Notes on: Nowicki, M., & Skrzypczyński, P. (2015): Indoor navigation with a smartphone fusing inertial and wifi data via factor graph optimization

View on DOI

@article{Nowicki_factor_15,
  author    = {Nowicki, Michał and Skrzypczyński, Piotr},
  title     = {Indoor Navigation with a Smartphone Fusing Inertial and WiFi
               Data via Factor Graph Optimization},
  journal   = {Mobile Computing, Applications, and Services},
  pages     = {280–298},
  year      = {2015},
  doi       = {10.1007/978-3-319-29003-4_16},
  url       = {http://dx.doi.org/10.1007/978-3-319-29003-4_16},
  isbn      = {9783319290034},
  issn      = {1867-822X},
  publisher = {Springer International Publishing},
}

3.3 WiFi Fingerprinting

It is possible to:

  1. compare the current WiFi scan to a pre-existing database of fingerprints taken at known locations,

  2. compare the current WiFi scan to the WiFi fingerprints recorded earlier during the system operation.

The former method assumes that before the start of a navigation task a small set of WiFi scans is taken in known poses of the user, and these fingerprints are stored in the memory of the smartphone. The procedure to obtain these WiFi fingerprints is straightforward and fast, as only WiFi scans in few selected locations are needed. The latter operation mode assumes no prior knowledge of the environment as only the WiFi scans taken during the system operation are compared. The matching of WiFi fingerprints provides localization constraints equivalent to the loop closure mechanism in SLAM, as it detects previously visited locations (either known a priori or discovered) and allows to reduce the dead reckoning drift.