Pre-Processing Of Geo-Spatial Sensor Data

• US 20150268929A1
• Filed: 03/19/2014
• Published: 09/24/2015
• Est. Priority Date: 03/19/2014
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Foebel, Florian, Kazmaier, Gerrit Simon, Licht, Nico, Abraham, Torsten, Gruschko, Boris, Lahmer, Christian, Lieberenz, Marcus, Volker, Lars
• Granted Patent: US 9,665,631 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/25 Integrating or interfacing systems involving database management systems
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/86 Mapping to a database

Abstract

Data is received that is derived from a plurality of geo-spatial sensors that respectively generate data characterizing a plurality of sources within a zone of interest. The data includes series time-stamped frames for each of the sensors and at least one of the sources has two or more associated sensors. The received data can be sorted and processed, for each sensor on a sensor-by-sensor basis, using a sliding window. The sorted and processed data can then be correlated and written into a data storage application. Related apparatus, systems, techniques and articles are also described.

Claims

1. A method comprising:

receiving data derived from a plurality of geo-spatial sensors that respectively generate data characterizing a plurality of sources within a zone of interest, the data comprising a series time-stamped frames for each of the sensors, at least one of the sources having two or more associated sensors;

sorting and processing, for each sensor on a sensor-by-sensor basis, the received data using a sliding window;

correlating the sorted and processed data for each source; and

writing the correlated data into a data storage application.

2. A method as in claim 1, wherein the sorting and processing comprises:

filtering frames based on their respective timestamps.

3. A method as in claim 2, wherein filtering the frames comprises:

recording, for each sensor, a system time and a time of a first recorded time-stamped frame;

determining an expected timestamp for arriving frames; and

discarding frames having a timestamp outside a pre-defined time window relative to the expected timestamp.

4. A method as in claim 1, wherein the sorting and processing comprises:

subdividing a time bar into buckets according to a given sample rate;

assigning each filtered frame into a corresponding bucket; and

interpolating missing values if an amount of time between two valid sensor values span multiple buckets.

5. A method as in claim 4, wherein the interpolating is linear.

6. A method as in claim 5, wherein correlating the sorted and processed data for each source comprises:

assigning data in each bucket to a source by transposition using a mapping table.

7. A method as in claim 1 further comprising:

feeding the sorted and correlated data into a pre-processing library prior to writing the data into the data storage application.

8. A method as in claim 1, wherein the data storage application comprises an in-memory database.

9. A method as in claim 8, wherein the in-memory database stores data in a columnwise fashion.

10. A method as in claim 1, wherein the receiving, sorting and processing, correlating, and writing are implemented by at least one data processor forming part of at least one computing system.

11. A non-transitory computer program product storing instructions which, when executed by at least one data processor forming part of at least one computing system, result in operations comprising:

receiving data derived from a plurality of geo-spatial sensors that respectively generate data characterizing a plurality of sources within a zone of interest, the data comprising a series time-stamped frames for each of the sensors, at least one of the sources having two or more associated sensors;

sorting and processing, for each sensor on a sensor-by-sensor basis, the received data using a sliding window;

correlating the sorted and processed data for each source; and

writing the correlated data into a data storage application.

12. A computer program product as in claim 11, wherein the sorting and processing comprises:

filtering frames based on their respective timestamps.

13. A computer program product as in claim 12, wherein filtering the frames comprises:

recording, for each sensor, a system time and a time of a first recorded time-stamped frame;

determining an expected timestamp for arriving frames; and

discarding frames having a timestamp outside a pre-defined time window relative to the expected timestamp.

14. A computer program product as in claim 11, wherein the sorting and processing comprises:

subdividing a time bar into buckets according to a given sample rate;

assigning each filtered frame into a corresponding bucket; and

interpolating missing values if an amount of time between two valid sensor values span multiple buckets.

15. A computer program product as in claim 14, wherein the interpolating is linear.

16. A computer program product as in claim 15, wherein correlating the sorted and processed data for each source comprises:

assigning data in each bucket to a source by transposition using a mapping table.

17. A computer program product as in claim 11, wherein the operations further comprise:

feeding the sorted and correlated data into a pre-processing library prior to writing the data into the data storage application.

18. A computer program product as in claim 11, wherein the data storage application comprises an in-memory database.

19. A computer program product as in claim 18, wherein the in-memory database stores data in a columnwise fashion.

20. A system comprising:

an in-memory database;

at least one data processor; and

memory storing instructions which, when executed by the at least one data processor, result in operations comprising;

receiving data derived from a plurality of geo-spatial sensors that respectively generate data characterizing a plurality of sources within a zone of interest, the data comprising a series time-stamped frames for each of the sensors, at least one of the sources having two or more associated sensors;

sorting and processing, for each sensor on a sensor-by-sensor basis, the received data using a sliding window;

correlating the sorted and processed data for each source; and

writing the correlated data into the in-memory database.