Invention Grant
- Patent Title: Methods and apparatus for distributed learning of parameters of a fingerprint prediction map model
- Patent Title (中): 指纹预测图模型参数分布式学习的方法与装置
-
Application No.: US13009779Application Date: 2011-01-19
-
Publication No.: US08692667B2Publication Date: 2014-04-08
- Inventor: Thomas Richardson , Aleksandar Jovicic , Cyril Measson
- Applicant: Thomas Richardson , Aleksandar Jovicic , Cyril Measson
- Applicant Address: US CA San Diego
- Assignee: QUALCOMM Incorporated
- Current Assignee: QUALCOMM Incorporated
- Current Assignee Address: US CA San Diego
- Agent James K. O'Hare
- Main IPC: G08B1/08
- IPC: G08B1/08 ; H04B17/00 ; G01S3/02

Abstract:
In accordance with various embodiments, mobile wireless devices which use prediction maps, e.g., RF prediction maps, to determine their location also contribute to the updating and distribution of parameters relating to prediction map generation. After a device determines its location, it uses the signal measurements used in determining the device's location to determine updated parameters which can be used in generating signal prediction maps in the future. The generated parameters are then transmitted to another device for use in generating prediction maps and/or distribution to other devices. In this manner parameters corresponding to a particular location can be updated and distributed without the need for a survey team and/or dedicated equipment. The devices using the prediction map become the devices which also aid in the updating of the parameters used to facilitate generation of a prediction map at some future time. In some embodiments, statistical information corresponding to the updated parameters is also generated and communicated.
Public/Granted literature
- US20120182144A1 METHODS AND APPARATUS FOR DISTRIBUTED LEARNING OF PARAMETERS OF A FINGERPRINT PREDICTION MAP MODEL Public/Granted day:2012-07-19
Information query