Learned RF signal compressor for spectrum and signals
This technology uses a pair of neural networks — an encoder and a decoder — to compress radio frequency signals collected at a radio antenna, store them compactly, and then reconstruct them with high fidelity for analysis. The encoder learns the most efficient mathematical basis for representing any given RF signal, rather than relying on fixed compression rules. The compressed files can later be decompressed and fed into standard signal analysis tools to extract information or detect events. The system trains itself by minimizing both the reconstruction error and the size of the compressed output simultaneously, so it continuously improves the compression-quality tradeoff.
What you could build
A software module or edge-compute appliance that sits at radio head hardware — spectrum sensors, base stations, or signals intelligence collectors — compressing raw RF captures before storage or backhaul, then reconstructing them on demand for downstream processing. Primary buyers would be defense/signals intelligence contractors, spectrum monitoring vendors, and telecom network equipment suppliers.
Who in Virginia should care
Northern Virginia and the Hampton Roads corridor host dense concentrations of defense signals intelligence contractors (Leidos, SAIC, Peraton, Booz Allen) and telecom infrastructure operators who manage large RF sensor networks — natural licensing or partnership targets.
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
- Granted
- May 6, 2025
- Status
- Granted patent
- Patent number
- 12293297
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.