3D-VRTGR: A novel 3D virtual relay tunnel-based geographic routing for flying ad hoc networks
TL;DR
Simulation results under varying UAV density, mobility, and traffic load demonstrate that 3D-VRTGR consistently outperforms GPSR, UF-GPSR, and A-Geo and confirm the robustness and efficiency of 3D-VRTGR in highly dynamic FANET environments.
Abstract
Flying ad hoc networks (FANETs), composed of unmanned aerial vehicles (UAVs), are a key component of next-generation wireless networks that support various public safety and civilian applications in highly dynamic three-dimensional environments. However, high UAV mobility, highly dynamic topology, and limited energy and processing resources make efficient routing a challenging task. Traditional geographic routing protocols, such as GPSR, suffer from local minima, routing holes, unstable links, and high recovery overhead, reducing their effectiveness in FANETs. To address these challenges, this paper proposes a novel three-dimensional virtual relay tunnel-based geographic routing (3D-VRTGR) protocol. Unlike existing virtual tunnel-based routing approaches that use tunnels mainly as forwarding constraints, 3D-VRTGR employs a dynamically evolving virtual tunnel that continuously adapts to local network conditions and guides packet forwarding toward favorable regions. Therefore, the tunnel acts as a route-shaping mechanism rather than a simple filtering region. Specifically, the tunnel is reshaped using a distributed attraction-repulsion mechanism based on node density, residual energy, buffer status, geometric progress toward the destination, and void-history information. This mechanism guides the tunnel toward high-quality network regions while avoiding low-quality or potential void areas. Furthermore, a constant-time geometric pruning method provides efficient membership checking, and a bounded local self-repair mechanism addresses rare void-region scenarios without triggering perimeter recovery mode. Simulation results under varying UAV density, mobility, and traffic load demonstrate that 3D-VRTGR consistently outperforms GPSR, UF-GPSR, and A-Geo. The proposed protocol improves packet delivery ratio and throughput by up to 8.4% and 11.76%, respectively, while reducing latency, local minima occurrences, and link failures by up to 11.20%, 15.97%, and 14.54%, respectively. These results confirm the robustness and efficiency of 3D-VRTGR in highly dynamic FANET environments.