Particle detection and velocity measurement in laser Doppler velocimetry using Kalman filters

The paper discusses a processing technique for LDV data, based on the use of two Kalman filters, enabling the presence of particles to be detected and their velocity to be inferred. This method turns out to be suitable for the design of real-time integrated velocimeters. A first estimator, based on the use of a Kalman filter, deals with the amplitude of the Doppler signal. A second one, using an extended Kalman filter, allows particle velocity estimation, which is assumed to be a constant. Finally, the estimator is studied by means of Monte Carlo trials obtained from synthesized signals, and its performance is then compared to the Cramer-Rao bound.