Urban Air Mobility (UAM) and Unmanned Traffic Management (UTM) have advanced in recent years, and it becomes clear that within these concepts, reliable communication systems are necessary for safe, scalable, and autonomous flight. Direct drone-to-drone (D2D) communication in particular is essential for safety-critical applications such as collision avoidance. However, urban environments impose severe challenges for the wireless channel, including dense multipath, non-line-of-sight propagation, high node mobility, spectrum scarcity and stringent latency and reliability requirements.
Since 2024, the German Aerospace Center (DLR), together with industry partners, has been developing a first demonstrator of DroneCAST (Drone Communications and Surveillance System), a D2D system whose design is tailored to these conditions. DroneCAST's core is a robust waveform design based on Orthogonal Frequency-Division Multiplexing (OFDM), Dual-Carrier Modulation (DCM), midambles, and LDPC codes to account for the rich multipath environment and rapidly changing channel conditions. It operates in the C-band (5,030–5,091 MHz) with up to 5 MHz bandwidth, foreseen for future unmanned aircraft systems (UAS) communication.
In this work, we present the first real-hardware implementation and flight demonstration of DroneCAST, moving from prior measurement-based simulation to a working over-the-air system. The physical layer is implemented in GNU Radio, running on Ettus USRP B210 software-defined radios (SDRs) and Intel NUC computers integrated onto our hexacopters. Following laboratory validation, we conducted an initial flight trial with two drones and one ground station in an urban collision avoidance scenario. Both drones broadcast their position and ID via DroneCAST and received each other’s signals, while the ground station tracked both. We successfully tracked the drones in the air and on the ground with sub-10 ms latency and consistent packet reception, with only occasional packet loss. We present our approach, discuss the challenges encountered and the solutions developed, and outline the next steps toward developing a full DroneCAST demonstrator.