FIR filter representations of reduced-rank noise reduction
- 1 June 1998
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 46 (6), 1737-1741
- https://doi.org/10.1109/78.678511
Abstract
We show that the reduced-rank output signal computed via truncated (Q)SVD is identical to that from an array of parallelly connected analysis-synthesis finite impulse response (FIR) filter pairs. The filter coefficients are determined by the (Q)SVD, and the filters provide an explicit description of the reduced-rank noise reduction algorithm in the frequency domainKeywords
This publication has 11 references indexed in Scilit:
- Rank-Deficient and Discrete Ill-Posed ProblemsPublished by Society for Industrial & Applied Mathematics (SIAM) ,1998
- Data adaptive rank-shaping methods for solving least squares problemsIEEE Transactions on Signal Processing, 1995
- Reduction of broad-band noise in speech by truncated QSVDIEEE Transactions on Speech and Audio Processing, 1995
- Enhanced resolution based on minimum variance estimation and exponential data modelingSignal Processing, 1993
- The SVD and reduced rank signal processingSignal Processing, 1991
- Physical interpretation of signal reconstruction from reduced rank matricesIEEE Transactions on Signal Processing, 1991
- Speech enhancement from noise: A regenerative approachSpeech Communication, 1991
- Rank reduction for modeling stationary signalsIEEE Transactions on Acoustics, Speech, and Signal Processing, 1987
- Data adaptive signal estimation by singular value decomposition of a data matrixProceedings of the IEEE, 1982
- Estimation of frequencies of multiple sinusoids: Making linear prediction perform like maximum likelihoodProceedings of the IEEE, 1982