Granted Patent B2

US 12,593,230 · App. 19/256,862 · Granted Mar 31, 2026

Inventors: David William Kleinbeck (Lees Summit, MO); Ronald C. Dzierwa (Baltimore, MD); Daniel Carbajal (Severna Park, MD)

Assignee: Digital Global Systems, Inc.

H04W24/08H04B17/23H04B17/27H04B17/318H04B17/347H04W4/029H04W16/14H04W24/10H04W52/0203H04W64/006H04W72/0453H04W72/0473H04B17/3911

Abstract

Claims (59)

1 . An apparatus for spectrum data management for a radio frequency (RF) environment, comprising:

at least one receiver and an automatic signal detection (ASD) module;

wherein the apparatus is operable to detect at least one signal in the RF environment using the ASD module;

wherein a sensitivity of the ASD module is operable to be fine-tuned based on dynamic feedback to determine if the at least one signal is a signal of interest;

wherein the apparatus is operable to smooth a real-time spectral sweep with a correction vector and a first smoothing filter;

wherein the apparatus is operable to perform a second smoothing filter only on frequencies outside a frequency range of the at least one signal;

wherein the apparatus is operable to apply a gradient detection algorithm to smoothed signal data based on a machine learning (ML) algorithm;

wherein the apparatus is operable to use gradients from the gradient detection algorithm and the first smoothing filter to create a calibration vector;

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

wherein the apparatus is operable to subtract a baseline from the real-time spectral sweep to reveal the at least one signal;

wherein the apparatus is configured for conflict recognition and anomaly identification wherein the apparatus is operable to process power level measurements of the RF environment using compressed data for deltas;

wherein the apparatus is operable to determine if the at least one signal is noise based on a pre-recognition delay;

wherein the apparatus is operable to calculate signal degradation data for the at least one signal based at least in part on 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 deltas are differentials from the baseline.

3 . The apparatus of claim 1 , wherein the ASD module is operable to identify a number of users associated with the at least one signal.

4 . The apparatus of claim 1 , wherein the at least one receiver comprises a primary receiver and a secondary receiver, wherein the primary receiver is configured to generate In-Phase and Quadrature (I/Q) data for at least one target bandwidth based on a learning and conflict detection engine, and wherein the secondary receiver is configured to perform a fast Fourier transform (FFT) based on a wideband sweep of the RF environment.

5 . The apparatus of claim 4 , further comprising an I/Q buffer, wherein the apparatus is operable to determine whether to keep the I/Q data in the I/Q buffer.

6 . The apparatus of claim 4 , further comprising a demodulator configured to distill the I/Q data and store actionable I/Q data, wherein the actionable I/Q data comprises signal metrics, protocol data, radio identification (ID), network ID and layer 3 data.

7 . The apparatus of claim 1 , wherein the apparatus is operable to add the at least one signal to a signal observation database based on the pre-recognition delay.

8 . The apparatus of claim 1 , wherein the apparatus is operable to tune the ASD module automatically.

9 . The apparatus of claim 1 , wherein the apparatus is configured to detect at least one anomaly based on the ML algorithm.

10 . The apparatus of claim 1 , wherein the ASD module is operable for signal recognition based on temporal feature extraction (TFE).

11 . The apparatus of claim 1 , wherein the ASD module is operable to detect a narrowband signal with a bandwidth from 1 kHz to 60 kHz inside a wideband signal with a bandwidth up to 100 MHz across a 6 GHz spectrum.

12 . The apparatus of claim 1 , wherein the apparatus further comprises a learning and conflict detection engine.

13 . An apparatus for spectrum data management for a radio frequency (RF) environment, comprising:

at least one receiver;

wherein the apparatus is operable to detect at least one signal in the RF environment;

wherein the apparatus is operable to smooth a real-time spectral sweep with a correction vector and a first smoothing filter;

wherein the apparatus is operable to perform a second smoothing filter only on frequencies outside a frequency range of the at least one signal;

wherein the apparatus is operable to apply a gradient detection algorithm to smoothed signal data based on a machine learning (ML) algorithm;

wherein the apparatus is operable to use gradients from the gradient detection algorithm and the first smoothing filter to create a calibration vector;

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

wherein the apparatus is operable to subtract a baseline from the real-time spectral sweep to reveal the at least one signal;

wherein the apparatus is operable to add the at least one signal to a signal observation database based on a pre-recognition delay;

wherein the apparatus is operable to process power level measurements of the RF environment using compressed data for deltas;

wherein the apparatus is operable to determine if the at least one signal is noise based on the pre-recognition delay;

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

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

wherein the apparatus is configured for conflict recognition and anomaly identification.

14 . The apparatus of claim 13 , wherein the apparatus is configured to detect at least one anomaly based on the ML algorithm.

15 . A method of spectrum data management for a radio-frequency (RF) environment, comprising:

a node device comprising at least one receiver creating learning data of the RF environment;

the node device detecting at least one signal in the RF environment based on a machine learning (ML) algorithm;

fine-tuning a sensitivity of a signal detection module of the node device based on dynamic feedback to determine if the at least one signal is a signal of interest;

the node device averaging a real-time spectral sweep, removing areas identified by matched positive and negative gradients, and connecting points between removed areas to determine a baseline;

the node device smoothing the real-time spectral sweep with a correction vector and a first smoothing filter;

the node device applying a second smoothing filter only on frequencies outside a frequency range of the at least one signal;

the node device using a calibration vector to de-bias raw signal data based on the ML algorithm;

the node device subtracting the baseline from the real-time spectral sweep to reveal the at least one signal;

the node device processing power level measurements of the RF environment using compressed data for deltas;

the node device determining if the at least one signal is noise based on a pre-recognition delay;

the node device calculating signal degradation data for the at least one signal based at least in part on hardware parameters;

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

the node device detecting at least one anomaly based on the ML algorithm.

16 . The method of claim 15 , further comprising the node device extracting time-frequency features of the RF environment.

17 . The method of claim 15 , wherein the signal detection module is operable to detect a narrowband signal with a bandwidth from 1 kHz to 60 kHz inside a wideband signal with a bandwidth up to 100 MHz across a 6 GHz spectrum.

18 . The method of claim 15 , further comprising the node device generating at least one report for the RF environment.

19 . The method of claim 15 , further comprising the node device generating In-Phase and Quadrature (I/Q) data for at least one target bandwidth determined by a learning and conflict detection engine; the node device distilling the I/Q data and storing actionable I/Q data; and the node device performing a fast Fourier transform (FFT) based on a wideband sweep of the RF environment and extracting metadata based on FFT data.

Assignments (3)

Continuity (30)

References Cited (400)