Optimal blind equalization of Gaussian channels
In this paper we show that a recurrent version of a Radial Basis Functions network can compute optimal symbol-by-symbol decissions for equalizing Gaussian channels in digital communication systems, when the (linear or not) channel response and the noise variance are known. In order to do this, the recurrent RBF (RRBF) computes several statistics about the possible states of the channel, which can be used to estimate the channel response during operation. Taking advantage of this fact, a novel technique for learning the channel parameters in a non-supervised way is theoretically derived, resulting a simple and fast algorithm that can be used for tracking in time variant environments, or for blind equalization purposes.
Finally, we show several simulation results which support the theoretically derived results.
Unable to display preview. Download preview PDF.
- S. Chen et al, “Adaptive Equalization of Finite Non-linear Channels Using Multilayer Perceptrons”, Signal Processing, vol. 20, no. 2, pp. 107–119; 1990.Google Scholar
- S. Chen et al, “Reconstruction of Binary Signals Using an Adaptive Radial Basis Function Equalizer”; Signal Processing, vol. 22, no. 2, pp. 77–93; 1991.Google Scholar
- G. J. Gibson et al, “The Application of Nonlinear Structures to the Reconstruction of Binary sSignals”, IEEETrans. Signal Processing, vol. 39, no. 8, pp. 1887–1884; 1991.Google Scholar
- S. Arcens, J. Cid-Sueiro, A. R. Figueiras-Vidal, “Pao Networks for Data Transmission Equalization”; Proc. of the International Joint Conference on Neural Networks, vol. 2, pp. 963–967; Baltimore, MA, Jun. 1992.Google Scholar
- T. Kohonen, K. Raivio, O. Simula, O. Venta, J. Henriksson, “Combining Linear Equalization and Self-Organizing Adaptation in Dinamic Discrete-Signal Detection”, Proc. of the International Joint Conference on Neural Networks, vol. 1, pp. 223–228; San Diego, CA, Jun 1990.Google Scholar
- J. Cid-Sueiro, A. R. Figueiras-Vidal, “Optimal Symbol-by Symbol and Sequential Nonlinear Channel Equalizers Using Neural Network Structures”; First Cost #229 WG. 4 Workshop; Leysin (Switzerland), March 1992.Google Scholar
- S. Haykin: Adaptive Filter Theory (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall; 1991.Google Scholar
- L. Weruaga-Prieto, J. Cid-Sueiro, A. R. Figueiras-Vidal, “Optimal Variable-Step LMS Look-Up-Table Plus Transversal Filter Nonlinear Echo Cancellers”; Proc. of the IEEE ICASSPP'92; vol. 4, pp. 229–232; San Francisco (CA), March 1992.Google Scholar
- L. Weruaga-Prieto, J. Cid-Sueiro, A. R. Figueiras-Vidal, “A Fast RLS Look-Up-Table Plus FIR Adaptive Echo Cancellers”; Proc. of the European Signal Processing Conference (EUSIPCO'92), vol. 3, pp. 1615–1618; Brussels (Belgium), Aug. 1992.Google Scholar
- J. Cid-Sueiro, A. R. Figueiras-Vidal, “Igualación no Lineal Optima para Comunicaciones Digitales por Satélite”, Actas del II Congreso INTA, pp. 485–490; Madrid (Spain), Oct. 1992.Google Scholar
- S. Chen, B. Mulgrew, S. Mc Laughlin, P.M. Grant, “Adaptive Bayesian Equaliser with Feedback for Mobile Radio Channels”; 2nd Cost #229 WG. 4 Workshop on Adaptive Algorithms in Communications; Bordeaux (France), Oct. 1992.Google Scholar
- J. F. Hayes, T.M. Cover, J.B. Riera, “Optimal Sequence Detection and Optimal Symbol-by Symbol Detection: Similar Algorithms”, IEEE Trans. on Communications, vol. COM-30, pp. 152–157, Jan. 1982.Google Scholar