Drone Swarm Surveillance using Hybrid Routing: Simulation and Performance Analysis of FANET

This paper presents the design, simulation, and performance analysis of a Flying Adhoc Network (FANET) for drone swarm surveillance. A 7-node topology comprising six drone nodes and one ground base station is implemented and evaluated in the NS-3 network simulator using a hybrid routing strategy that combines Ad-hoc On-Demand Distance Vector (AODV) routing with static pre-configured paths. A proximity-based dynamic gateway election mechanism enables the closest drone to the base station to act as the communication relay, with re-election occurring every 0.5 seconds. Network resilience is validated through a drone failure injection at t = 30 s (Drone-4 kill), and self-healing recovery is measured. Performance metrics including Packet Delivery Ratio (PDR), throughput, end-to-end latency, jitter, hop count, and routing efficiency are evaluated across drone speeds from 2 to 12 m/step. Key findings show that the 50–80 m communication range achieves greater than 98% PDR (the “Goldilocks Zone”), the network recovers from node failure within approximately 15 s via AODV reconvergence, and jitter remains below the 20 ms real-time video threshold across all tested speeds. A basic ML layer using Isolation Forest anomaly detection and Random Forest classification achieves over 95% accuracy in rogue drone identification.