We aim to recover a multi-frequency component nonstationary signal from its broadband noise-corrupted measurements using a time-varying optimal Wiener filter. A new method for realizing the Wiener filter is proposed, based on our multiresolution parametric spectral estimator (MPSE). Conventional estimators for contaminated AR processes are all fixed resolution based methods, which are mostly suitable for stationary situations. In nonstationary applications, the estimator must not only locate the signal components in frequency but also in time. MPSE offers better time resolution than conventional fixed resolution parametric estimators. The MPSE frequency band splitting reduces necessary model orders and improves the SNR. The Wiener filter is given in terms of the MPSE parameters. Experiments show that the performance of the MPSE Wiener filter lies much closer to the ideally possible performance than for a Wiener filter based on fixed resolution AR modeling.
[1]
Min Xie,et al.
Optimal wavelet design via multiresolutional parametric spectral estimation
,
1995,
Defense, Security, and Sensing.
[2]
A.V. Oppenheim,et al.
Enhancement and bandwidth compression of noisy speech
,
1979,
Proceedings of the IEEE.
[3]
Charles W. Therrien,et al.
Discrete Random Signals and Statistical Signal Processing
,
1992
.
[4]
Alan V. Oppenheim,et al.
All-pole modeling of degraded speech
,
1978
.
[5]
A.A. Beex,et al.
Multiresolution parametric spectral estimation
,
1994,
Proceedings of 26th Southeastern Symposium on System Theory.
[6]
Zeungnam Bien,et al.
Sliding-windowed weighted recursive least-squares method for parameter estimation
,
1989
.