Robust Linear Filtering for Multivariable Stationary Time Series. Part 3.
CONNECTICUT UNIV STORRS DEPT OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE
Pagination or Media Count:
The problem of asymptotic non-causal linear filtering for statistically contaminated multivariable second order stationary time series is considered. It is assumed that the spectra of both the signal and the noise components of the observation process are uncertain. This uncertainty is modeled by requiring that the spectra belong to two well defined spectral classes. Subsequently, a game theoretic formalization is adopted, and for some specific spectral classes saddle point solutions are found and analyzed. Author
- Statistics and Probability