Granted Patent B2

US 12,639,606 · App. 18/322,777 · Granted May 26, 2026

Inventors: Chad Lau (Melbourne, FL); Mark D. Rahmes (Melbourne, FL); David B. Chester (Palm Bay, FL); Michael C. Garrett (Melbourne, FL)

Assignee: EAGLE TECHNOLOGY, LLC

G06N10/20G06N3/0455G06F15/82G06F17/14G06F17/18G06N3/044G06N3/08G06N7/01G06N10/60

Abstract

Claims (42)

1 . A radio frequency (RF) signal classification device comprising:

an RF receiver configured to receive RF signals;

a quantum computing circuit configured to perform quantum subset summing; and

a processor configured to

generate a game theory reward matrix for a plurality of different deep learning models,

cooperate with the quantum computing circuit to perform quantum subset summing of the game theory reward matrix,

select a deep learning model from the plurality thereof based upon the quantum subset summing of the game theory reward matrix, and

process the RF signals using the selected deep learning model for RF signal classification.

2 . The RF signal classification device of claim 1 wherein the game theory reward matrix includes rows corresponding to different signal classes.

3 . The RF signal classification device of claim 1 wherein the game theory reward matrix includes columns corresponding to classification probabilities associated with the different deep learning models.

4 . The RF signal classification device of claim 3 wherein the classification probabilities comprise VAE cluster Z-test scores.

5 . The RF signal classification device of claim 3 wherein the classification probabilities comprise ResNet classifications.

6 . The RF signal classification device of claim 3 wherein the classification probabilities comprise probabilities from a plurality of distilled classification networks.

7 . The RF signal classification device of claim 6 wherein the distilled classification networks have different distillation temperatures associated therewith.

8 . The RF signal classification device of claim 3 wherein the classification probabilities comprise at least one of modulation class probabilities and waveform class probabilities.

9 . A radio frequency (RF) signal classification device comprising:

an RF receiver configured to receive RF signals;

a quantum computing circuit configured to perform quantum subset summing; and

a processor configured to

generate a game theory reward matrix for a plurality of different deep learning models, the game theory reward matrix including rows corresponding to different signal classes and columns corresponding to classification probabilities associated with the different deep learning models,

cooperate with the quantum computing circuit to perform quantum subset summing of the game theory reward matrix,

select a deep learning model from the plurality thereof based upon the quantum subset summing of the game theory reward matrix, and

process the RF signals using the selected deep learning model for RF signal classification.

10 . The RF signal classification device of claim 9 wherein the classification probabilities comprise VAE cluster Z-test scores.

11 . The RF signal classification device of claim 9 wherein the classification probabilities comprise ResNet classifications.

12 . The RF signal classification device of claim 9 wherein the classification probabilities comprise probabilities from a plurality of distilled classification networks.

13 . The RF signal classification device of claim 12 wherein the distilled classification networks have different distillation temperatures associated therewith.

14 . The RF signal classification device of claim 9 wherein the classification probabilities comprise at least one of modulation class probabilities and waveform class probabilities.

15 . A radio frequency (RF) signal classification method comprising:

receiving RF signals at an RF receiver;

using a processor to

generate a game theory reward matrix for a plurality of different deep learning models,

cooperate with a quantum computing circuit to perform quantum subset summing of the game theory reward matrix,

select a deep learning model from the plurality thereof based upon the quantum subset summing of the game theory reward matrix, and

process the RF signals using the selected deep learning model for RF signal classification.

16 . The method of claim 15 wherein the game theory reward matrix includes rows corresponding to different signal classes.

17 . The method of claim 15 wherein the game theory reward matrix includes columns corresponding to classification probabilities associated with the different deep learning models.

18 . The method of claim 17 wherein the classification probabilities comprise VAE cluster Z-test scores.

19 . The method of claim 17 wherein the classification probabilities comprise ResNet classifications.

20 . The method of claim 17 wherein the classification probabilities comprise probabilities from a plurality of distilled classification networks.

21 . The method of claim 20 wherein the distilled classification networks have different distillation temperatures associated therewith.

22 . The method of claim 17 wherein the classification probabilities comprise at least one of modulation class probabilities and waveform class probabilities.

Assignments (1)

Continuity (2)

References Cited (165)