人工智能的紫外光通信调制方式研究  

Research on modulation mode of ultraviolet communication based on artificial intelligence

作　　者：王奇[1] 宋思思 刘毅娟[1] WANG Qi;SONG Sisi;LIU Yijuan(Jitang College of North China University of Science and Technology,Tangshan Hebei 063000,China;North China University of Science and Technology,Tangshan Hebei 063000,China)

机构地区：[1]华北理工大学冀唐学院,河北唐山063000 [2]华北理工大学,河北唐山063000

出　　处：《激光杂志》2022年第11期118-123,共6页Laser Journal

基　　金：河北省科技厅项目(No.18211844)。

摘　　要：为了提高紫外光通信系统传输可靠性,提出一种考虑人工智能的紫外光通信调制方式。信号经FPGA芯片的编码纠错、加密加扰、帧头插入处理后,传输至分离双脉冲位置调制器,经串/并转换后保存于存储器中,利用离散小波变换方法分解紫外光信号,去除紫外光信号中的噪声,利用稠密网络、长短期记忆网络构建紫外光通信调制识别模型,实现紫外光通信信号的调制识别,输出的紫外光调制信号经滤光片、紫外探测器以及信号预处理后,采用SDPPM方式解调,结合解密、干扰解除操作,获取紫外光的原始语音信号。实验结果表明:本方法可以消除紫外光通信信号的工频干扰噪声,获取高质量的紫外光信号,在带宽、功率、传输容量、平均符号长度等方面具有显著优势。In order to improve the transmission reliability of UV communication system, a UV communication modulation method considering artificial intelligence is proposed. After the signal is processed by encoding and error correction, encryption and scrambling and frame header insertion of FPGA chip, it is transmitted to the separated dual pulse position modulator, which is stored in the memory after serial/parallel conversion. The discrete wavelet transform method is used to decompose the UV signal to remove the noise in the UV signal, and the dense network and long-term and short-term memory network are used to construct the UV communication modulation recognition model, The modulation recognition of UV communication signal is realized. After the output UV modulation signal is processed by filter, UV detector and signal preprocessing, it is demodulated by sdppm, combined with decryption and interference cancellation to obtain the original voice signal of UV. The experimental results show that this method can eliminate the power frequency interference noise of UV communication signal and obtain high-quality UV signal. It has significant advantages in bandwidth, power, transmission capacity, average symbol length and so on.

关 键 词：人工智能 紫外光通信 调制方式 离散小波变换 调制识别模型 大气信道 

分 类 号：TN97[电子电信—信号与信息处理]