Patent · US Active

Maintaining a trained neural network in magnetic fingerprint based indoor navigation

US11402233B2 · kind B2 · utility

0Cited by
4References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJul 23, 2019
Grant dateAug 2, 2022
Priority date
Expiry dateMay 3, 2040

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04W4/33
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method and system of maintaining a trained neural network for mobile device indoor navigation and positioning. The method comprises receiving a set of magnetic measured parameters acquired at a mobile device positioned at a first location of an indoor area; computing, at an output layer of a trained neural network, an output error based on comparing a magnetic input feature in accordance with the magnetic measured parameters to a magnetic output feature generated at the output layer, the magnetic output feature being generated at least in part based on a matrix of weights associated with at least a first neural network layer; and if the output error exceeds a threshold value, re-training the neural network based at least in part upon re-initializing the matrix of weights associated with the at least a first neural network layer.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.