IP Library › Granted Patent US 12675550
Granted Patent B2
US 12675550 · App. 17/712,630 · Granted Jul 7, 2026

Pattern recognition in signals

Inventors: Brandt J. Lomen (Beford, NH); Paul Monahan (Acton, MA); Matthew Anderson (Mont Vernon, NH)
Assignee: BAE Systems Information and Electronic Systems Integration Inc.
G06F17/18
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12675550
App. No.
17/712,630
Filed
Apr 4, 2022
Granted
Jul 7, 2026
Kind
B2
Art Unit
2151
USPC
708/212
Abstract

A method of recognizing a pattern in a signal is provided. The method includes receiving, by a processor and at an input, the signal comprising data; converting, by the processor and based on a plurality of features, the data from the signal into a collection of features; identifying, by the processor, one or more feature-to-feature transitions between features in the collection of features; plotting, by the processor, a directed graph of each of the feature-to-feature transitions, wherein each unique feature in the collection of features is represented by a single vertex and each feature-to-feature transition is represented by a directed edge between two vertices weighted by a corresponding number of each of the feature-to-feature transitions; and detecting, by the processor, a pattern in the data by comparing the weight of each of the feature-to-feature transitions in the directed graph to a user-defined threshold.

Claims (36)

1 . A method of recognizing a pattern in a signal, the method comprising:

receiving, by a processor and at an input, the signal comprising data;

converting, by the processor and based on a plurality of features, the data from the signal into a collection of features;

identifying, by the processor, one or more feature-to-feature transitions between features in the collection of features;

plotting, by the processor, a directed graph of each of the feature-to-feature transitions, wherein each unique feature in the collection of features is represented by a single vertex and each feature-to-feature transition is represented by a directed edge between two vertices weighted by a corresponding number of each of the feature-to-feature transitions; and

detecting, by the processor, a pattern in the data by comparing a weight of each of the feature-to-feature transitions in the directed graph to a user-defined threshold;

wherein the converting includes comparing the data to a first sequence of values, a second sequence of values, and a delta time between occurrence of the first sequence of values and the second sequence of values in the data, and wherein the collection of features includes two or more of the features in sequence based on the comparison.

2 . The method of claim 1 , further comprising defining, by the processor, each of the plurality of features as a unique sequence of values.

3 . The method of claim 1 , wherein each of the plurality of features includes a unique sequence of values.

4 . The method of claim 1 , wherein each of the feature-to-feature transitions is defined by the first sequence of values, the second sequence of values, and the delta time.

5 . The method of claim 4 , wherein the user-defined threshold is at least two feature-to-feature transitions from the first sequence of values to the second sequence of values.

6 . The method of claim 1 , further comprising causing a display device to display the detected pattern.

7 . A pattern recognition system comprising:

a data storage; and

at least one processor coupled to the data storage and configured to execute a process comprising:

receiving, at an input, a signal comprising data;

converting, based on a plurality of features, the data from the signal into a collection of features;

identifying one or more feature-to-feature transitions between features in the collection of features;

plotting a directed graph of each of the feature-to-feature transitions, wherein each unique feature in the collection of features is represented by a single vertex and each feature-to-feature transition is represented by a directed edge between two vertices weighted by a corresponding number of each of the feature-to-feature transitions; and

detecting a pattern in the data by comparing a weight of each of the feature-to-feature transitions in the directed graph to a user-defined threshold;

wherein the converting includes comparing the data to a first sequence of values, a second sequence of values, and a delta time between occurrence of the first sequence of values and the second sequence of values in the data, and wherein the collection of features includes two or more of the features in sequence based on the comparison.

8 . The system of claim 7 , wherein the process further comprises defining each of the plurality of features as a unique sequence of values.

9 . The system of claim 7 , wherein each of the plurality of features includes a unique sequence of values.

10 . The system of claim 7 , wherein each of the feature-to-feature transitions is defined by the first sequence of values, the second sequence of values, and the delta time.

11 . The system of claim 7 , wherein the user-defined threshold is at least two feature-to-feature transitions from the first sequence of values to the second sequence of values.

12 . The system of claim 7 , wherein the process further comprises causing a display device to display the detected pattern.

13 . A computer program product including one or more non-transitory machine-readable mediums encoded with instructions that when executed by one or more processors cause a process to be carried out for correlating signals, the process comprising:

converting, based on a plurality of features, data from a signal into a collection of features;

identifying one or more feature-to-feature transitions between features in the collection of features;

plotting a directed graph of each of the feature-to-feature transitions, wherein each unique feature in the collection of features is represented by a single vertex and each feature-to-feature transition is represented by a directed edge between two vertices weighted by a corresponding number of each of the feature-to-feature transitions; and

detecting a pattern in the data by comparing a weight of each of the feature-to-feature transitions in the directed graph to a user-defined threshold;

wherein the converting includes comparing the data to a first sequence of values, a second sequence of values, and a delta time between occurrence of the first sequence of values and the second sequence of values in the data, and wherein the collection of features includes two or more of the features in sequence based on the comparison.

14 . The computer program product of claim 13 , wherein each of the plurality of features includes a unique sequence of values.

15 . The computer program product of claim 13 , wherein each of the feature-to-feature transitions is defined by the first sequence of values, the second sequence of values, and the delta time.

16 . The computer program product of claim 13 , wherein the user-defined threshold is at least two feature-to-feature transitions from the first sequence of values to the second sequence of values.

17 . The computer program product of claim 13 , wherein the process further comprises causing a display device to display the detected pattern.