Abstract | ||
---|---|---|
Recurrent neural networks have been successfully applied to communications channel equalization because of their capability of modelling nonlinear dynamic systems. The major problems of gradient descent learning techniques, commonly employed to train recurrent neural networks, are slow convergence rates and long training sequences. This paper presents a decision feedback equalizer using a recurrent neural network trained with unscented Kalman filter (UKF). The main features of the proposed recurrent neural equalizer are fast convergence and good performance using relatively short training symbols. Experimental results for time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer. |
Year | DOI | Venue |
---|---|---|
2003 | 10.1109/ICC.2003.1204038 | ICC |
Keywords | Field | DocType |
unscented kalman filter,recurrent neural network,gradient descent,communication channels,convergence rate | Convergence (routing),Gradient descent,Equalization (audio),Control theory,Fading,Computer science,Recurrent neural network,Kalman filter,Time delay neural network,Artificial neural network | Conference |
Volume | ISBN | Citations |
5 | 0-7803-7802-4 | 3 |
PageRank | References | Authors |
0.57 | 8 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jong-Soo Choi | 1 | 147 | 30.10 |
Antonio Cezar de Castro Lima | 2 | 30 | 1.77 |
Simon Haykin | 3 | 5558 | 711.64 |