System and method for determination of proximity between wireless devices

• US 9,313,722 B2
• Filed: 08/23/2013
• Issued: 04/12/2016
• Est. Priority Date: 05/28/2013
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Carranza, Marcos A., Hirsch, Dan F., Oviedo, Marcos J.
• Primary Examiner(s): MITCHELL, NATHAN A
• Application Number:
  US13/974,560
• Publication Number:
  US 20140357267A1
• Time in Patent Office: 963 Days
• US Class Current: 1/1
• CPC Class Codes: H04W 48/16 Discovering, processing access restriction or access information

Abstract

Methods and systems to determine the proximity of wireless devices to each other. In an embodiment, the proximity may be determined as a membership in a classification, e.g., high, medium, or low proximity of one wireless device to a second. The process for determining proximity may include a training phase, where a set of wireless devices may perform scans to determine the identities and number of wireless access points, such as wireless routers. From this, statistical features may be extracted and used to perform training; the output of this training may be a set of trained models. The trained models may be used in an operational phase. Here, scan data from wireless devices may be submitted to a feature extraction process. The extracted features may then be used to determine the trained model most closely fitting the extracted features.

Claims

1. An apparatus, comprising, a processor and memory configured to:

receive a first set of scan data that includes information regarding wireless access points available to each device of the first group;

compute a set of statistics for each of multiple subsets of devices of the first group based on the first set of scan data;

classify each of the subsets of devices in one of multiple proximity classes;

train a model to classify the subsets of devices in the respective proximity classes based on the respective sets of statistics;

receive a second set of scan data that includes information regarding wireless access points available to each device of a second group;

compute the set of statistics for the second group of devices based on the second set of scan data; and

classify the second group devices in one of the proximity classes based on the set of statistics of the second group and the model.

2. The apparatus of claim 1, wherein the first set of scan data includes indications of signal strengths of the wireless access points accessible to the respective devices of the first group, and wherein the processor and memory are further configured to:

compute the set of statistics for each subset of devices of the first group based on the wireless access points accessible to the devices of the respective subset and the respective signal strengths.

3. The apparatus of claim 1, wherein the processor and memory are further configured to compute the set of statistics for each subset of devices of the first group to include one or more of:

a number of wireless access points that are accessible to multiple devices of the respective subset,

a difference in signal strengths of a wireless access point that is accessible to multiple devices of the respective subset,

a number of wireless access points that are not accessible to multiple devices of the respective subset,

a percentage of a total signal strength of a wireless access point detected by each of multiple devices of the respective subset,

a percentage of wireless access points accessible to multiple devices of the respective subset, relative to wireless access points accessible to at least one device of the respective subset, and

an indication if a wireless access point provides a highest signal strength to all devices of the respective subset of devices.

4. The apparatus of claim 1, wherein the processor and memory are further configured to compute the set of statistics for pairs of devices of the first group.

5. The apparatus of claim 1, wherein a device of a third group of devices is configured to exchange scan data with another device of the third group, compute the set of statistics for the third group of devices based on the exchanged scan data, and classify the third group of devices in one of the proximity classes based on the set of statistics of the third group and the model.

6. The apparatus of claim 1, wherein the proximity classes correspond to proximities of the respective devices relative to one another.

7. A non-transitory computer readable medium encoded with a computer program that includes instructions to cause a processor to:

receive a first set of scan data that includes information regarding wireless access points available to each device of the first group;

compute a set of statistics for each of multiple subsets of devices of the first group based on the first set of scan data;

classify each of the subsets of devices in one of multiple proximity classes;

train a model to classify the subsets of devices in the respective proximity classes based on the respective sets of statistics;

receive a second set of scan data that includes information regarding wireless access points available to each device of a second group;

compute the set of statistics for the second group of devices based on the second set of scan data; and

classify the second group of devices in one of the proximity classes based on the set of statistics of the second group and the model.

8. The non-transitory computer readable medium of claim 7, wherein the first set of scan data includes indications of signal strengths of the wireless access points accessible to the respective devices of the first group, and wherein the instructions include instructions to cause the processor to:

compute the set of statistics for each subset of devices of the first group based on the wireless access points accessible to the devices of the respective subset and the respective signal strengths.

9. The non-transitory computer readable medium of claim 7, further including instructions to cause the processor to compute the set of statistics for each subset of devices of the first group to include one or more of:

a number of wireless access points that are accessible to multiple devices of the respective subset,

a difference in signal strengths of a wireless access point that is accessible to multiple devices of the respective subset,

a number of wireless access points that are not accessible to multiple devices of the respective subset,

a percentage of a total signal strength of a wireless access point detected by each of multiple devices of the respective subset,

a percentage of wireless access points accessible to multiple devices of the respective subset, relative to wireless access points accessible to at least one device of the respective subset, and

an indication if a wireless access point provides a highest signal strength to all devices of the respective subset of devices.

10. The non-transitory computer readable medium of claim 7, further including instructions to cause the processor to:

compute the set of statistics for pairs of devices of the first group.

11. The non-transitory computer readable medium of claim 7, further including instructions to cause a processor of a device of a third group of devices to exchange scan data with another device of the third group, compute the set of statistics for the third group of devices based on the exchanged scan data, and classify the third group of devices in one of the proximity classes based on the set of statistics of the third group and the model.

12. The non-transitory computer readable medium of claim 7, wherein the proximity classes correspond to proximities of the respective devices relative to one another.

13. A machine-implemented method, comprising:

receiving a first set of scan data that includes information regarding wireless access points available to each device of the first group;

computing a set of statistics for each of multiple subsets of devices of the first group based on the first set of scan data;

classifying each of the subsets of devices in one of multiple proximity classes;

training a model to classify the subsets of devices in the respective proximity classes based on the respective sets of statistics;

receiving a second set of scan data that includes information regarding wireless access points available to each device of a second group;

computing the set of statistics for the second group of devices based on the second set of scan data; and

classifying the second group of devices in one of the proximity classes based on the set of statistics of the second group and the model.

14. The machine-implemented method of claim 13, wherein the first set of scan data includes indications of signal strengths of the wireless access points accessible to the respective devices of the first group, and wherein the computing a set of statistics includes:

computing the set of statistics for each subset of devices of the first group based on the wireless access points accessible to the devices of the respective subset and the respective signal strengths.

15. The machine-implemented method of claim 13, wherein the computing a set of statistics includes compute the set of statistics for each subset of devices of the first group to include one or more of:

a number of wireless access points that are accessible to multiple devices of the respective subset,

a difference in signal strengths of a wireless access point that is accessible to multiple devices of the respective subset,

a number of wireless access points that are not accessible to multiple devices of the respective subset,

a percentage of a total signal strength of a wireless access point detected by each of multiple devices of the respective subset,

a percentage of wireless access points accessible to multiple devices of the respective subset, relative to wireless access points accessible to at least one device of the respective subset, and

an indication if a wireless access point provides a highest signal strength to all devices of the respective subset of devices.

16. The machine-implemented method of claim 13, wherein the computing a set of statistics includes computing the set of statistics for pairs of devices of the first group.

17. The machine-implemented method of claim 13, further including:

performing the receiving a first set of scan data, the computing a set of statistics, the classifying each of the subsets, the training, and the outputting on a server system; and

performing the receiving a second set of scan data, the computing the set of statistics for the second group of devices, and the classifying the second group of devices on a device of the second group;

wherein the receiving a second set of scan data further includes exchanging scan data with the second device.

18. The machine-implemented method of claim 13, wherein the proximity classes correspond to proximities of the respective devices relative to one another.