IP Library Granted Patent US 11,906,330
Granted Patent B1
US 11,906,330 · App. 17/889,718 · Granted Feb 20, 2024

Efficient compression of sensor data

Inventor: Gokulmuthu Narayanaswamy (Bangalore, IN)
Assignee: Itron, Inc.
G01D4/004G01R22/063G01R29/04
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,906,330
App. No.
17/889,718
Granted
Feb 20, 2024
Kind
B1
Abstract

Techniques for efficient compression of sensor data are described herein. In an example, metrology data is received from a metrology device, the metrology data comprising one or more of: voltage (V) data; current (A) data; resistive power (W) data; and volt-amps reactive power (VAR) data. The metrology data is processed, wherein the processing comprises: performing peak-detection on the metrology data, to create data-signals comprising: a timestamped peak-values data-signal; and a peak-removed data-signal. Median-filtering is performed on the peak-removed data-signal, wherein a median-filtered data-signal is created. Level-shift detection is performed on the median-filtered data-signal, wherein a timestamped level-shift data-signal is created. The data is sent to a server. The data may include the timestamped peak-values data-signal and the timestamped level-shift data-signal.

Claims (124)

1. A method, comprising:

receiving metrology data from a metrology device, the metrology data comprising one or more of:

voltage (V) data;

current (A) data;

resistive power (W) data; and

volt-amps reactive power (VAR) data;

processing the metrology data, wherein the processing comprises:

performing peak-detection on the metrology data, to create data-signals comprising:

a timestamped peak-values data-signal; and

a peak-removed data-signal;

performing median-filtering on the peak-removed data-signal, wherein a median-filtered data-signal is created; and

performing level-shift detection on the median-filtered data-signal, wherein a timestamped level-shift data-signal is created;

sending data to a server, the data comprising:

the timestamped peak-values data-signal; and

the timestamped level-shift data-signal.

2. The method of claim 1 , wherein performing peak-detection comprises:

configuring a peak threshold of a peak detector;

using the peak detector to identify data associated with at least one peak; and

separating relevant data from less-relevant data based at least in part on the at least one peak.

3. The method of claim 1 , wherein performing median-filtering comprises:

configuring a time-interval window of a median filter;

using the median filter to distinguish noise from relevant data; and

separating the relevant data from less-relevant data based at least in part on operation of the median filter.

4. The method of claim 1 , wherein performing level-shift detection comprises:

configuring a shift threshold of a level-shift detector;

using the level-shift detector to distinguish two or more consumption levels measured by the metrology device; and

separating relevant data from less-relevant data based at least in part on the two or more consumption levels measured by the metrology device.

5. The method of claim 1 , additionally comprising:

determining if less-relevant data is needed by the server; and

based on a need for the less-relevant data, performing actions comprising:

performing signal subtraction, wherein the timestamped peak-values data-signal and the timestamped level-shift data-signal are subtracted from the metrology data to thereby create a less-relevant data-signal;

performing a short-time Fourier transform on the less-relevant data-signal to result in a lossy compressed data-signal; and

sending the lossy compressed data-signal to the server.

6. The method of claim 5 , additionally comprising:

performing a reverse Fourier transform, wherein the lossy compressed data-signal is transformed into a decompressed less-relevant data-signal; and

performing a signal addition, wherein the decompressed less-relevant data-signal, timestamped peak-values data-signal, and timestamped level-shift data-signal are combined to form a restored data-signal.

7. The method of claim 1 , additionally comprising:

receiving information indicating changed data needs of the server or changed network conditions of a network to which the server is connected;

responsive to the information, performing at least one action from a group of actions comprising:

adjusting a parameter governing a minimum peak height to be detected by a peak detector;

adjusting a window size to be used with a median filter;

adjusting a parameter governing a threshold level of change to be detected by a level-shift detector; and

adjusting a window size to be used with a short-time Fourier transform.

8. A network device, comprising:

a processor;

a metrology device configured to measure utility consumption and to thereby produce metrology data;

one or more memory devices in communication with the processor;

applications defined in the one or more memory devices and executed by the processor, and wherein the applications comprise:

a peak detector application, configured to receive the metrology data from the metrology device as input, and configured to create data comprising:

a timestamped peak-values data-signal; and

a peak-removed data-signal;

a median filter application, configured to receive the peak-removed data-signal, and configured to create a median-filtered data-signal; and

a level-shift detector application, configured to receive the median-filtered data-signal, and configured to create a timestamped level-shift data-signal; and

a connection to a network, wherein the processor is configured to cause transmission of the timestamped peak-values data-signal and the timestamped level-shift data-signal over the connection to the network to a server.

9. The network device as recited in claim 8 , additionally comprising instructions stored in the one or more memory devices, that when executed by the processor, cause the processor to perform actions comprising:

responsive to a need to detect peaks of any duration, performing peak detection prior to performing median filtering; and

responsive to a need to detect only peaks over a threshold duration, performing median filtering prior to performing peak detection.

10. The network device as recited in claim 8 , wherein the applications additionally comprise:

a filter selection application to determine which filters are to be used and an order of their use.

11. The network device as recited in claim 8 , wherein the applications additionally comprise:

a filter configuration application to determine parameters for use by each filter selected for operation.

