In recent weeks, there have been increasing sightings of unknown drones in several European countries. Incidents have occurred particularly in northern and eastern Europe, resulting in temporary airspace closures and flight diversions. The exact origin of many of these aircraft remains unclear, with some suspecting a connection to reconnaissance or disruptive activities.
The increasing number of such incidents shows how important reliable and early detection of unmanned aerial objects has become. At JOANNEUM RESEARCH, the Intelligent Acoustic Solutions research group has many years of expertise in acoustic drone detection, which has several advantages over other detection technologies.
3D microphone arrays in hemispherical shape
‘Unlike active systems such as radar, the acoustic method works passively – it does not emit any signals itself and therefore remains undetected. This makes it particularly valuable in security-critical scenarios,’ explains project manager Martin Blass from the DIGITAL Institute. A current development focus is on hemispherical 3D microphone arrays that enable precise localisation of drones at altitude. The system is continuously improved through the use of AI: models are trained with acoustic signatures of different drone types and can thus reliably recognise typical noise patterns – even in environments with high background noise. Research group leader and inventor of the so-called ‘tunnel ears’ (acoustic accident detection in tunnels – AKUT®) Franz Graf explains: ‘Microphone arrays are used to pick up sound sources from different directions, and the position of a drone can be determined from the time differences between the signals.’
Neues Forschungsprojekt UAS-CheckPoint zur akustischen Drohnendetektion
Following initial long-term successes, researchers are continuing to pursue the development of acoustic drone detection: the UAS-CheckPoint project will launch in December 2025. The aim is to set up distributed, multimodal sensor nodes that are installed as checkpoints on buildings, masts or vehicles and are powered by energy-autonomous batteries. Each node combines audio, video, GNSS and remote ID modules with a computing unit (edge computing), enabling multi-sensory verification of unmanned aerial vehicles in what is known as U-space.
Sensor node ‘checkpoints’ have several advantages:
- Long-range surveillance: Distributed sensor technology allows drone movements to be detected over large areas and tracked over long distances.
- Robustness: Decentralised, energy-self-sufficient systems operate independently of central infrastructure and remain functional even if power and communication networks fail.
- Tactical security: Acoustic sensors operate passively, are small and easy to conceal, and are therefore undetectable to potential attackers.
Partners of the project UAS-CheckPoint are AIT Austrian Institute of Technology, AIRlabs Austria, Frequentis AG und Austro Control, as well as drone manufacturers twins GmbH and Flightec GmbH. This project is funded by Österreichische Forschungsförderungsgesellschaft FFG as part of the Take Off 2024 programme.
Our expertise in acoustic signal processing and AI-based evaluation forms the foundation of these developments.
‘We combine physical understanding with data-driven intelligence,’ emphasises Blass. ‘This enables us to create systems that not only hear, but also understand what they hear.’
These capabilities will become increasingly important for airspace surveillance, as incidents in European airspace have far-reaching consequences. For both civil and military actors, it is not only a question of airspace safety, but also of protecting critical infrastructure and avoiding false alarms. Precise detection and classification of drone movements is therefore becoming a key prerequisite for the stability and operational capability of European states.
