Motion localization based on channel response characteristics

• US 10,380,856 B2
• Filed: 10/01/2018
• Issued: 08/13/2019
• Est. Priority Date: 11/16/2017
• Status: Active Grant

• Current Assignee: Cognitive Systems Corp.
• Original Assignee: Cognitive Systems Corp.
• Inventors: Devison, Stephen Arnold, Omer, Mohammad, Piao, Yunfeng, Nogueira, Marco
• Primary Examiner(s): Wang, Jack K
• Application Number:
  US16/148,705
• Publication Number:
  US 20190147713A1
• Time in Patent Office: 316 Days
• Field of Search: 340552
• CPC Class Codes: G08B 13/24 by interference with electromagnetic field distribution
• CPC Class Codes: G08B 13/2491 Intrusion detection systems, i.e. where the body of an intruder causes the interference with the electromagnetic field
• CPC Class Codes: H04L 25/022 of frequency response

Abstract

In a general aspect, detected motion is localized based on channel response characteristics. In some aspects, channel responses based on wireless signals transmitted through a space between wireless communication devices are obtained. A motion detection process is executed to detect motion of an object in the space based on the channel responses, and the channel responses are analyzed to identify a location of the detected motion within one of a plurality of regions within the space.

Claims

1. A motion detection method comprising:

obtaining channel responses based on wireless signals transmitted through a space between wireless communication devices;

executing a motion detection process to detect motion of an object in the space based on the channel responses; and

by operation of one or more processors, analyzing the channel responses to identify a location of the detected motion within one of a plurality of regions within the space.

2. The method of claim 1, wherein analyzing the channel responses to identify the location of the detected motion comprises:

identifying a characteristic of one or more of the channel responses; and

identifying the location based on comparing the identified characteristic with reference characteristics associated with multiple distinct locations within the space.

3. The method of claim 2, wherein identifying the characteristic of one or more of the channel responses comprises identifying a shape of a contour defined by one or more of the channel responses.

4. The method of claim 2, wherein the reference characteristics comprise entries in a motion detection database that associates each of the reference characteristics with one distinct region from among the plurality of regions within the space.

5. The method of claim 2, comprising identifying a change in the characteristic over time and identifying a change in the identified location based on the change in the characteristic over time.

6. The method of claim 1, wherein the channel responses are associated with a series of time points, and the method comprises identifying the location of the motion based on a characteristic shared by channel responses associated with distinct time points in the series.

7. The method of claim 6, wherein the motion detection process detects the motion based on changes over the time series, between channel responses associated with distinct time points.

8. The method of claim 1, wherein each of the wireless signals is transmitted by a first wireless communication device and received by a second wireless communication device, and each of the channel responses is generated from a respective one of the wireless signals received by the second wireless communication device.

9. The method of claim 1, wherein the space comprises a building, and each of the plurality of regions within the space comprise at least one of:

a floor of the building, a room in the building, or a set of rooms in the building, wherein the set of rooms in the building includes less than a totality of the building.

10. The method of claim 1, wherein analyzing the channel responses to identify the location of the detected motion comprises providing the channel responses as input to a neural network and identifying the location of the detected motion based on an output of the neural network.

11. A motion detection system comprising:

multiple wireless communication devices, each wireless communication device configured to detect motion of an object in the space based on a series of wireless signals received from other wireless communication devices; and

a data processing apparatus communicably coupled to the wireless communication devices and configured to;

obtain channel responses based on wireless signals transmitted through a space between the wireless communication devices;

execute a motion detection process to detect motion of an object in the space based on the channel responses; and

analyze the channel responses to identify a location of the motion within one of a plurality of regions within the space.

12. The system of claim 11, wherein the data processing apparatus is configured to analyze the channel responses to identify the location by:

identifying a characteristic of one or more of the channel responses; and

identifying the location based on comparing the identified characteristic with reference characteristics associated with multiple distinct locations within the space.

13. The system of claim 12, wherein the data processing apparatus is configured to identify the characteristic of one or more of the channel responses by identifying a shape of a contour defined by one or more of the channel responses.

14. The system of claim 12, wherein the reference characteristics comprise entries in a motion detection database that associates each of the reference characteristics with one distinct region from among the plurality of regions within the space.

15. The system of claim 12, wherein the data processing apparatus is configured to identify a change in the characteristic over time and identify a change in the identified location based on the change in the characteristic over time.

16. The system of claim 11, wherein the channel responses are associated with a series of time points, and the data processing apparatus is configured to identify the location of the motion based on a characteristic shared by channel responses associated with distinct time points in the series.

17. The system of claim 16, wherein executing the motion detection process comprises detecting the motion based on changes over the time series, between channel responses associated with distinct time points.

18. The system of claim 11, wherein the space comprises a building, and each of the plurality of regions within the space comprise at least one of:

a floor of the building, a room in the building, or a set of rooms in the building, wherein the set of rooms in the building includes less than a totality of the building.

19. The system of claim 11, wherein the data processing apparatus is configured to analyze the channel responses to identify the location of the detected motion by providing the channel responses as input to a neural network and identifying the location of the detected motion based on an output of the neural network.

20. A non-transitory computer-readable medium comprising instructions which, when executed by a processor, cause the processor to:

obtain channel responses based on wireless signals transmitted through a space between wireless communication devices;

execute a motion detection process to detect motion of an object in the space based on the channel responses; and

analyze the channel responses to identify a location of the detected motion within one of a plurality of regions within the space.

21. The medium of claim 20, wherein analyzing the channel responses to identify the location of the detected motion comprises:

identifying a characteristic of one or more of the channel responses; and

identifying the location based on comparing the identified characteristic with reference characteristics associated with multiple distinct locations within the space.

22. The medium of claim 21, wherein identifying the characteristic of one or more of the channel responses comprises identifying a shape of a contour defined by one or more of the channel responses.

23. The medium of claim 21, wherein the reference characteristics comprise entries in a motion detection database that associates each of the reference characteristics with one distinct region from among the plurality of regions within the space.

24. The medium of claim 21, comprising instructions to identify a change in the characteristic over time and identify a change in the identified location based on the change in the characteristic over time.

25. The medium of claim 20, wherein the channel responses are associated with a series of time points, and the medium comprises instructions to identify the location of the motion based on a characteristic shared by channel responses associated with distinct time points in the series.

26. The medium of claim 25, wherein executing the motion detection process comprises detecting the motion based on changes over the time series, between channel responses associated with distinct time points.

27. The medium of claim 20, wherein each of the wireless signals is transmitted by a first wireless communication device and received by a second wireless communication device, and each of the channel responses is generated from a respective one of the wireless signals received by the second wireless communication device.

28. The medium of claim 20, wherein the space comprises a building, and each of the plurality of regions within the space comprise at least one of:

a floor of the building, a room in the building, or a set of rooms in the building, wherein the set of rooms in the building includes less than a totality of the building.

29. The medium of claim 20, wherein analyzing the channel responses to identify the location of the detected motion comprises providing the channel responses as input to a neural network and identify the location of the detected motion based on an output of the neural network.