Title :
A complex version of a decision feedback recurrent neural equalizer as an infinite impulse response filter
Author :
Ong, Sunghwan ; Choi, Sooyong ; You, Cheolwoo ; Hong, Daesik
Author_Institution :
Dept. of Electron. Eng., Yonsei Univ., Seoul, South Korea
Abstract :
A decision feedback recurrent neural equalizer (DFRNE), a nonlinear infinite impulse response (IIR) filter, has already been proposed successfully for the channel equalization problem (Ong et al. 1997). In this paper, we introduce a complex version of a DFRNE. Its small size and high performance makes it attractive for channel equalization. We evaluate its performance through computer simulations for various complex channels and derive that its convergence is more stable and robust than that of other equalizers. It is also shown that the complex DFRNE outperforms other equalization schemes when the channel distortion is profound
Keywords :
IIR filters; convergence of numerical methods; decision feedback equalisers; nonlinear filters; recurrent neural nets; telecommunication channels; telecommunication computing; DFRNE; channel distortion; channel equalization; complex version; convergence; decision feedback recurrent neural equalizer; infinite impulse response filter; nonlinear infinite impulse response filter; performance; size; Computer simulation; Decision feedback equalizers; Ear; Finite impulse response filter; IIR filters; Neural networks; Neurofeedback; Nonlinear distortion; Recurrent neural networks; Robustness;
Conference_Titel :
Global Telecommunications Conference, 1997. GLOBECOM '97., IEEE
Conference_Location :
Phoenix, AZ
Print_ISBN :
0-7803-4198-8
DOI :
10.1109/GLOCOM.1997.632512
