Network slicing and edge-driven UAV surveillance in 5G private networks for industrial digital transformation in the aerospace sector

This paper presents a 5G-enabled private network solution designed to support industrial digital transformation within the aerospace sector. Building upon prior research that identified 5G adoption challenges in the Ayrshire region of Scotland, this work introduces a practical use case demonstrating an integrated 5G and edge computing system for autonomous aerial surveillance. The proposed system employs a 5G private network with network slicing to guarantee ultra-reliable low-latency communication (URLLC) even under network congestion. A fleet of 5G-enabled Unmanned Aerial Vehicles (UAVs) form a resilient mesh network using the Better Approach To Mobile Adhoc Networking (BATMAN) protocol to maintain continuous connectivity. At the network edge, a Human Detector AI module performs real-time intruder detection using a custom deep learning model, UWS-YOLO. Experimental evaluations demonstrate that network slicing ensures service continuity for mission-critical applications, highlighting the potential of 5G MPNs as a key enabler of digital transformation for the aerospace sector and the wider industrial landscape.