Initializing the Kalman Filter for Nonstationary State Space Models
نویسنده
چکیده
The Kalman filter cannot be used with nonstationary state space models. To circumvent this difficulty, a conditional state space model and a new algorithm, calIed the conditional Kalman filter, can be used. The conditional state space model is obtained by first selecting adequately that part oof the initial state vector which has an unspecified distribution and then conditioning on O. Using the conditional Kalman filter with an initial stretch of the data, an estimator o of ois obtained that allows the conditional Kalman filter to be colIapsed to the ordinary Kalman filter. Considerable simplification can be achieved by careful selection of the nonstationary part O. An application is made to the ARIMA (p,d,q) model.
منابع مشابه
An Effective Attack-Resilient Kalman Filter-Based Approach for Dynamic State Estimation of Synchronous Machine
Kalman filtering has been widely considered for dynamic state estimation in smart grids. Despite its unique merits, the Kalman Filter (KF)-based dynamic state estimation can be undesirably influenced by cyber adversarial attacks that can potentially be launched against the communication links in the Cyber-Physical System (CPS). To enhance the security of KF-based state estimation, in this paper...
متن کاملResidual noise compensation for robust speech recognition in nonstationary noise
We present a model-based noise compensation algorithm for robust speech recognition in nonstationary noisy environments. The effect of noise is split into a stationary part, compensated by parallel model combination, and a time varying residual. The evolution of residual noise parameters is represented by a set of state space models. The state space models are updated by Kalman prediction and t...
متن کاملRotated Unscented Kalman Filter for Two State Nonlinear Systems
In the several past years, Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF) havebecame basic algorithm for state-variables and parameters estimation of discrete nonlinear systems.The UKF has consistently outperformed for estimation. Sometimes least estimation error doesn't yieldwith UKF for the most nonlinear systems. In this paper, we use a new approach for a two variablestate no...
متن کاملA KNN Based Kalman Filter Gaussian Process Regression
The standard Gaussian process (GP) regression is often intractable when a data set is large or spatially nonstationary. In this paper, we address these challenging data properties by designing a novel K nearest neighbor based Kalman filter Gaussian process (KNN-KFGP) regression. Based on a state space model established by the KNN driven data grouping, our KNN-KFGP recursively filters out the la...
متن کامل