AI-Powered Anti-Drone System for Protecting Critical Infrastructure Sites: Case Study – Port Sudan International Airport

This paper presents an AI-powered Counter-Unmanned Aerial System (C-UAS) architecture customized for Port Sudan International Airport to enhance protection of airspace and ground assets from unauthorized drones. The system integrates perimeter radar sensors, a sensor-fusion pipeline, and Convolutional Neural Network (CNN)-based classification, coupled with directional RF jamming for real-time threat mitigation. Site-specific factors including coastal geography, runway configuration (16/34; ≈ 2,500 × 45 m), and environmental stressors (heat, humidity, and episodic dust events) are incorporated into the simulation framework. We simulate various scenarios of unauthorized drone incursions into restricted airspace and evaluate the system’s detection accuracy, classification reliability, and mitigation effectiveness. The results demonstrate that the proposed multi-sensor AI-driven system can: Reliably detect small drones at ranges between 1–2 km, Achieve classification accuracies exceeding 95 %, and Successfully neutralize incursions through targeted radio-frequency jamming within a 5-second command-loop latency Keywords: Counter-UAV; Airport security; Radar detection; Sensor fusion; CNN; Jamming; Critical infrastructure protection; Port Sudan; Environmental modeling.