IDENTIFICATION OF FAILED SENSORS IN A SYSTEM OF INTERCONNECTED DEVICES

• US 20170067763A1
• Filed: 09/04/2015
• Published: 03/09/2017
• Est. Priority Date: 09/04/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kodeswaran, Palanivel A., Kokku, Ravindranath, Sen, Sayandeep, Srivatsa, Mudhakar
• Granted Patent: US 10,378,936 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01D 18/00 Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
• CPC Class Codes: G01D 3/08 with provision for safeguarding the apparatus, e.g. against abnormal operation, against breakdown
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Methods and arrangements for identifying failed sensors in a system of interconnected devices. A contemplated method includes: utilizing at least one processor to execute computer code that performs the steps of: receiving data from a first plurality of sensors, each sensor being operatively coupled to a device within a system of interconnected devices; associating the data received with an activity; comparing the data received with previously stored data associated with the activity; detecting, based on the comparing, an anomaly associated with at least one of the first plurality of sensors, wherein the anomaly indicates a problem with the at least one of the first plurality of sensors; and recording, at a data storage location, the anomaly, wherein the data storage location stores data associated with previously identified anomalies. Other variants and embodiments are broadly contemplated herein.

Claims

1. A method of identifying failed sensors in a system of interconnected devices, said method comprising:

utilizing at least one processor to execute computer code that performs the steps of;

receiving data from a first plurality of sensors, each sensor being operatively coupled to a device within a system of interconnected devices;

associating the data received with an activity;

comparing the data received with previously stored data associated with the activity;

detecting, based on the comparing, an anomaly associated with at least one of the plurality of sensors, wherein the anomaly indicates a problem with the at least one of the plurality of sensors; and

recording, at a data storage location, the anomaly, wherein the data storage location stores data associated with previously identified anomalies.

2. The method of claim 1, further comprising identifying at least one failed sensor within the first plurality of sensors, wherein the at least one failed sensor is identified as the at least one of the first plurality of sensors for which the anomaly was detected.

3. The method of claim 2, wherein the identifying comprises calculating an improbability associated with the at least one of the first plurality of sensors, wherein the improbability identifies a likelihood of sensor failure.

4. The method of claim 2, wherein the identifying comprises:

calculating an improbability associated with the at least one of the first plurality of sensors; and

determining that the improbability exceeds a predetermined threshold.

5. The method of claim 2, wherein the recording comprises increasing a counter associated with the at least one of the first plurality of sensors for which an anomaly is detected.

6. The method of claim 5, wherein the identifying comprises identifying that the counter associated with the at least one of the first plurality of sensors exceeds a predetermined threshold.

7. The method of claim 2, further comprising:

receiving additional data, associated with a second activity, from a second plurality of sensors, wherein at least one sensor is common between the first plurality of sensors and the second plurality of sensors; and

wherein the identifying comprises identifying that the anomaly is associated with the at least one common sensor.

8. The method of claim 1, wherein the anomaly comprises receiving no signal from the at least one of the first plurality of sensors when a signal is expected based upon the previously stored data associated with the activity.

9. The method of claim 1, wherein the anomaly comprises receiving at least one spurious signal from the at least one of the first plurality of sensors when a signal is not expected based upon the previously stored data associated with the activity.

10. An apparatus for identifying failed sensors in a system of interconnected devices, said apparatus comprising:

at least one processor; and

a computer readable storage medium having computer readable program code embodied therewith and executable by the at least one processor, the computer readable program code comprising;

computer readable program code that receives data from a first plurality of sensors, each sensor being operatively coupled to a device within a system of interconnected devices;

computer readable program code that associates the data received with an activity;

computer readable program code that compares the data received with previously stored data associated with the activity;

computer readable program code that detects, based on the comparing, an anomaly associated with at least one of the first plurality of sensors, wherein the anomaly indicates a problem with the at least one of the first plurality of sensors; and

computer readable program code that records, at a data storage location, the anomaly, wherein the data storage location stores data associated with previously identified anomalies.

11. A computer program product for identifying failed sensors in a system of interconnected devices, comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code that receives data from a first plurality of sensors, each sensor being operatively coupled to a device within a system of interconnected devices;

computer readable program code that associates the data received with an activity;

computer readable program code that compares the data received with previously stored data associated with the activity;

computer readable program code that detects, based on the comparing, an anomaly associated with at least one of the first plurality of sensors, wherein the anomaly indicates a problem with the at least one of the first plurality of sensors; and

computer readable program code that records, at a data storage location, the anomaly, wherein the data storage location stores data associated with previously identified anomalies.

12. The computer program product of claim 11, wherein the computer readable program code further identifies at least one failed sensor within the first plurality of sensors, wherein the at least one failed sensor is identified as the at least one of the first plurality of sensors for which the anomaly was detected.

13. The computer program product of claim 12, wherein the computer readable program code that identifies comprises calculating an improbability associated with the at least one of the first plurality of sensors, wherein the improbability identifies a likelihood of sensor failure.

14. The computer program product of claim 12, wherein the computer readable program code that identifies comprises:

calculating an improbability associated with the at least one of the first plurality of sensors; and

determining that the improbability exceeds a predetermined threshold.

15. The computer program product of claim 12, wherein the computer readable program code that records comprises increasing a counter associated with the at least one of the first plurality of sensors for which an anomaly is detected.

16. The computer program product of claim 15, wherein the computer readable program code that identifies comprises identifying that the counter associated with the at least one of the first plurality of sensors exceeds a predetermined threshold.

17. The computer program product of claim 12, wherein the computer readable program code further:

receives additional data, associated with a second activity, from a second plurality of sensors, wherein at least one sensor is common between the first plurality of sensors and the second plurality of sensors; and

wherein the computer readable program code that identifies comprises identifying that the anomaly is associated with the at least one common sensor.

18. The computer program product of claim 11, wherein the anomaly comprises receiving no signal from the at least one of the first plurality of sensors when a signal is expected based upon the previously stored data associated with the activity.

19. The computer program product of claim 11, wherein the anomaly comprises receiving at least one spurious signal from the at least one of the first plurality of sensors when a signal is not expected based upon the previously stored data associated with the activity.

20. A method of identifying failed sensors in a system of interconnected devices, said method comprising:

utilizing at least one processor to execute computer code that performs the steps of;

receiving data from a first plurality of sensors, each sensor being operatively coupled to a device within a system of interconnected devices;

associating the data received with an activity;

comparing the data received with previously stored data associated with the activity;

receiving additional data from a second plurality of sensors, wherein at least one sensor is common between the first plurality of sensors and the second plurality of sensors;

associating the additional data received with a second activity;

comparing the additional data received with previously stored data associated with the second activity;

detecting, based on the comparing, an anomaly associated with the at least one sensor which is common between the first plurality of sensors and the second plurality of sensors, wherein the anomaly indicates a problem with the at least one sensor which is common;

recording, at a data storage location, the anomaly, wherein the data storage location stores data associated with identified anomalies; and

identifying at least one failed sensor within the first plurality of sensors, wherein the at least one failed sensor is identified as the at least one of the first plurality of sensors for which the anomaly was detected;

the identifying comprising calculating an improbability associated with the at least one of the first plurality of sensors, wherein the improbability identifies a likelihood of sensor failure and wherein the identifying further comprises identifying that the improbability exceeds a predetermined threshold.