AI-driven RF signal compression for wireless and SIGINT
This technology uses a pair of neural networks — one to compress radio frequency signals and one to reconstruct them — trained together so the system learns the smallest possible representation of an RF signal without losing the information that matters. During training, the system automatically balances how much it compresses against how accurately it can rebuild the original signal. The result is an AI model that can be deployed in real wireless systems to shrink the data burden of transmitting or storing RF signals. This is essentially learned compression, tuned specifically for the statistical structure of radio signals rather than relying on traditional handcrafted encoding schemes.
What you could build
A software library or edge-deployable AI module that compresses RF signal streams for use in spectrum monitoring, cellular network diagnostics, or signals intelligence pipelines; likely buyers are wireless infrastructure vendors, spectrum analytics companies, and defense signal processing integrators.
Who in Virginia should care
Defense and intelligence contractors in Northern Virginia — particularly those supporting spectrum monitoring, SIGINT, or electronic warfare programs — would have direct operational interest.
Readiness: Concept
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
- Filed
- Patent pending — filed April 29, 2025
- Status
- Application
- Publication number
- US20260154551A1
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.