Low-cost handheld detection of non-cooperative UAS with edge machine learning
WhitepaperOctober 14, 2025
The proliferation of commercially available unmanned aerial systems (UAS) presents unprecedented security challenges.
Mitigating these threats requires expensive radar or camera systems, detailed knowledge of the UAS C2 protocol or cooperative telemetry. Bulky, fixed installations can be highly effective, but struggle to deliver results for mobile personnel, especially when seeking to remain inconspicuous.
TTP has developed a proof-of-concept pocket-sized passive RF drone detector that addresses this need. Our system combines low-cost RF hardware with low-power Edge AI for detection. With the model trained to detect the characteristics of UAS RF emissions, it has the potential to detect non-cooperative and previously unseen drones. Unlike other AI-powered solutions, our lightweight models can run on the processing power found in a mobile phone, and can be imagined as a massproduced solution for routine security or defence scenarios.
Leveraging an extensive pre-training step and a synthetic data generation pipeline significantly reduces the need for large-scale data collection and labelling activities, overcoming the common scalability issues of machine learning (ML) powered systems. When issued to personnel or used to protect critical infrastructure, the device provides increased situational awareness and robust detection performance in complex, noisy environments.
