Abstract
Many contrast-based independent component analysis (ICA) algorithms can be directly applied to single-channel blind deconvolution tasks without modification. For prewhitened signals, however, an allpass constraint on the filter impulse response is more appropriate than a unit-norm constraint. In this paper, we show how a finite-impulse-response filter can be adapted to maximize or minimize an arbitrary contrast function while approximately maintaining an allpass characteristic in a self-stabilizing fashion. Both on-line time domain and block-based frequency domain algorithms are described. Simulations verify the superior performance behaviors of the new algorithms over those of standard ICA algorithms in prewhitened blind deconvolution tasks.
Original language | English (US) |
---|---|
Pages | 273-282 |
Number of pages | 10 |
State | Published - 1999 |
Externally published | Yes |
Event | Proceedings of the 1999 9th IEEE Workshop on Neural Networks for Signal Processing (NNSP'99) - Madison, WI, USA Duration: Aug 23 1999 → Aug 25 1999 |
Other
Other | Proceedings of the 1999 9th IEEE Workshop on Neural Networks for Signal Processing (NNSP'99) |
---|---|
City | Madison, WI, USA |
Period | 8/23/99 → 8/25/99 |
All Science Journal Classification (ASJC) codes
- Signal Processing
- Software
- Electrical and Electronic Engineering