Fast frequency tracking using an adaptive lattice filter for a vortex flowmeter signal

Adaptive lattice filters for detecting signals have been thoroughly investigated in the recent years (1), (2), (3), (4), but, to our knowledge, it does not seem that a lot of applications have been designed, especially in very noisy and non-stationary environment. This paper proposes an application of an adaptive lattice filter and the study of its limitations to a problem of continuous flow measurement. The basic idea consists in looking at the compartment of the filter in function of the signal to noise ratio, the bandwidth of the process and finally of the fading memory parameter of the filter, then provide some hints about its choice to improve the rapidity of convergence sometimes at the trade off the precision of the measurement.