Recurrent neural equalization for communication channels in impulsive noise environments

In some communication systems, the transmitted signal is contaminated by impulsive noise with a non-Gaussian distribution. Non-Gaussian noise causes significant performance degradation to communication receivers. In this paper, we apply a recurrent neural equalizer to impulsive noise channels, for which the performance of neural network equalizers has never been evaluated. This new application is motivated from the fact that the unscented Kalman filter (UKF), which is suited for training of the recurrent neural equalizer, provides a higher accuracy than the extended Kalman filter (EKF) in capturing the statistical characteristics for non-Gaussian random variables. The performance of the recurrent neural equalizer is evaluated for impulsive noise channels through Monte Carlo simulations. The results support the superiority of the UKF to the EKF in compensating the effect of non-Gaussian impulsive noise.

[1]  Gonzalo R. Arce,et al.  Robust techniques for wireless communications in non-gaussian environments , 1997 .

[2]  Hugh F. Durrant-Whyte,et al.  A new method for the nonlinear transformation of means and covariances in filters and estimators , 2000, IEEE Trans. Autom. Control..

[3]  Bernard Mulgrew,et al.  Adaptive Bayesian decision feedback equaliser for alpha-stable noise environments , 2001, Signal Process..

[4]  David Middleton,et al.  Non-Gaussian Noise Models in Signal Processing for Telecommunications: New Methods and Results for Class A and Class B Noise Models , 1999, IEEE Trans. Inf. Theory.

[5]  C. L. Nikias,et al.  Signal processing with fractional lower order moments: stable processes and their applications , 1993, Proc. IEEE.

[6]  Bayan S. Sharif,et al.  Nonlinear decorrelator for multiuser detection in non-Gaussian impulsive environments , 2000 .

[7]  Chrysostomos L. Nikias,et al.  Performance of optimum and suboptimum receivers in the presence of impulsive noise modeled as an alpha-stable process , 1995, IEEE Trans. Commun..

[8]  Seokho Yoon,et al.  Code acquisition for DS/SS communications in non-Gaussian impulsive channels , 2004, IEEE Transactions on Communications.

[9]  Simon Haykin,et al.  Unscented Kalman filter-trained recurrent neural equalizer for time-varying channels , 2003, IEEE International Conference on Communications, 2003. ICC '03..

[10]  Azzedine Zerguine,et al.  Multilayer perceptron-based DFE with lattice structure , 2001, IEEE Trans. Neural Networks.

[11]  Rudolph van der Merwe,et al.  The Unscented Kalman Filter , 2002 .

[12]  Bayan S. Sharif,et al.  Robust decorrelating decision-feedback multiuser detection in non-Gaussian channels , 2001, Signal Process..

[13]  Daesik Hong,et al.  A decision feedback recurrent neural equalizer as an infinite impulse response filter , 1997, IEEE Trans. Signal Process..

[14]  Tet Hin Yeap,et al.  Online State–Space Modeling Using Recurrent Multilayer Perceptrons with Unscented Kalman Filter , 2005, Neural Processing Letters.

[15]  Tet Hin Yeap,et al.  Decision feedback recurrent neural equalization with fast convergence rate , 2005, IEEE Transactions on Neural Networks.

[16]  Chrysostomos L. Nikias,et al.  Parameter estimation and blind channel identification in impulsive signal environments , 1995, IEEE Trans. Signal Process..

[17]  Sheng Chen,et al.  Adaptive Bayesian equalizer with decision feedback , 1993, IEEE Trans. Signal Process..

[18]  Simon Haykin,et al.  Kalman filter-trained recurrent neural equalizers for time-varying channels , 2005, IEEE Transactions on Communications.