Evaluating SDN against AODV for dynamic topology management in disaster-relief UAV networks

In the aftermath of natural disasters, the rapid deployment of Unmanned Aerial Vehicles (UAVs) is critical for re-establishing communication services. However, managing the network topology in these scenarios is a significant challenge, primarily due to the high mobility of both the UAVs and the ground users they serve. While dynamic routing protocols have been the traditional approach for ad hoc networks, the centralized control paradigm of Software-Defined Networking (SDN) presents a compelling alternative for dynamic topology management. This paper investigates the efficacy of an SDN-based management solution for UAV networks, a topic for which the existing literature lacks a definitive performance consensus. An SDN architecture tailored for UAV-based emergency networks is proposed and evaluated, with performance benchmarked against the well-established Ad hoc On-Demand Distance Vector (AODV) protocol. Through simulations in the OMNeT++ framework, key performance indicators were analyzed, including end-to-end latency, packet delivery ratio, infrastructure utilization, and packet drop rates, under varying user loads. The results demonstrate a clear performance advantage for the proposed SDN solution. Specifically, the architecture achieved a packet delivery ratio of up to 17.5 % higher than AODV and reduced the end-to-end data packet latency by up to 10 ms. These findings provide strong evidence that a centralized software-defined approach can offer superior reliability and efficiency for managing highly dynamic UAV communication networks in critical disaster recovery operations.