Correction of sensor data in a multi-sensor internet of things environment

1. A method, comprising:

obtaining sensor data from a first sensor;

applying, by at least one processing device, an anomaly detection technique to detect one or more anomalous sensor values in the sensor data from the first sensor based on additional sensor data from one or more of the first sensor and at least one additional sensor in proximity to the first sensor, wherein a given anomalous sensor value in the sensor data from the first sensor is detected by evaluating (i) a first difference between a first predicted sensor value for the first sensor based at least in part on historical data from the first sensor and a measured sensor value from the first sensor, and (ii) a second difference between a second predicted sensor value for the first sensor based at least in part on at least one measured sensor value from the at least one additional sensor and the measured sensor value from the first sensor; and

correcting, by the at least one processing device, at least one of the one or more anomalous sensor values in the sensor data from the first sensor using additional sensor data from one or more of the first sensor and the at least one additional sensor.

2. The method of claim 1 , wherein the anomaly detection technique employs a prediction method to detect the anomaly in the sensor data from the first sensor based on the additional sensor data from the first sensor.

3. The method of claim 1 , wherein the at least one additional sensor in proximity to the first sensor comprises a plurality of additional sensors that satisfy one or more neighboring criteria with respect to the first sensor and wherein the additional sensor data from the plurality of additional sensors is used to predict the sensor data from the first sensor.

4. The method of claim 1 , wherein the first sensor and the at least one additional sensor comprise sensors that have a substantially similar sensor type and a substantially similar sensor location.

5. The method of claim 1 , wherein the step of applying the anomaly detection technique to detect the one or more anomalous sensor values in the sensor data further comprises estimating a value of the sensor data from the first sensor based on a predefined source of information.

6. The method of claim 1 , wherein the step of correcting the anomalous sensor data is performed substantially close in time to the step of applying the anomaly detection technique to detect the anomaly in the sensor data.

7. The method of claim 1 , wherein the correcting the one or more anomalous values in the sensor data comprises determining at least one corrected sensor value based at least in part on a difference between a predicted sensor value and a measured sensor value.

8. The method of claim 1 , wherein the given anomalous sensor value in the sensor data from the first sensor is classified as anomalous in response to both the first difference and the second difference indicating an anomaly.

9. The method of claim 1 , wherein the correcting the one or more anomalous sensor values in the sensor data comprises using a weighted average of the first predicted sensor value and the second predicted sensor value.

10. A system, comprising:

a memory; and

at least one processing device, coupled to the memory, operative to implement the following steps:

obtaining sensor data from a first sensor;

applying an anomaly detection technique to detect one or more anomalous sensor values in the sensor data from the first sensor based on additional sensor data from one or more of the first sensor and at least one additional sensor in proximity to the first sensor, wherein a given anomalous sensor value in the sensor data from the first sensor is detected by evaluating (i) a first difference between a first predicted sensor value for the first sensor based at least in part on historical data from the first sensor and a measured sensor value from the first sensor, and (ii) a second difference between a second predicted sensor value for the first sensor based at least in part on at least one measured sensor value from the at least one additional sensor and the measured sensor value from the first sensor; and

correcting at least one of the one or more anomalous sensor values in the sensor data from the first sensor using additional sensor data from one or more of the first sensor and the at least one additional sensor.

11. The system of claim 10 , wherein the anomaly detection technique employs a prediction method to detect the anomaly in the sensor data from the first sensor based on the additional sensor data from the first sensor.

12. The system of claim 10 , wherein the at least one additional sensor in proximity to the first sensor comprises a plurality of additional sensors that satisfy one or more neighboring criteria with respect to the first sensor and wherein the additional sensor data from the plurality of additional sensors is used to predict the sensor data from the first sensor.

13. The system of claim 10 , wherein the first sensor and the at least one additional sensor comprise sensors that have a substantially similar sensor type and a substantially similar sensor location.

14. The system of claim 10 , wherein the step of applying the anomaly detection technique to detect the one or more anomalous sensor values in the sensor data further comprises estimating a value of the sensor data from the first sensor based on a predefined source of information.

15. The system of claim 10 , wherein the step of correcting the anomalous sensor data is performed substantially close in time to the step of applying the anomaly detection technique to detect the anomaly in the sensor data.

16. A non-transitory machine-readable storage medium having encoded therein executable code of one or more software programs, wherein the one or more software programs when executed by at least one processing device perform the following steps:

obtaining sensor data from a first sensor;

applying an anomaly detection technique to detect one or more anomalous sensor values in the sensor data from the first sensor based on additional sensor data from one or more of the first sensor and at least one additional sensor in proximity to the first sensor, wherein a given anomalous sensor value in the sensor data from the first sensor is detected by evaluating (i) a first difference between a first predicted sensor value for the first sensor based at least in part on historical data from the first sensor and a measured sensor value from the first sensor, and (ii) a second difference between a second predicted sensor value for the first sensor based at least in part on at least one measured sensor value from the at least one additional sensor and the measured sensor value from the first sensor; and

correcting at least one of the one or more anomalous sensor values in the sensor data from the first sensor using additional sensor data from one or more of the first sensor and the at least one additional sensor.

17. The computer program product of claim 16 , wherein the anomaly detection technique employs a prediction method to detect the anomaly in the sensor data from the first sensor based on the additional sensor data from the first sensor.

18. The computer program product of claim 16 , wherein the at least one additional sensor in proximity to the first sensor comprises a plurality of additional sensors that satisfy one or more neighboring criteria with respect to the first sensor and wherein the additional sensor data from the plurality of additional sensors is used to predict the sensor data from the first sensor.

19. The computer program product of claim 16 , wherein the first sensor and the at least one additional sensor comprise sensors that have a substantially similar sensor type and a substantially similar sensor location.

20. The computer program product of claim 16 , wherein the step of correcting the anomalous sensor data is performed substantially close in time to the step of applying the anomaly detection technique to detect the anomaly in the sensor data.