Contextualized anomaly analyzer

Patent metadata

Field	Value
Publication number	US-12423207-B2
Application number	US-202117559852-A
Country	US
Kind code	B2
Filing date	Dec 22, 2021
Priority date	Dec 22, 2021
Publication date	Sep 23, 2025
Grant date	Sep 23, 2025

Abstract

Methods, systems and apparatuses may provide for technology that transforms input data into a set of reference waveforms, defines a context space for the set of reference waveforms, and determines whether sample data is an anomaly based on the context space.

First claim

I claim:

1. A computing system comprising:
a network controller;
a processor coupled to the network controller; and
a memory coupled to the processor, the memory including a set of instructions, which when executed by the processor, causes the processor to:
convert an input signal into input data,
transform the input data into a set of reference waveforms, wherein the input data comprises multidimensional audio or video data,
define a context space for the set of reference waveforms, and
determine whether sample data is an anomaly based on the context space, wherein the sample data comprising multidimensional audio or video data, wherein determining sample data is an anomaly is to:
transform the sample data into a set of sample waveforms,
define a sample space for the set of sample waveforms, and
identify the sample data as the anomaly upon an overlap between the sample space and one or more of the context space or a subset of vertices associated with the context space below a threshold.

2. The computing system of claim 1 , wherein to define the context space, the instructions, when executed, further cause the computing system to combine the set of reference waveforms.

3. The computing system of claim 2 , wherein to combine the set of reference waveforms, the instructions, when executed, further cause the computing system to determine a maximum profile and a minimum profile of the set of reference waveforms.

4. The computing system of claim 1 , wherein the context space is to contain a fewer number of dimensions than the input data.

5. A semiconductor apparatus comprising:
one or more substrates; and
hardware logic coupled to the one or more substrates, wherein the hardware logic is implemented at least partly in one or more of configurable or fixed-functionality hardware, the hardware logic to:
convert an input signal into input data,
transform input data into a set of reference waveforms, wherein the input data comprises multidimensional audio or video data;
define a context space for the set of reference waveforms; and
determine whether sample data is an anomaly based on the context space, wherein the sample data comprising multidimensional audio or video data, wherein determining sample data is an anomaly is to:
transform the sample data into a set of sample waveforms,
define a sample space for the set of sample waveforms, and
identify the sample data as the anomaly upon an overlap between the sample space and one or more of the context space or a subset of vertices associated with the context space below a threshold.

6. The semiconductor apparatus of claim 5 , wherein to define the context space, the hardware logic is to combine the set of reference waveforms.

7. The semiconductor apparatus of claim 6 , wherein to combine the set of reference waveforms, the hardware logic is to determine a maximum profile and a minimum profile of the set of reference waveforms.

8. The semiconductor apparatus of claim 5 , wherein the context space is to contain a fewer number of dimensions than the input data.

9. At least one non-transitory computer readable storage medium comprising a set of instructions, which when executed by hardware logic, causes the hardware logic to:
convert an input signal into input data,
transform the input data into a set of reference waveforms, wherein the input data comprises multidimensional audio or video data;
define a context space for the set of reference waveforms; and
determine whether sample data is an anomaly based on the context space, wherein the sample data comprising multidimensional audio or video data, wherein determining sample data is an anomaly is to:
transform the sample data into a set of sample waveforms,
define a sample space for the set of sample waveforms, and
identify the sample data as the anomaly upon an overlap between the sample space and one or more of the context space or a subset of vertices associated with the context space below a threshold.

10. The at least one non-transitory computer readable storage medium of claim 9 , wherein to define the context space, the instructions, when executed, further cause the hardware logic to combine the set of reference waveforms.

11. The at least one non-transitory computer readable storage medium of claim 10 , wherein to combine the set of reference waveforms, the instructions, when executed, further cause the hardware logic to determine a maximum profile and a minimum profile of the set of reference waveforms.

12. The at least one non-transitory computer readable storage medium of claim 9 , wherein the context space is to contain a fewer number of dimensions than the input data.

13. A method performed by a computing system, comprising:
convert an input signal into input data;
transforming the input data into a set of reference waveforms, wherein the input data comprises multidimensional audio or video data;
defining a context space for the set of reference waveforms; and
determining whether sample data is an anomaly based on the context space, wherein the sample data comprising multidimensional audio or video data, wherein determining sample data is an anomaly is to:
transform the sample data into a set of sample waveforms,
define a sample space for the set of sample waveforms, and
identify the sample data as the anomaly upon an overlap between the sample space and one or more of the context space or a subset of vertices associated with the context space below a threshold.

14. The method of claim 13 , wherein defining the context space includes combining the set of reference waveforms.

15. The method of claim 14 , wherein combining the set of reference waveforms includes determining a maximum profile and a minimum profile of the set of reference waveforms.

16. The method of claim 13 , wherein the context space contains a fewer number of dimensions than the input data.

Assignees

• Intel Corp

Inventors

• Calabria Luca

Classifications

• G06F11/3006

  where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50) · CPC title

• G06F11/076

  by exceeding a count or rate limit, e.g. word- or bit count limit · CPC title

• G06F11/0772

  Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers · CPC title

• G06F11/322 Primary

  Display of waveforms, e.g. of logic analysers (G06F11/323 takes precedence) · CPC title