一种基于LSTM的OFDM-VLC后失真补偿方法  被引量:7

Post distortion compensation of OFDM-VLC based on LSTM

在线阅读下载全文

作  者:许俊翔 林邦姜[3] 汤璇[1] 申晓欢 赖奇伟 XU Junxiang;LIN Bangjiang;TANG Xuan;SHEN Xiaohuan;LAI Qiwei(Quanzhou Institute of Equipment Manufacturing,Chinese Academy of Sciences,Quanzhou Fujian 362000,China;School of Big Data,North University of China,Taiyuan 030051,China)

机构地区:[1]中国科学院海西研究泉州装备制造研究所物联网通信技术实验室,福建泉州362000 [2]中北大学大数据学院,太原030051 [3]中国科学院海西研究泉州装备制造研究所,福建泉州362000

出  处:《光通信技术》2020年第1期54-57,共4页Optical Communication Technology

基  金:中国科学院科研仪器设备研制项目(YJKYYQ20170052)资助;福建省科技计划对外合作项目(2017I01010012)资助;国家自然科学基金(61601439)资助。

摘  要:非线性是限制可见光通信(VLC)系统传输性能的一个重要因素。正交频分复用(OFDM)技术虽然可以有效提升VLC的频谱效率,但其较高的峰均值比使得系统受到更多非线性损伤的影响。提出了一种基于长短期记忆网络(LSTM)的后失真补偿方法,建立局部信号的联系,补偿VLC系统中的线性与非线性传输损伤。仿真表明:视距传输下,相比于基于线性递推最小二乘法、基于沃尔泰拉级数和基于记忆多项式的后失真补偿方法,基于LSTM的后失真补偿方法改善了信号的前向误差幅度性能,在达到相同误码率条件下提高了信号的偏置电压和动态范围,降低了约4 dB的信噪比需求。Nonlinearity is an important factor limiting the transmission performance of visible light communication(VLC)systems.Although orthogonal frequency division multiplexing(OFDM)technology can effectively improve the spectral efficiency of VLC,it also makes systems more sensible to nonlinearity due to its high peak-to-average power ratio.This paper proposes a post distortion method based on long short term memory(LSTM)network,which takes the local correlation of signals into account and can effectively compensate linear and nonlinear transmission impairments in VLC systems.In the case of line-of-sight transmission,compared with the post-distortion based on linear recursive least square,simulation results show that Volterra series and memory polynomials,the post distortion method based on LSTM not only can improve the performance of error vector magnitude,but also can increase signal bias voltage and dynamic range and decrease the demand for signal and noise ratio by about 4 dB with the same bit error rate.

关 键 词:可见光通信 非线性补偿 长短期记忆网络 

分 类 号:TN929.12[电子电信—通信与信息系统]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象