the role of rCDN for data re-usability within the Smart City, and highlight existing challenges.