12. The network device as recited in claim 8 , additionally comprising:

a signal subtractor, wherein the signal subtractor is configured to subtract the timestamped peak-values data-signal and the timestamped level-shift data-signal from the metrology data, to thereby create a less-relevant data-signal; and

a short-time Fourier transform application, wherein the short-time Fourier transform application is configured to transform the less-relevant data-signal into a lossy compressed data-signal.

13. The network device as recited in claim 12 , additionally comprising instructions stored in the one or more memory devices, that when executed by the processor, cause the processor to perform actions comprising:

receiving information indicating changed data needs of the server or changed network conditions; and

responsive to the information, performing at least one action from a group of actions comprising:

adjusting a parameter governing a minimum peak height to be detected by a peak detector;

adjusting a window size to be used with a median filter;

adjusting a parameter governing a threshold level change to be detected by a level-shift detector; and

adjusting a window size of the short-time Fourier transform application.

14. The network, comprising the network device as recited in claim 12 and a computing device in communication with the network device over the network, wherein the computing device comprises:

a processor;

one or more memory devices in communication with the processor;

applications respectively defined in the one or more memory devices, and executed by the processor, the applications comprising:

a reverse Fourier transform to decompress the lossy compressed data-signal, thereby creating a decompressed less-relevant data-signal; and

a signal adder, to receive signals comprising:

the timestamped peak-values data-signal;

the timestamped level-shift data-signal; and

the decompressed less-relevant data-signal,

wherein the signal adder outputs a restored data-signal comprising data from the metrology data.

15. The network device as recited in claim 14 , additionally comprising instructions stored in the one or more memory devices, that when executed by the processor, cause the processor to perform actions comprising:

determining a need for use or non-use of the signal subtractor and the short-time Fourier transform;

responsive to a need to use the signal subtractor and the short-time Fourier transform, creating the lossy compressed data-signal; and

responsive to no need to use the signal subtractor and the short-time Fourier transform, creating the restored data-signal without using the decompressed less-relevant data-signal.

16. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by one or more processors, configure a computing device to perform acts comprising:

receiving metrology data from a metrology device, the metrology data comprising one or more of:

voltage (V);

current (A);

resistive power (W); and

volt-amps reactive power (VAR) data;

processing the metrology data, wherein the processing comprises:

performing peak-detection on the metrology data, to create data-signals comprising:

a timestamped peak-values data-signal; and

a peak-removed data-signal;

performing median-filtering on the peak-removed data-signal, wherein a median-filtered data-signal is created; and

performing level-shift detection on the median-filtered data-signal, wherein a timestamped level-shift data-signal is created; and

sending data to a server, the data comprising:

the timestamped peak-values data-signal; and

the timestamped level-shift data-signal.

17. The one or more non-transitory computer-readable media as recited in claim 16 , wherein the acts additionally comprise:

determining that a currently available network bandwidth is greater than, or less than, a previously available network bandwidth that was prior to the currently available network bandwidth; and

responsive to the determining, performing at least one action from a group of actions comprising:

adjusting a parameter governing a minimum peak height to be detected by a peak detector;

adjusting a window size to be used with a median filter;

adjusting a parameter governing a threshold level of change to be detected by a level shift detector; and

adjusting a window size to be used with a short-time Fourier transform.

18. The one or more non-transitory computer-readable media as recited in claim 16 , wherein the acts additionally comprise:

determining a measurement of an available network bandwidth; and

responsive to the measurement, adjusting a parameter of one or more tools comprising:

a median filter;

a level-shift detector;

a peak detector; and

a short-time Fourier transform;

wherein if the measurement of the available network bandwidth has increased over a previous measurement of the available network bandwidth, then a first parameter of the one or more tools is adjusted to increase data sent to the server; and

wherein if the measurement of the available network bandwidth has decreased over the previous measurement of the available network bandwidth, then a second parameter of the one or more tools is adjusted to decrease data sent to the server.

19. The one or more non-transitory computer-readable media as recited in claim 16 , wherein the acts additionally comprise:

performing signal subtraction, wherein the timestamped peak-values data-signal and the timestamped level-shift data-signal are subtracted from the metrology data to thereby create a less-relevant data-signal;

performing a short-time Fourier transform on the less-relevant data-signal to result in a lossy compressed data-signal; and

sending the lossy compressed data-signal to the server.

20. The one or more non-transitory computer-readable media as recited in claim 19 , wherein the acts additionally comprise:

receiving the timestamped peak-values data-signal, the timestamped level-shift data-signal, and the lossy compressed data-signal at the server;

performing a reverse Fourier transform, wherein the lossy compressed data-signal is decompressed to result in an approximation of the less-relevant data-signal; and

performing signal addition, wherein the timestamped peak-values data-signal, the timestamped level-shift data-signal, and the less-relevant data-signal are combined into a restored data-signal.

Assignments (2)
SECURITY INTEREST Recorded Sep 15, 2025
From: ITRON, INC.
To: WELLS FARGO BANK, NATIONAL ASSOCIATION
Reel/Frame 072870/0873 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 7, 2024
From: NARAYANASWAMY, GOKULMUTHU
To: ITRON, INC.
Reel/Frame 067656/0153 →
Cited By (1)
US 12,695,656