Machine-learned RF signal compression for wireless systems
This technology uses machine learning to compress radio frequency (RF) signals more efficiently than traditional methods. Two neural networks work together: one compresses the original RF signal into a compact form, and another reconstructs it as accurately as possible. The system trains itself by measuring both how much it compressed the signal and how closely the reconstruction matches the original, then adjusts both networks to optimize that balance. The result is a self-improving compression engine specifically tuned for RF signals rather than generic data.
What you could build
A software library or embedded module for wireless infrastructure equipment—such as base stations, software-defined radios, or spectrum monitoring systems—that reduces the data load of RF signal processing pipelines. Telecom equipment vendors, defense electronics integrators, and spectrum intelligence companies would be the primary buyers.
Who in Virginia should care
Northern Virginia's dense concentration of defense contractors, signals intelligence firms, and telecom infrastructure companies—many supporting DoD and intelligence community spectrum operations—would have direct interest.
Readiness: Lab validated
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
- —
- Granted
- February 25, 2020
- Status
- Granted patent
- Patent number
- 102800
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.