Locations filtering

[giohappy]

Location recordings are affected by noise and artefacts due to GPS signal reflections, scattering, etc.
Literature reports erros on the total distance of tracks even of 400%. This can sariously affect any downstream analysis and calculations (like SMB indexes).

Beside backend data processing I propose to include some filtering at the source, i.e. during recording on the app. A widely adopted approach is Kalman filters, which can integrate data from sensors (accelerometer, gyroscope, magnetometer, etc.).
Here it is an implementation, dedicated to GPS filtering, which can be adopted or taken in consideration for SMB app:

https://blog.maddevs.io/reduce-gps-data-error-on-android-with-kalman-filter-and-accelerometer-43594faed19c

The github repo contains the core library and a sample application using the library