RFID-BASED ENTERPRISE INTELLIGENCE

• US 20090315678A1
• Filed: 06/18/2008
• Published: 12/24/2009
• Est. Priority Date: 06/18/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Agrawal, Piyush, Sivalingam, Lenin Ravindranath, Padmanabhan, Venkata
• Granted Patent: US 7,982,609 B2
• Time in Patent Office: Days
• US Class Current: 340/10.100
• CPC Class Codes: H04Q 2209/47 using RFID associated with sensors
• CPC Class Codes: H04Q 2209/845 where the measuring is synchronized between sensing devices
• CPC Class Codes: H04Q 9/00 Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom

Abstract

An “RFID-Based Inference Platform” provides various techniques for using RFID tags in combination with other enterprise sensors to track users and objects, infer their interactions, and provide these inferences for enabling further applications. Specifically, observations are collected from combinations of RFID tag reads and other enterprise sensors including electronic calendars, user presence identifiers, cardkey access logs, computer logins, etc. Given sufficient observations, the RFID-Based Inference Platform automatically differentiates between tags associated with or affixed to people and tags affixed to objects. The RFID-Based Inference Platform then infers additional information including identities of people, ownership of specific objects, the nature of different “zones” in a workspace (e.g., private office versus conference room). These inferences are then used to enable various applications including object tracking, automated object ownership determinations, automated object cataloging, automated misplaced object alerts, video annotations, automated conference room scheduling, semi-automated object image catalogs, object interaction query systems, etc.

Claims

1. A method for automatically inferring correlations between entities having RFID tags, comprising steps for:

repeatedly scanning a location having one or more zones of interest using one or more RFID tag readers;

storing all RFID tag data captured by any of the RFID tag readers to a computer readable storage medium;

repeatedly evaluating the stored RFID tag data, including co-location of RFID tags within particular zones, to infer correspondences between two or more RFID tags;

repeatedly evaluating the stored RFID tag data, including RFID tag movement and RFID tag co-movement between two or more zones, to infer relationships between RFID tags; and

repeatedly evaluating the RFID tag correspondences and relationships to infer a type of RFID tag, including which of the RFID tags represent people (“

person tags”

), and which of the RFID tags represent objects (“

object tags).

2. The method of claim 1 further comprising steps for inferring a zone type for each zone from a set of predefined zone types by repeatedly evaluating the RFID tag correspondences, relationships and types.

3. The method of claim 1 further comprising steps for inferring an association between one or more specific object tags and one or more specific person tags by repeatedly evaluating the RFID tag correspondences, relationships and types.

4. The method of claim 3 further comprising steps for providing an automated misplaced object alert to a person linked with a particular person tag when one or more of the RFID readers receives RFID tag data from one of the object tags, having an inferred association by that person tag, in a zone different from the zone in which one or more other RFID readers receives RFID tag data from that person tag.

5. The method of claim 1 wherein inferring correspondences between two or more RFID tags further includes repeatedly evaluating RFID tag interactions within a particular zone.

6. The method of claim 5 further comprising correlating event data to the RFID tag data within each zone based on time-stamps associated with the event data, time-stamps associated with the RFID tag data, and any inferences made from the event data and RFID tag data.

7. The method claim 6 further comprising automatically inferring an identity of persons associated with one or more of the person tags by evaluating the correlated event data and RFID data to associate identity information from one or more elements of the event data to particular person tags.

8. The method of claim 1 further comprising steps for:

storing all inferences to an inference database; and

providing a set of APIs for allowing one or more external programs to directly interact with the inference database.

9. A system for using RFID tag data to infer information regarding tagged entities, comprising:

tagging a plurality of entities with one or more random RFID tags, where the entities include one or more persons and one or more objects;

a device for automatically collecting RFID tag data captured by repeated scans of the RFID tags in one or more zones of interest; and

a device for repeatedly evaluating the RFID tag data for automatically inferring which of the RFID tags are linked to persons (“

person tags”

) and which of the RFID tags are linked to objects (“

object tags”

).

10. The system of claim 9 further comprising a device for automatically inferring correlations between tagged entities by evaluating RFID tag data selected from a group including one or more of RFID tag co-location in one or more of the zones, RFID tag interaction within a particular zone, and RFID tag co-movement between two or more zones.

11. The system of claim 10 further comprising a device for automatically indexing one or more multimedia recordings of one or more areas in one or more of the zones of interest based on the inferred correlations.

12. The system of claim 10 further comprising a device for automatically inferring associations between one or more of the RFID tags based on event data collected from a plurality of sensors, services and monitors within any of the zones of interest.

13. The system of claim 12 further comprising a device for automatically inferring an identity of persons associated with one or more of the person tags by evaluating the inferred associations between the RFID tags and the event data.

14. The system of claim 13 further comprising a device for

storing all inferences to an inference database;

and

providing a set of APIs for allowing one or more external programs to directly interact with the inference database.

15. A computer-readable medium having computer-executable instructions stored thereon for inferring correlations between entities associated with random RFID tags, said instructions comprising:

automatically collecting RFID tag data captured by repeated scans of the RFID tags in one or more zones of interest;

automatically inferring correlations between tagged entities, by evaluating RFID tag data selected from a group comprising one or more of RFID tag co-location in one or more of the zones, RFID tag interaction within a particular zone, and RFID tag co-movement between two or more zones; and

repeatedly evaluating the collected RFID tag data and inferred correlations for automatically inferring which of the entities are persons and which of the entities are objects.

16. The computer-readable medium of claim 15 for automatically inferring zone type for each zone from a set of predefined zone types by repeatedly evaluating the inferred correlations.

17. The computer-readable medium of claim 15 further comprising instructions for automatically inferring associations between one or more of the entities based on event data collected from a plurality of sensors, services and monitors within any of the zones of interest in combination with the inferred correlations and inference of which of the entities are persons and which of the entities are objects.

18. The computer-readable medium of claim 17 further comprising instructions for:

storing all inferences to an inference database; and

providing a set of APIs for allowing one or more external applications to directly interact with the inference database.

19. The computer-readable medium of claim 18 wherein the external applications include one or more of a group comprising:

an automated object tracking application;

an automated object cataloging application;

an automated misplaced object alert application

an automated multimedia indexing application;

an annotated video query application;

an automated conference room scheduling application;

a semi-automated object image cataloging application;

an object ownership query application;

an object interaction query application;

an object location query application;

a person location query application;

a person interaction query application;

an automated object-based security alert application; and

an automated person-based security alert application.

20. The computer-readable medium of claim 15 further comprising instructions for associating a specific identity of one or more persons with one or more of the random RFID tags.