Title :
Robust time-varying Wiener filters: theory and time-frequency formulation
Author :
Matz, Gerald ; Hlawatsch, Ranz
Author_Institution :
Inst. of Commun. & Radio-Frequency Eng., Tech. Univ. Wien, Austria
Abstract :
We propose a minimax robust time-varying Wiener filter that is based on a novel uncertainty model for nonstationary random processes. This filter maintains a certain performance for all second-order statistics within prescribed uncertainty classes. Furthermore, it requires less detailed prior knowledge than the ordinary Wiener filter. We also present an intuitively appealing time-frequency formulation of the robust time-varying Wiener filter in which signal subspaces are replaced with time-frequency regions
Keywords :
Wiener filters; filtering theory; minimax techniques; random processes; statistical analysis; time-frequency analysis; time-varying filters; minimax robust time-varying Wiener filter; nonstationary random processes; performance; second-order statistics; signal subspaces; time-frequency formulation; time-frequency regions; uncertainty model; Electronic mail; Kernel; Minimax techniques; Random processes; Robustness; Signal processing; Time frequency analysis; Time varying systems; Uncertainty; Wiener filter;
Conference_Titel :
Time-Frequency and Time-Scale Analysis, 1998. Proceedings of the IEEE-SP International Symposium on
Conference_Location :
Pittsburgh, PA
Print_ISBN :
0-7803-5073-1
DOI :
10.1109/TFSA.1998.721446
