摘要
针对光纤通信系统在传输过程中易受光纤线路、传输设备等因素干扰的问题,提出了一种利用基于卷积神经网络的编码器-解码器来降低接收端信号误码率的方法,从而提升光纤通信系统的抗干扰能力。具体地,利用伪随机二进制序列(Pseudo-Random Binary Sequence)码作为标签信号,将受干扰后的信号作为输入信号,训练出具有恢复受扰信号能力的卷积编码器-解码器。实验结果表明,卷积编码器-解码器能降低受一定程度干扰的信号误码率,具备提升光纤通信系统抗干扰性能的能力。
Aiming at the problem that the optical fiber communication system is easy to be interfered by the factors such as optical fiber line and transmission equipment in the transmission process,a method is proposed to reduce the BER(Bit Error Rate)of the received signal by using the encoder-decoder based on CNN(Convolutional Neural Network),so as to improve the anti-interference ability of the optical fiber communication system.Specifically,PRBS(Pseudo-Random Binary Sequence)code is used as the label signal and the disturbed signal as the input signal to train the CNN encoder-decoder with the ability to recover the disturbed signal.The experimental results indicate that the CNN encoder-decoder can reduce the BER of the signal disturbed to some extent,and thus has the ability to improve the anti-interference performance of the optical fiber communication system.
作者
别芳宇
张建国
高勇
BIE Fang-yu;ZHANG Jian-guo;GAO Yong(College of Electronics and Information Engineering,Sichuan University,Chengdu Sichuan 610065,China;Chengdu Superxon Communication Technology Co.,Ltd.,Chengdu Sichuan 610041,China)
出处
《通信技术》
2020年第3期551-556,共6页
Communications Technology
关键词
光纤通信系统
卷积神经网络
编码器-解码器
抗干扰
误码率
optical fiber communication system
CNN(convolutional neural network)
encoder-decoder
anti-interference
BER(bit error rate)
