Transmitting Information with Global-designation of Emergency Routes for Edge Video Processing.

Large volumes of video feeds are generated by systems of Unmanned Aerial Vehicles (UAV) or city intersections with cameras in edge applications such as border security, crime mitigation, precision agriculture and smart city traffic management. Due to compute/network resource demands in video processing in a reliable and scalable manner to enable situational awareness for application consumers of video feeds, these systems need to rely on network services, edge computing devices and cloud infrastructure resources. Consequently, there is a need for effective integration of point-solutions that can transmit information from the system edge to the cloud platforms. In this paper, we present a novel framework viz., TIGER (Transmitting Information with Global-designation of Emergency Routes) that leverages the merits of Software-Defined Networking (SDN) and Programming Protocol-Independent Packet Processors (P4) to provide intelligent network services. Specifically, TIGER framework comprises two network services ‘priority routing’ and ‘congestion control’ that distinguish emergency traffic transmission based on source-based routing, and select congestion-free links based on multi-hop routing inspection as network cross-traffic increases, respectively. Our experiment results show that TIGER framework achieves priority routing by reducing packet jitter by 52%, corresponding to robust real-time data transmission. In addition, it achieves congestion control through traffic reduction by 45% across multiple pathways.