CLASSIFYING THE POSITION OF A WIRELESS DEVICE

• US 20120149388A1
• Filed: 12/14/2010
• Published: 06/14/2012
• Est. Priority Date: 12/14/2010
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: West, Earle, West, Nathan E.
• Granted Patent: US 8,509,806 B2
• Time in Patent Office: Days
• US Class Current: 455/456.1
• CPC Class Codes: G01S 5/0252 Radio frequency fingerprinting
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/022 with dynamic range variability
• CPC Class Codes: H04W 4/025 using location based information parameters
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/21 for social networking applications

Abstract

Aspects describe using mobile device telemetry data with location reports from a small fraction of reporting devices to pre-compute a set of categorizing formulas that can quickly assign devices of unknown latitude, longitude, and/or altitude to specific values of latitude, longitude, and/or altitude. The described aspects perform a large percentage of complicated calculations in advance of a particular device of unknown location presenting itself on the radio network. Thus, the categorizing formulas can be applied to categorize the device into a particular range of latitude, longitude, and/or altitude with minimal delays. Various aspects relate to enabling an efficient and highly scalable implementation of mobile device advertising, social networking, and other applications that may reply in rapid, real time delivery of “WITHIN” or NOT WITHIN a “fenced-in” region or geography or shape of space.

Claims

1. A method of locating a user device in a communication network, comprising:

computing a set of hyperplanes using historical data of known timing measurements and associated location information;

receiving a reported time measurement from the user device;

applying the reported time measurement to the set of hyperplanes; and

determining a location of the user device as a function of the applying.

2. The method of claim 1, wherein the computing comprises:

obtaining telemetry data from a plurality of mobile devices, wherein the plurality of mobile devices are location aware devices;

combining the telemetry data with the location information from the plurality of mobile devices; and

computing a subset of categorizing formulas based on the telemetry data and the location information.

3. The method of claim 1, wherein the applying comprises:

receiving a collection of timing offset measures from the user device; and

assigning a location value to the user device, wherein the location value comprises at least one of a latitude, a longitude, or an altitude.

4. The method of claim 1, wherein the receiving comprises receiving at least one timing offset measurement from the user device, the applying comprises comparing the at least one timing offset measurement with the set of hyperplanes, and the determining comprises determining a classification of the at least one timing offset measurement based on the comparing, wherein the classification provides an indication of the location of the user device.

5. The method of claim 1, further comprising retaining the set of hyperplanes after the computing, wherein the set of hyperplanes are updated periodically.

6. The method of claim 1, wherein the calculating comprises solving a least squares optimization problem that positions each hyperplane between points that are closest to a hyperplane boundary.

7. The method of claim 1, wherein the applying comprises classifying the reported time measurement in a sequence of one or more classification operations that provide bounds for upper and lower limits of latitude, longitude, or altitude, or combinations thereof.

8. The method of claim 1, wherein the determining comprises individually evaluating a latitude, a longitude, and an altitude of the user device.

9. The method of claim 1, wherein the computing comprises using a kernel function to map a vector of measurements received from a mobile device to a set of measurements located in a higher dimension.

10. The method of claim 1, further comprising performing an error estimate for at least one hyperplane in the set of hyperplanes.

11. The method of claim 1, wherein the determining comprises employing a support vector machine.

12. A communication apparatus, comprising:

a location manager component configured to create a location model based on at least one measurement received from a location aware mobile device; and

a categorization processing component configured to utilize the location model to determine a location of a non-location aware mobile device.

13. The communication apparatus of claim 12, wherein the location manager component is further configured to determine at least one classification of values for one or more dimensions of latitude, longitude or altitude using an array of measurements reported by the location aware mobile device.

14. The communication apparatus of claim 13, wherein the categorization processing component is further configured to utilize the at least one classification of values to determine the location of the non-location aware mobile device.

15. The communication apparatus of claim 12, wherein the categorization processing component is further configured to receive the at least one measurement from the non-location aware mobile device and classify the at least one measurement in a sequence of classification operations that provide upper and lower bounds for at least one of a latitude, a longitude, and an altitude.

16. The communication apparatus of claim 12, wherein the categorization processing component is further configured to utilize one or more kernel functions that map a vector of measurements provided by the non-location aware mobile device to a set of measurements in a higher dimension.

17. The communication apparatus of claim 12, wherein the categorization processing component is further configured to individually evaluate a latitude, a longitude, and an altitude to determine the location of the non-location aware mobile device.

18. A method for classifying a mobile device as a function of a point of interest, comprising:

receiving geographic boundary information;

determining expected timing differences at points bracketing each side of a geographic boundary;

individually categorizing the expected timing differences;

ascertaining coefficients of a linear model;

classifying a data vector as a function of the coefficients; and

sending a notification to a network if the classifying indicates that the mobile device is near the point of interest.

19. The method of claim 18, wherein the classifying comprises classifying the data vector as “

near”

or “

not near”

the point of interest defined by the geographic boundary information.

20. The method of claim 18, wherein the categorizing comprises deriving a training set from the individual categorizations, wherein the training set brackets sides of the geographic boundary.