反馈神经网络卷积码解码器在光纤通信中的研究

Study on Recurrent Neural Network Convolution Decoder in Optical Fiber Communications

反馈神经网络卷积码解码器(RNN)的性能接近传统的Viterbi解码器,RNN的复杂度是随约束长度成线性增加的,而Viterbi的复杂度是成指数增加的,RNN的性能已经在加性白噪声信道中得到了肯定。对RNN解码器在以放大自发发射(ASE)噪声为主的光纤信道中的性能进行了研究,在以ASE噪声为主的3种信道模型中,同时研究了RNN解码器和Viterbi解码器的性能,发现RNN的性能同样很好,在解码过程中还有优势。

Recurrent Neural Networks （RNN） convolution decoder was proposed as an applicable decoder performing close to the conventional Viterbi decoder,and the RNN decoder＇s complexity increases linearly with the constraint length while Viterbi＇s increases exponentially. The performances of RNN decoder were studied on Additive White Gaussian Noise （AWGN） channels. In this paper,we investigate the performance of RNN decoder on intensity modulated optical fiber channels where the Amplified Spontaneous Emission （ASE） noise dominates all other noises. We investigate the performances of RNN and Viterbi decoder on three channel models for the ASE noise dominated channel and find that the RNN decoder also performs very well and is better in some sense.