Multistage Estimation of Dynamical and Colored Noise States in Continuous-Time Linear Systems.
UNIVERSITY OF SOUTHERN CALIFORNIA LOS ANGELES DEPT OF ELECTRICAL ENGINEERING
Pagination or Media Count:
This paper extends Friedlands bias estimation technique to the problem of estimating dynamical states and colored noise states. An exact multistage decomposition is derived not only for the state estimation equations, but also for the associated error covariance equations. Additionally, a second-order suboptimal multistage estimator is derived using a perturbation technique. Whereas a high-order matrix Riccati equation must be solved when the exact results are used, a matrix Riccati equation, of the dimension of the colored noise states, must be solved when the suboptimal results are used.
- Statistics and Probability