Title :
Restoration of nonlinearly distorted magnetic recordings
Author :
Preis, U. ; Polchlopek, H.
Author_Institution :
Tufts University, Medford, Massachusetts, USA
Abstract :
An efficient method to restore nonlinearly distorted magnetic recordings is presented. Given the distorted recording and the actual recording device or a model of the record/reproduce process, the original input signal can be computed iteratively. Results from several computer simulations using a variety of nonlinear and linear distortions with different test signals are included. The algorithm converged with fewer than 40 iterations in all simulated restorations.
Keywords :
Convergence; Cutoff frequency; Iterative algorithms; Magnetic recording; Mathematical model; Maximum likelihood detection; Nonlinear distortion; Nonlinear filters; Saturation magnetization; Stability;
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '83.
DOI :
10.1109/ICASSP.1983.1172014
