i have a tricky project where i have a sensor on and animal and have a collection of data..
now what i want to do is see whatht e output of the BMA4XX sensor would be ( interupts etc) if the sesnor had detected it itself, like the gesture filters and step counting etc
the reason for doing it this way is that i cannot connect a cable and laptop top the animal. and the data is not repeatable ( the nature of animals)
and no i cannot make a mechanical fixture to replicate the animal movement
The gesture is defined by human hands movement which is defined in Android phone / watch application.
The step counter pattern is also based on human natual walking pattern.
So if it is put on random place of animal, the output will be unpredicible.
What is exactly motion style you want to monitor for this application?
Normally, i will suggest to just take the raw acceleration sensor data then develop your own algorithm.