Crowdsourcing method to detect broken WiFi indoor locationing model

Patent metadata

Field	Value
Publication number	US-9107045-B2
Application number	US-201313937378-A
Country	US
Kind code	B2
Filing date	Jul 9, 2013
Priority date	Jul 9, 2013
Publication date	Aug 11, 2015
Grant date	Aug 11, 2015

Abstract

Aspects of the present disclosure provide techniques for detecting breaks in a wireless network data model. An exemplary method includes determining neighboring access points from scans of network access points in a space. Each neighboring access point occurs together in a scan of a particular level of the space. Wireless data is received from a plurality of mobile devices moving through a space. A set of all access points for the space is identified based on the wireless data. A ratio is derived based on a difference between the neighboring access points and the set of all access points. The ratio represents a percentage of missing access points for the particular level of the space.

First claim

The invention claimed is:

1. A method, comprising:
determining a set of neighboring access points from scans of network access points in a space, wherein each neighboring access point occurs together in a scan of a particular level of the space;
receiving wireless data from a plurality of mobile devices moving through the space;
identifying a set of all access points for the space based on the wireless data; and
deriving, using a processor, a ratio based on the size of a dif…

Assignees

• Google Inc

Inventors

• Zhou Shuchang
• Mantri Anup

Classifications

• G01S5/0242 Primary

  Physics · mapped topic

• H04W64/003

  Electricity · mapped topic

• H04W4/043 Primary

  Electricity · mapped topic

• H04W16/20

  Electricity · mapped topic

• H04W24/02

  Electricity · mapped topic