Artificial Intelligence Edge Applications in 5G Networks

TL;DR

This paper evaluates the feasibility of real-time computer vision applications on small devices using 5G networks, demonstrating improved latency, throughput, and high reliability compared to previous generations.

Contribution

It provides an empirical analysis of deploying real-time computer vision on small devices over 5G, highlighting performance improvements and reliability.

Findings

Latency and throughput are significantly improved with 5G.

High availability and reliability are achievable in real-time applications.

Feasibility of deploying computer vision on small devices using 5G is confirmed.

Abstract

In recent years, the 5th Generation of mobile communications has been thoroughly researched to improve the previous 4G capabilities. As opposed to earlier architectures, 5G Networks provide low latency access to services with high reliability. Additionally, they allow exploring new opportunities for applications that need to offload computing load in the network with a real-time response. This paper analyzes the feasibility of a real-time Computer Vision use case model in small devices using a fully deployed 5G Network. The results show an improvement in Latency and Throughput over previous generations, and a high percentage of Availability and Reliability in the analyzed use case.