Neural network encoder-decoder firmware for adaptive wireless
This technology uses neural networks to automatically learn the best way to encode information before transmission and decode it after reception over a radio frequency channel. Instead of relying on hand-engineered modulation and coding schemes, the system trains encoder and decoder networks together so they adapt to the actual impairments of a specific channel — noise, interference, fading — without requiring engineers to model those impairments mathematically. The trained networks are then deployed at the transmitter and receiver ends to convert data into RF signals and reconstruct the original data on the other side. In effect, the communication system learns its own language optimized for the specific physical environment it operates in.
What you could build
A software-defined radio module or chipset firmware that replaces conventional modulation/coding blocks with trained neural network encoders and decoders, sold to wireless equipment OEMs or defense contractors building radios for unpredictable or contested RF environments.
Who in Virginia should care
Defense contractors and federal systems integrators in Northern Virginia and the Hampton Roads corridor — particularly those building tactical communications, electronic warfare, or spectrum management systems for DoD — 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
- Timothy James O`Shea
- Granted
- August 23, 2022
- Status
- Granted patent
- Patent number
- 11423301
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.