Classification and detection of multiple UAVs using rational Gaussian wavelet neural networks
arXiv:2605.26310v1 Announce Type: new Abstract: The detection of unmanned aerial vehicles (UAVs) is important for the protection of civilian and military infrastructure. In this paper we propose a cost effective UAV detection system using sound signals obtained from microphones. The recorded signals are passed through a signal processing pipeline which employs interpretable adaptive feature extractors using so-called rational Gaussian wavelets. These adaptive wavelet transformations are embedded into and trained together with an underlying small neural network which detects and classifies UAVs
Advances in AI, particularly neural networks and wavelet analysis, coupled with the increasing proliferation of UAVs, create an urgent need and technical capability for effective detection systems.
This development offers a cost-effective and interpretable method for UAV detection, directly addressing security vulnerabilities in critical infrastructure and defence settings.
The proposed system provides an alternative to traditional radar-based detection, leveraging sound signals and AI for potentially wider and more flexible deployment.
- · Defence contractors
- · Security firms
- · AI/ML researchers
- · Government agencies
- · UAV operators (malicious)
- · Traditional security system providers (if they don't adapt)
Improved detection capabilities for unauthorized drones over protected areas.
Increased adoption of acoustic-based AI detection systems in both civilian and military security protocols.
Potential for new regulations and standards around acoustic 'fingerprinting' of drones for identification and tracking.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG