I have some simulated data from a model of an IMU's gyroscopes. I am trying to use Allan deviation to verify that we have correctly modelled various sources of error in the sensor data. I was able to verify angular random walk, and have now moved on to bias instability.
I'm following a tutorial on mathworks which describes finding a critical point in the Allan deviation curve, depicted as such:
However, my sensors do not output this single, easily spotted critical point. My simulated data looks like so:
Is this normal? If so, are there any techniques to choose the best critical point to get an instability reading? If not, what might cause it?
