The paper presents theoretical considerations and simulation analysis of some modifications applied to the linear blind equalization procedure. The proposed idea of the equalizer order estimation together with modified equalizer structures improves their working flexibility, understood as initial automatic adjustment of the equalizer parameters in accordance with the unknown channel conditions. The results of the optimal whitening filtering performed prior to equalization procedure are the entry point to develop proposed solutions. Numerical experiments confirm that for the determined equalizer parameters, the assumed distortion level in the receiver output can be achieved. The analysis was done under the baud-spaced equalizer and single input single output (SISO) communication scheme.
[1]
J.E. Mazo,et al.
Digital communications
,
1985,
Proceedings of the IEEE.
[3]
S. Haykin,et al.
Adaptive Filter Theory
,
1986
.
[4]
Lang Tong,et al.
Relationships Between the Constant Modulus and Wiener Receivers
,
1998,
IEEE Trans. Inf. Theory.
[5]
Christophe Laot,et al.
Adaptive decision feedback equalization: can you skip the training period?
,
1998,
IEEE Trans. Commun..
[6]
Fernando Pérez-González,et al.
Efficient reinitialization of the prewhitened constant modulus algorithm
,
2001,
IEEE Communications Letters.