CNN-based real-time RF signal detection and classification
This technology uses a convolutional neural network to automatically scan radio frequency spectrum data, identify specific types of signals, and pinpoint exactly where they appear in both frequency and time. Rather than requiring human analysts to manually inspect spectrograms, the system learns to recognize signal patterns and draws bounding boxes around detected signals on a visual display — much like object detection in images, but applied to radio waves. It can classify signals against a library of known signal types and apply different identification policies depending on context. The result is faster, automated RF signal awareness without needing a trained spectrum analyst watching a screen.
What you could build
An automated spectrum monitoring and signal intelligence platform that ingests RF sensor data and delivers real-time signal detection, classification, and visualization — sold to defense and intelligence contractors, spectrum regulators, or telecom operators managing interference.
Who in Virginia should care
Northern Virginia defense and intelligence contractors (e.g., SAIC, Leidos, Booz Allen, ManTech) and DoD/IC spectrum management programs would be natural buyers or partners.
Readiness: Prototype likely
Concept — described but not yet demonstrated. Lab validated — supported by experimental results in the patent. Prototype likely — the text describes a built, working embodiment.
Readiness is inferred from the patent text, not from a lab visit.
The record
- Inventors
- Timothy James O`Shea, Tamoghna Roy
- Granted
- March 17, 2026
- Status
- Granted patent
- Patent number
- 12581463
Ready to talk?
Virginia Tech Intellectual Properties handles licensing for this technology.
Prosim summaries are generated from public patent text and are not legal advice.