ADAPTIVE FILTERING METHOD Russian patent published in 2022 - IPC H03H21/00 H03H19/00 

FIELD: adaptive systems.

SUBSTANCE: invention relates to the field of adaptive systems. To achieve the effect, a method for adaptive filtering is proposed, in which the Kalman filtering of the state vector of the observed object is carried out in the presence of current discrete noisy measurements. Wherein the adaptive determination of the filter gain is carried out in the process of current estimation of stochastic signals and state parameters of stochastic systems based on accurate measurements coming at irregular (or random) times by using the transition matrix of the system state, the measurement matrix, the covariance matrix, the transposed transition matrix of the system state , the transposed measurement matrix, and the inverse dispersion matrix of measurement noise, after which the Kalman filtering of the state vector of the observed object is carried out according to the current discrete noisy measurements with the newly formed matrix of filter adaptation coefficients until the next accurate measurements arrive, after which the procedure is repeated.

EFFECT: ensuring stability and improving the accuracy of Kalman filtering.

1 cl