Invention Grant
US09591454B2 Computational complexity reduction of training wireless strength-based probabilistic models from big data
有权
从大数据中训练基于无线强度的概率模型的计算复杂度降低
- Patent Title: Computational complexity reduction of training wireless strength-based probabilistic models from big data
- Patent Title (中): 从大数据中训练基于无线强度的概率模型的计算复杂度降低
-
Application No.: US14843935Application Date: 2015-09-02
-
Publication No.: US09591454B2Publication Date: 2017-03-07
- Inventor: Brian John Julian , Etienne Le Grand , Brian Patrick Williams
- Applicant: Google Inc.
- Applicant Address: US CA Mountain View
- Assignee: Google Inc.
- Current Assignee: Google Inc.
- Current Assignee Address: US CA Mountain View
- Agency: McDonnell Boehnen Hulbert & Berghoff LLP
- Main IPC: G01R31/08
- IPC: G01R31/08 ; H04W4/02 ; H04W24/08 ; H04W64/00

Abstract:
Disclosed are apparatus and methods for providing outputs; e.g., location estimates, based on signal strength measurements. A computing device can receive a particular signal strength measurement, which can include a wireless-signal-emitter (WSE) identifier and a signal strength value and can be associated with a measurement location. The computing device can determine one or more bins; each bin including statistics for WSEs and associated with a bin location. The statistics can include mean and standard deviation values. The computing device can: determine a particular bin whose bin location is associated with the measurement location for the particular signal strength measurement, determine particular statistics of the particular bin associated with a wireless signal emitter identified by the WSE identifier of the particular signal strength measurement, and update the particular statistics based on the signal strength value. The computing device can provide an estimated location output based on the bins.
Public/Granted literature
Information query