Class WindowedAccelerationTriadNoiseEstimator

java.lang.Object
com.irurueta.navigation.inertial.calibration.noise.WindowedTriadNoiseEstimator<com.irurueta.units.AccelerationUnit,com.irurueta.units.Acceleration,AccelerationTriad,WindowedAccelerationTriadNoiseEstimator,WindowedAccelerationTriadNoiseEstimatorListener>
com.irurueta.navigation.inertial.calibration.noise.WindowedAccelerationTriadNoiseEstimator
All Implemented Interfaces:
AccelerometerNoiseRootPsdSource

public class WindowedAccelerationTriadNoiseEstimator extends WindowedTriadNoiseEstimator<com.irurueta.units.AccelerationUnit,com.irurueta.units.Acceleration,AccelerationTriad,WindowedAccelerationTriadNoiseEstimator,WindowedAccelerationTriadNoiseEstimatorListener> implements AccelerometerNoiseRootPsdSource
Estimates accelerometer noise variances and PSD's (Power Spectral Densities) along with the accelerometer average values for a windowed amount of samples. This estimator must be used when the body where the accelerometer is attached remains static on the same position with zero velocity while capturing data. To compute PSD's, this estimator assumes that accelerometer samples are obtained at a constant provided rate equal to WindowedTriadNoiseEstimator.getTimeInterval() seconds. If not available, accelerometer sampling rate average can be estimated using TimeIntervalEstimator. This estimator does NOT require the knowledge of current location and body orientation. Because body location and orientation is not known, estimated average values cannot be used to determine biases. Only norm of noise estimations can be (variance or standard deviation) safely used. Notice that if there are less than WindowedTriadNoiseEstimator.getWindowSize() processed samples in the window, this estimator will assume that the remaining ones until the window is completed have zero values. This implementation of noise estimator will use the following units: - meters per squared second (m/s^2) for acceleration, average or standard deviation values. - (m^2/s^4) for acceleration variances. - (m^2 * s^-3) for accelerometer PSD (Power Spectral Density). - (m * s^-1.5) for accelerometer root PSD (Power Spectral Density).