Multi-sensor edge computing system

1. A system comprising a sensor assembly connected to an edge computing device in a single building and to a downstream host positioned external to the single building, the sensor assembly comprising a first sensor and a second sensor, the edge computing device comprising a local controller configured to identify a false reading of the first sensor in response to the local controller determining motion detected in the single building by the second sensor is artificial, not real, and will interfere with operation of the first sensor.

2. The system of claim 1 , wherein the edge computing device is physically connected to the sensor assembly within a single room of the single building.

3. The system of claim 1 , wherein the first sensor is a particle counting sensor and the second sensor is a camera.

4. The system of claim 1 , wherein each sensor is positioned within a housing with the local controller and network adapter.

5. The system of claim 4 , wherein each sensor is connected to a remote controller of the downstream host via the respective network adapters.

6. A method comprising:

connecting a first sensor to a first edge computing device in a single building;

connecting a second sensor to a second edge computing device in the single building, each edge computing device connected to a downstream host;

detecting motion with the first edge computing device from data collected by the first sensor;

classifying the detected motion as artificial, not real, and likely to interfere with the second sensor by a local controller of the first edge computing device; and

identifying a false reading of the second sensor in response to the detected artificial motion, the second sensor functioning normally.

7. The method of claim 6 , wherein the first sensor detects different conditions in the single building than the second sensor.

8. The method of claim 6 , wherein the first edge computing device initializes motion detection to accumulate a weighted average between at least two frames of data collected by the first sensor.

9. The method of claim 8 , wherein the first edge computing device computes differences in at least one frame of data collected by the first sensor.

10. The method of claim 6 , wherein the first edge computing device defines contours in the at least two frames of data.

11. The method of claim 10 , wherein the first edge computing device eliminates at least one frame-to-frame characteristic from the at least two frames of data.

12. The method of claim 6 , wherein the first edge computing device employs a contour mapping algorithm to classify the detected motion.

13. A method comprising:

connecting a first sensor to a first edge computing device in a single building;

connecting a second sensor to a second edge computing device in the single building, each edge computing device connected to a downstream host;

detecting motion with the first edge computing device from data collected by the first sensor;

classifying the detected motion as artificial and not real;

predicting a false reading of the second sensor in response to the detected artificial motion; and

ignoring readings from the second sensor until the detection motion stops.

14. The method of claim 13 , wherein the motion is classified by the first edge computing device.

15. The method of claim 13 , wherein the detected motion is correlated to an increased likelihood of motion by the first edge computing device.

16. The method of claim 15 , wherein the downstream host identifies the increased likelihood of motion increases a likelihood of artificial readings.

17. The method of claim 13 , wherein the prediction of a false reading results from a prediction of motion interference above a predetermined accuracy threshold.

18. The method of claim 13 , wherein the first edge computing device proactively alters a recording parameter in response to the predicted false reading.

19. The method of claim 13 , wherein the first edge computing device proactively activates a secondary sensor in response to the predicted false reading.

20. The method of claim 19 , wherein the secondary sensor is a redundant sensor to the first sensor.