Title :
Channel equalization using complex-valued recurrent neural network
Author :
Wang, Xiaoqiu ; Lin, Hua ; Lu, Jianming ; Yahagi, Takashi
Author_Institution :
Yahagi & Lu Lab., Chiba Univ., Japan
Abstract :
Recurrent neural network (RNN) is a kind of neural network with one or more feedback loops. In this paper, a complex-valued fully connected RNN with real-time recurrent learning is presented for the equalization of complex-valued systems, such as quadrature amplitude modulation (QAM), in the presence of intersymbol interference and nonlinear distortions. Simulation results show that the proposed scheme is quite effective in channel equalization when facing the nonlinear distortions
Keywords :
equalisers; feedback; intersymbol interference; nonlinear distortion; quadrature amplitude modulation; recurrent neural nets; QAM; channel equalization; complex-valued recurrent neural network; feedback loops; fully connected RNN; intersymbol interference; nonlinear distortions; quadrature amplitude modulation; Delay; Feedback loop; Multi-layer neural network; Neural networks; Neurofeedback; Neurons; Nonlinear distortion; Pattern recognition; Quadrature amplitude modulation; Recurrent neural networks;
Conference_Titel :
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on
Conference_Location :
Beijing
Print_ISBN :
0-7803-7010-4
DOI :
10.1109/ICII.2001.983106
