Granted Patent B2

US 12,267,688 · App. 18/778,403 · Granted Apr 1, 2025

Inventor: Armando Montalvo (Winter Garden, FL)

Assignee: Digital Global Systems, Inc.

H04W16/10G06F30/27G06N3/02G06N5/022G06N5/04G06N20/00G06N20/10G06N20/20H04L41/0893H04L41/16H04W24/02H04W72/0453G06N3/042G06N3/045H04L41/0894H04W16/14H04W24/08

Abstract

Claims (38)

1. A system for spectrum analysis in an electromagnetic environment comprising:

a programmable channelizer configured to analyze data from the electromagnetic environment;

a classification engine; and

a learning engine;

wherein the classification engine is operable to classify at least one signal of interest based on the analyzed data from the programmable channelizer;

wherein the learning engine is operable to provide learning data to the classification engine for classifying the at least one signal; and

wherein the learning engine is operable to utilize at least one learning technique to generate the learning data.

2. The system of claim 1 , wherein the at least one learning technique includes machine learning (ML), artificial intelligence (AI), deep learning (DL), neural networks (NNs), artificial neural networks (ANNs), support vector machines (SVMs), Markov decision process (MDP), and/or natural language processing (NLP).

3. The system of claim 1 , wherein the programmable channelizer includes channelization selector logic for a lookup of filter coefficient vectors and/or channelization vectors.

4. The system of claim 3 , wherein data from the lookup of filter coefficient and/or channelization vectors undergoes preprocessing with a mix circular rotator to produce a plurality of blocks of a plurality of points.

5. The system of claim 1 , wherein the analyzed data from the programmable channelizer undergoes an N point fast Fourier transform (FFT), wherein a power spectral density (PSD) is calculated for the N point FFT, wherein a complex average FFT is obtained for a plurality of blocks of the N point FFT.

6. The system of claim 1 , wherein the programmable channelizer includes at least one channel definition, at least one fast Fourier transform (FFT) configuration, at least one detector configuration, and/or at least one channel detection.

7. The system of claim 1 , wherein the programmable channelizer includes at least one deference matrix operable to identify at least one narrowband channel that is a subset of at least one wideband channel.

8. The system of claim 1 , wherein the programmable channelizer includes buffer services, pre-processing of fast Fourier transform (FFT) bin samples, bin selection, at least one band pass filter (BPF), an inverse fast Fourier transform (IFFT) function to produce at least one IFFT, decomposition, and/or phase correction.

9. The system of claim 1 , wherein the programmable channelizer includes at least one channelization vector, wherein the at least one channelization vector is operable to specify normalized power levels per fast Fourier transform (FFT) bin for at least one channel.

10. The system of claim 1 , wherein the system is combined in a chip, on a chipset, or on a circuit board.

11. A system for spectrum analysis in an electromagnetic environment comprising:

a channelizer; and

a learning engine;

wherein the channelizer and the learning engine are on at least one chip, a chipset, or at least one circuit board;

wherein the channelizer is configured to analyze data from the electromagnetic environment;

wherein the channelizer includes at least one fast Fourier transform (FFT) configuration, wherein the at least one FFT configuration is operable to detect at least two channels; and

wherein the learning engine is operable to learn the electromagnetic environment based on the analyzed data from the channelizer.

12. The system of claim 11 , wherein the channelizer is a programmable channelizer.

13. The system of claim 11 , wherein the at least one FFT configuration is further operable to identify a sufficient resolution bandwidth to resolve ambiguities between the at least two channels.

14. The system of claim 11 , wherein the channelizer includes at least one deference matrix operable to identify at least one narrowband channel that is a subset of at least one wideband channel.

15. The system of claim 11 , wherein the channelizer includes at least one channelization vector, wherein the at least one channelization vector is operable to specify normalized power levels per FFT bin for at least one channel.

16. A system for spectrum analysis in an electromagnetic environment comprising:

a channelizer; and

a learning engine;

wherein the channelizer and the learning engine are on at least one chip, a chipset, or at least one circuit board;

wherein the channelizer is configured to analyze data from the electromagnetic environment;

wherein the channelizer includes at least one deference matrix operable to identify at least one narrowband channel that is a subset of at least one wideband channel; and

wherein the learning engine is operable to provide data to the channelizer created based on at least one learning technique to generate data for identification of the at least one narrowband channel.

17. The system of claim 16 , wherein the channelizer is a programmable channelizer.

18. The system of claim 16 , wherein the channelizer includes at least one fast Fourier transform (FFT) configuration, wherein the at least one FFT configuration is operable to detect at least two channels.

19. The system of claim 18 , wherein the at least one FFT configuration is further operable to identify a sufficient resolution bandwidth to resolve ambiguities between the at least two channels.

20. The system of claim 16 , wherein the at least one learning technique includes machine learning (ML), artificial intelligence (AI), deep learning (DL), neural networks (NNs), artificial neural networks (ANNs), support vector machines (SVMs), Markov decision process (MDP), and/or natural language processing (NLP).

Assignments (1)

Continuity (10)

References Cited (400)