Granted Patent B2

US 12,267,714 · App. 18/888,822 · Granted Apr 1, 2025

Inventors: Ronald C. Dzierwa (Baltimore, MD); Daniel Carbajal (Severna Park, MD)

Assignee: Digital Global Systems, Inc.

H04W24/08H04B17/23H04B17/26H04B17/27H04B17/309H04B17/318H04W4/029H04W16/14H04W24/10H04W64/006H04B17/3911

Abstract

Claims (60)

1. An apparatus for automatic signal detection in an electromagnetic environment, comprising:

at least one receiver, at least one processor, and at least one memory;

wherein the apparatus is operable to create power level measurements of one or more frequency bins within the electromagnetic environment;

wherein the apparatus is operable to learn a baseline;

wherein the apparatus is operable to automatically detect at least one signal based on the baseline;

wherein the apparatus is operable to create a knowledge map of the electromagnetic environment based on a machine learning (ML) algorithm;

wherein the knowledge map is based on the power level measurements of the one or more frequency bins within the electromagnetic environment;

wherein the apparatus is operable to create deltas that are differentials from the baseline and minimize a data set or sample data required for comparisons and/or analytics;

wherein the apparatus is operable to reconstruct the at least one signal using the deltas and the baseline;

wherein the apparatus is operable to fill gaps during reconstruction of the at least one signal where signal data is absent;

wherein the apparatus is operable to smooth the signal data using a first smoothing filter;

wherein the apparatus is operable to use gradients and a second smoothing filter to the smoothed signal data to create a calibration vector;

wherein the apparatus is operable to use the calibration vector to de-bias raw signal data;

wherein the apparatus includes a Temporal Feature Extraction (TFE) system operable to automatically tune parameters and dynamically adjust sensitivity of the one or more frequency bins within the electromagnetic environment;

wherein the apparatus is operable to calculate signal degradation data for the at least one signal based at least in part on noise figure parameters and hardware parameters; and

wherein the hardware parameters comprise antenna position, antenna type, orientation, and/or effective isotropic radiated power (EIRP).

2. The apparatus of claim 1 , wherein the TFE system is operable to utilize ML.

3. The apparatus of claim 1 , wherein the apparatus is operable to determine if the at least one signal is moving using a frequency-locked loop by determining if there is a Doppler change in the at least one signal.

4. The apparatus of claim 1 , wherein the apparatus is operable to estimate a location of a signal emitting device from which the at least one signal is emitted based on in-phase and quadrature (I/Q) data generated from a spectral sweep.

5. The apparatus of claim 1 , wherein the deltas provide for interference identification, neighboring band identification, device identification, and/or signal optimization in near real time.

6. The apparatus of claim 1 , wherein the apparatus is operable to provide automatic modulation detection.

7. The apparatus of claim 1 , wherein the ML algorithm provides for automatic signal variance determination.

8. The apparatus of claim 1 , further comprising a temporal anomaly detector operable to learn channels in the electromagnetic environment.

9. The apparatus of claim 1 , wherein the apparatus is operable to implement a learning routine, wherein the learning routine includes the power level measurements as an index into each distribution column corresponding to each frequency bin and incrementing a counter in a location corresponding to a power level.

10. The apparatus of claim 1 , wherein the apparatus is operable to distinguish signals based on gradients from a moving noise floor without a fixed threshold bar.

11. The apparatus of claim 1 , wherein the ML algorithm stores a learning map that is operable to update at a predetermined timeframe.

12. The apparatus of claim 1 , wherein the second smoothing filter is performed only on frequencies outside a frequency range of the at least one signal.

13. A method for automatic signal detection in an electromagnetic environment, comprising:

receiving power level measurements of one or more frequency bins within the electromagnetic environment using at least one receiver;

learning a baseline of the electromagnetic environment based on the power level measurements;

automatically detecting at least one signal based on the baseline;

creating a knowledge map of the electromagnetic environment using a machine learning (ML) algorithm, wherein the knowledge map is based on the power level measurements of the one or more frequency bins;

generating deltas that represent differentials from the baseline to minimize data sets or sample data required for comparisons and analytics;

reconstructing the at least one signal using the deltas and the baseline;

filling gaps in the reconstruction of the at least one signal where signal data is absent;

smoothing the signal data using a first smoothing filter;

applying gradients and a second smoothing filter to the smoothed signal data to generate a calibration vector to de-bias raw signal data;

utilizing a Temporal Feature Extraction (TFE) system with ML to automatically tune parameters and dynamically adjust sensitivity of the one or more frequency bins within the electromagnetic environment; and

calculating signal degradation data for the at least one signal based at least in part on noise figure parameters and hardware parameters;

wherein the hardware parameters comprise antenna position, antenna type, orientation, and/or effective isotropic radiated power (EIRP).

14. The method of claim 13 , further comprising implementing a learning routine, wherein the learning routine includes taking the power level measurements as an index into each distribution column corresponding to each of the one or more frequency bins.

15. The method of claim 13 , wherein the second smoothing filter is performed only on frequencies outside a frequency range of the at least one signal.

16. The method of claim 13 , further comprising learning channels in the electromagnetic environment.

17. A method for automatic signal detection in an electromagnetic environment, comprising:

receiving power level measurements of one or more frequency bins within the electromagnetic environment using at least one receiver;

learning a baseline of the electromagnetic environment based on the power level measurements;

automatically detecting at least one signal based on the baseline;

creating a knowledge map of the electromagnetic environment using a machine learning (ML) algorithm, wherein the knowledge map is based on the power level measurements of the one or more frequency bins;

generating deltas that represent differentials from the baseline to minimize data sets or sample data required for comparisons and analytics;

reconstructing the at least one signal using the deltas and the baseline;

filling gaps in the reconstruction of the at least one signal where signal data is absent;

smoothing the signal data using a first smoothing filter;

applying gradients and a second smoothing filter to the smoothed signal data to generate a calibration vector to de-bias raw signal data;

utilizing a Temporal Feature Extraction (TFE) system to automatically tune parameters and dynamically adjust sensitivity of the one or more frequency bins within the electromagnetic environment; and

calculating signal degradation data for the at least one signal based at least in part on noise figure parameters and hardware parameters;

wherein the hardware parameters comprise antenna position, antenna type, orientation, and/or effective isotropic radiated power (EIRP); and

wherein the detecting the at least one signal in the environment comprises fine-tuning a threshold of power level on a segmented basis.

18. The method of claim 17 , further comprising implementing a learning routine, wherein the learning routine includes taking the power level measurements as an index into each distribution column corresponding to each of the one or more frequency bins.

19. The method of claim 17 , wherein the threshold of power level is operable to be variable between the one or more frequency bins.

20. The method of claim 17 , further comprising learning channels in the electromagnetic environment.

Assignments (2)

Continuity (19)

References Cited (400)