An alternative approach to the linear causal least-square filtering theory
IEEE Transactions on Information Theory
Robust detection of known signals
IEEE Transactions on Information Theory
Robust sequential detection of signals in noise
IEEE Transactions on Information Theory
Robust hypothesis testing for bounded classes of probability densities (Corresp.)
IEEE Transactions on Information Theory
Hi-index | 22.14 |
A representative selection of results from an extensive study of the performance of Wiener filtering under spectral uncertainty is presented. For a variety of spectral uncertainty models the Wiener filter is shown to have undesirable sensitivity to even small deviations from those signal and noise spectral densities which were used to design the filter. Performance of a robust filter (designed to have the best possible performance when the uncertainty is worst) is also examined. In most cases the robust filter's insensitivity to spectral uncertainty makes it preferable to the traditional Wiener filter.