Sensor data compression in a multi-sensor internet of things environment

1. A method, comprising:

obtaining non-image sensor data from a plurality of sensors;

applying, by at least one edge-based processing device, an image-based compression technique to the non-image sensor data to generate compressed sensor data responsive to the plurality of sensors satisfying one or more predefined sensor proximity criteria based on a distance between the plurality of sensors; and

providing, by the at least one edge-based processing device, the compressed sensor data to a data center.

2. The method of claim 1 , wherein the non-image sensor data from one or more of the plurality of sensors comprises non-image sensor data compressed using a prediction-based single sensor compression technique.

3. The method of claim 1 , further comprising predicting, using the at least one edge-based processing device, a future value of the non-image sensor data from one or more of the plurality of sensors and providing the predicted future value of the non-image sensor data to the one or more sensors.

4. The method of claim 1 , wherein the image-based compression technique comprises a discrete cosine transform technique, and further comprising normalizing the non-image sensor data prior to the applying step.

5. The method of claim 1 , wherein the non-image sensor data is obtained over time and wherein the image-based compression technique comprises a video compression technique.

6. The method of claim 1 , wherein the image-based compression technique employs an auto-encoder deep learning technique that utilizes one or more over-fitted bidirectional recurrent convolutional neural networks to compress the non-image sensor data obtained over time.

7. The method of claim 1 , wherein the plurality of sensors has a substantially similar sensor type and a substantially similar sensor location.

8. The method of claim 1 , further comprising estimating a value of the non-image sensor data from one or more of the plurality of sensors based on a predefined source of environmental information.

9. A system, comprising:

a memory; and

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

obtaining non-image sensor data from a plurality of sensors;

applying, by at least one edge-based processing device, an image-based compression technique to the non-image sensor data to generate compressed sensor data responsive to the plurality of sensors satisfying one or more predefined sensor proximity criteria based on a distance between the plurality of sensors; and

providing, by the at least one edge-based device, the compressed sensor data to a data center.

10. The system of claim 9 , wherein the non-image sensor data from one or more of the plurality of sensors comprises non-image sensor data compressed using a prediction-based single sensor compression technique.

11. The system of claim 9 , further comprising predicting, by the at least one edge-based device, a future value of the non-image sensor data from one or more of the plurality of sensors and providing the predicted future value of the non-image sensor data to the one or more sensors.

12. The system of claim 9 , wherein the image-based compression technique comprises a discrete cosine transform technique, and further comprising normalizing the non-image sensor data prior to the applying step.

13. The system of claim 9 , wherein the non-image sensor data is obtained over time and wherein the image-based compression technique comprises a video compression technique.

14. The system of claim 9 , wherein the image-based compression technique employs an auto-encoder deep learning technique that utilizes one or more over-fitted bidirectional recurrent convolutional neural networks to compress the non-image sensor data obtained over time.

15. A computer program product, comprising 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 non-image sensor data from a plurality of sensors;

applying, by at least one edge-based processing device, an image-based compression technique to the non-image sensor data to generate compressed sensor data responsive to the plurality of sensors satisfying one or more predefined sensor proximity criteria based on a distance between the plurality of sensors; and

providing, by the at least one edge-based device, the compressed sensor data to a data center.

16. The computer program product of claim 15 , wherein the non-image sensor data from one or more of the plurality of sensors comprises non-image sensor data compressed using a prediction-based single sensor compression technique.

17. The computer program product of claim 15 , further comprising predicting, by the at least one edge-based device, a future value of the non-image sensor data from one or more of the plurality of sensors and providing the predicted future value of the non-image sensor data to the one or more sensors.

18. The computer program product of claim 15 , wherein the image-based compression technique comprises a discrete cosine transform technique, and further comprising normalizing the non-image sensor data prior to the applying step.

19. The computer program product of claim 15 , wherein the non-image sensor data is obtained over time and wherein the image-based compression technique comprises a video compression technique.

20. The computer program product of claim 15 , wherein the image-based compression technique employs an auto-encoder deep learning technique that utilizes one or more over-fitted bidirectional recurrent convolutional neural networks to compress the non-image sensor data obtained over time.