AI waveform engine that self-optimizes for degraded RF links
This technology uses neural networks to automatically learn the best way to encode and transmit data over a radio frequency channel, and to decode it on the other end. Instead of engineers manually designing modulation and coding schemes, two AI models — one acting as a transmitter and one as a receiver — train together by measuring how much information is lost or distorted in transmission and continuously adjusting to minimize that error. The system learns to adapt to the specific characteristics of a real-world channel, including interference and noise, rather than relying on fixed, pre-designed waveforms. Over time, the paired networks converge on a communication strategy optimized for that channel.
What you could build
An AI-driven modem or waveform engine that self-optimizes for specific RF environments, sold as a software module or chipset to defense communications contractors, satellite operators, or private wireless network builders who need reliable links in contested or degraded conditions.
Who in Virginia should care
Defense primes and government contractors concentrated in Northern Virginia (Booz Allen, SAIC, Leidos, Peraton) and military communications programs at DoD installations would be natural partners or licensees.
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
- Timothy James O`Shea
- Granted
- February 26, 2019
- Status
- Granted patent
- Patent number
- 10217047
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.