Optimal and Suboptimal FIR Estimation of Linear and Nonlinear Models

 

Address: Stillwell Hall 103 (Compton Lounge)
Fenn College of Engineering, Cleveland State University
1960 East 24th Street • Cleveland, OH 44115
Refreshment and soft drink will be provided!
RSVP: Dr. Lili Dong • L.Dong34@csuohio • 216-687-5312
CPD Credit: 1 CPD hour is available.

 

Guest Speaker----- Professor Yuriy S. Shmaliy
Department of Electronics
DICIS, University of Guanajuato,
Salamanca, 36855, Mexico

Abstract: Although the optimal Kalman filter has become a part of estimation theory and optimal control, its practical implementation is often problematic due to commonly-unknown noise statistics. In this talk, we show that finite impulse response (FIR) filtering is able to circumvent this disadvantage in the p-shift iterative unbiased FIR (UFIR) estimation algorithm, ignoring noise and initial errors. The latter has a unified scheme for filtering (p=0), smoothing (p<0), and prediction (p>0). Its other advantages against the Kalman filter are: bounded input/bounded output (BIBO) stability, better robustness against temporary model uncertainties and round-off errors, and lower sensitivity to noise and initial conditions. We also show that errors in FIR estimators are well-bounded in the three-sigma sense via the noise power gain. The Kalman and FIR algorithms have been applied to linear and nonlinear models and compared on typical examples of state estimation, tracking, GPS-based timekeeping, model prediction, and denoising of signals and images.

Address: Stillwell Hall 103 (Compton Lounge)
Fenn College of Engineering, Cleveland State University
1960 East 24th Street • Cleveland, OH 44115

Refreshment and soft drink will be provided!

RSVP: Dr. Lili Dong • L.Dong34@csuohio • 216-687-5312

CPD Credit: 1 CPD hour is available.

Flyer Available Here

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer