基于高斯支持向量机的可见光通信解调方法

徐子震; 李建锋; 刘晓爽

doi:10.13756/j.gtxyj.2026.240166

您当前的位置：

首页 >

文章列表页 >

基于高斯支持向量机的可见光通信解调方法

光无线技术 | 更新时间：2026-03-03

- 基于高斯支持向量机的可见光通信解调方法
- Demodulation of Visible Light Communication based on Gaussian Support Vector Machine
- 光通信研究 2026年第1期
- 作者机构：
  
  1.河北经贸大学　管理科学与信息工程学院，石家庄　 050061
  2.天津理工大学　集成电路科学与工程学院，天津　 300384
- 作者简介：
  
  徐子震（1999-），男，河北廊坊人。硕士，主要研究方向为可见光通信技术。
  李建锋，副教授。E-mail：lijianfeng555@126.com
- 基金信息：
  
  河北省“三三三人才工程”资助项目(A202101015);河北省教育厅科学技术研究资助项目(ZD2021044)
- DOI：10.13756/j.gtxyj.2026.240166
  中图分类号： TN929.12
- 收稿：2024-08-05，
  
  修回：2024-09-05，
  
  纸质出版：2026-02-10
- 稿件说明：
移动端阅览
徐子震，李建锋，刘晓爽. 基于高斯支持向量机的可见光通信解调方法[J]. 光通信研究，2026(1): 240166.

Xu Z Z, Li J F, Liu X S. Demodulation of Visible Light Communication based on Gaussian Support Vector Machine[J]. Study on Optical Communications, 2026(1): 240166.
徐子震，李建锋，刘晓爽. 基于高斯支持向量机的可见光通信解调方法[J]. 光通信研究，2026(1): 240166. DOI： 10.13756/j.gtxyj.2026.240166.

Xu Z Z, Li J F, Liu X S. Demodulation of Visible Light Communication based on Gaussian Support Vector Machine[J]. Study on Optical Communications, 2026(1): 240166. DOI： 10.13756/j.gtxyj.2026.240166.

摘要

目的

针对基于正交频分复用（OFDM）的可见光通信（VLC）系统中的非线性效应，文章提出了一种基于高斯支持向量机（Gaussian SVM）的分类解调方法。

方法

文章采用机器学习（ML）中的支持向量机（SVM）并结合高斯核函数对解调前的信号进行非线性映射，为解调过程增添非线性处理能力，增强系统对复杂信道环境的适应性。把接收到的信号视为输入向量，通过Gaussian SVM的映射机制，将这些信号转换到高维特征空间。在高维特征空间内构造一个最优分类超平面，以最大化样本间的间隔，即最大化向量到超平面的总距离，来实现信号的高效分类解调。在发光二极管（LED）非线性信道下，通过不同调制阶数（16正交振幅调制（QAM）、32QAM和64QAM）进行蒙特卡罗仿真和性能分析实验，文章验证了该方法的显著优势。

结果

仿真表明，与传统硬解调的方法相比，Gaussian SVM的分类解调方法在应对非线性信道时，展现出更强的鲁棒性和更高的解调精度。特别在高阶调制下，如64QAM，该方法能有效缓解非线性效应对信号质量的负面影响，得到7 dB的信噪比增益。

结论

文章证明了随着非线性效应的加剧，该方法展现出的信噪比增益也随之增大，充分证明了在高阶调制阶数下，文章所提系统性能提升的巨大潜力。通过动态适应信道变化，该方法不仅提高了系统的整体性能，还显著增强了系统的可靠性。

Abstract

Objective

Considering the nonlinear effects in Visible Light Communication (VLC) systems based on Orthogonal Frequency Division Multiplexing (OFDM)

an advanced classification demodulation methodology based on Gaussian Support Vector Machines (Gaussian SVM) is presented.

Methods

In this paper

the Support Vector Machines (SVM) within Machine Learning (ML) and the Gaussian kernel function are utilized to map the signal before demodulation

thereby enhancing the nonlinear processing capability to the demodulation process and improving the system's adaptabilityto the intricate channel environment. The received signals are considered as input vectors and transfigured into the high-dimensional feature space via the Gaussian SVM mapping mechanism. An optimum classification hyperplane is fabricated in the high-dimensional feature space to maximize the margin between samples. It can maximize the total distance from the vector to the hyperplane

so as to accomplish efficient classification and demodulation of signals. Through Monte Carlo simulation under diverse modulation orders (16 Quadrature Amplitude Modulation (QAM)

32QAM

64QAM) in Light Emitting Diode (LED) nonlinear channels

the prominent advantages of this approach are validated in this paper.

Results

The simulation results show that

in contradistinction to the conventional hard demodulation method

the Gaussian SVM classification demodulation method provides enhanced robustness and higher demodulation accuracy when handling nonlinear channels. Particularly in high-order modulation

such as 64QAM

this method can effectively mitigate the adverse impact of the nonlinear effect on signal quality and can obtain a 7 dB gain of signal-to-noise ratio.

Conclusion

In this paper

it isdemonstrated thatthe signal-to-noise ratio gain of the method rises with the augmentation of the nonlinear effect

which shows the potential for enhancing system performance at high modulation orders. By dynamically adapting to channel fluctuations

this method not only enhances the overall performance of the system but also significantly improves the reliability of the system.

关键词

Keywords

references

申晓欢 , 林邦姜 , 汤璇 , 等 . PDMA的可见光通信系统 [J ] . 北京邮电大学学报 , 2020 , 43 ( 1 ): 111 - 115 .

Shen X H , Lin B J , Tang X , et al . Study on PDMA based Visible Light Communication Systems [J ] . Journal of Beijing University of Posts and Telecommunications , 2020 , 43 ( 1 ): 111 - 115 .

贾科军 , 杨博然 , 陆皓 , 等 . 可见光通信光正交频分复用系统符号分解技术抑制LED非线性失真研究 [J ] . 中国激光 , 2020 , 47 ( 4 ): 0406002 .

Jia K J , Yang B R , Lu H , et al . LED Nonlinearity Mitigation for Visible Light Communication Optical-Orthogonal Frequency Division Multiplexing System with Symbol Decomposing Techniques [J ] . Chinese Journal of Lasers , 2020 , 47 ( 4 ): 0406002 .

邹鹏 , 赵一衡 , 胡昉辰 , 等 . 基于机器学习的可见光通信信号处理研究现状 [J ] . 激光与光电子学进展 , 2020 , 57 ( 1 ): 010001 .

Zou P , Zhao Y H , Hu F C , et al . Research Status of Machine Learning based Signal Processing in Visible Light Communication [J ] . Laser & Optoelectronics Progress , 2020 , 57 ( 1 ): 010001 .

Armstrong J , Schmidt B J C . Comparison of Asymmetrically Clipped Optical OFDM and DC-biased Optical OFDM in AWGN [J ] . IEEE Communications Letters , 2008 , 12 ( 5 ): 343 - 345 .

李建锋 , 刘晓爽 , 任亚浩 , 等 . 可见光通信中翻转/单极性正交频分复用对偶最大似然接收机 [J ] . 激光与光电子学进展 , 2021 , 58 ( 23 ): 2306008 .

Li J F , Liu X S , Ren Y H , et al . Pairwise Maximum Likelihood Receiver for Flip/Unipolar Orthogonal Frequency Division Multiplexing in Visible Light Communication [J ] . Laser & Optoelectronics Progress , 2021 , 58 ( 23 ): 2306008 .

Dissanayake S D , Armstrong J . Comparison of ACO-OFDM, DCO-OFDM and ADO-OFDM in IM/DD Systems [J ] . Journal of Lightwave Technology , 2013 , 31 ( 7 ): 1063 - 1072 .

刘晓爽 , 李建锋 , 任亚浩 , 等 . 无线光通信中LA-CO-OFDM的非迭代检测接收方法 [J ] . 中国激光 , 2022 , 49 ( 11 ): 1106002 .

Liu X S , Li J F , Ren Y H , et al . Noniterative Detection Receiving Method for LACO-OFDM in Wireless Optical Communications [J ] . Chinese Journal of Lasers , 2022 , 49 ( 11 ): 1106002 .

贾科军 , 南钰桐 , 杨博然 , 等 . 可见光通信预编码O-OFDM自适应符号分解技术研究 [J ] . 光通信研究 , 2024 ( 4 ): 230017 .

Jia K J , Nan Y T , Yang B R , et al . Research on Adaptive Symbol Decomposition Technology of Precoding O-OFDM in Visible Light Communication [J ] . Study on Optical Communications , 2024 ( 4 ): 230017 .

Elgala H , Mesleh R , Haas H . An LED Model for Intensity-Modulated Optical Communication Systems [J ] . IEEE Photonics Technology Letters , 2010 , 22 ( 11 ): 835 - 837 .

Li J , Huang Z , Liu X , et al . Hybrid Time-Frequency Domain Equalization for LED Nonlinearity Mitigation in OFDM-based VLC Systems [J ] . Optics Express , 2015 , 23 ( 1 ): 611 - 619 .

Shieh W , Tang Y , Krongold B S . DFT-Spread OFDM for Optical Communications [C ] // Digest of the 9th International Conference on Optical Internet (COIN 2010) . Jeju, Korea : IEEE , 2010 : 5546595 .

Li Y , Huang T , Li J , et al . Nonlinear Post Equalizer for Visible Light Communication System based on Deep Residual Convolutional Neural Network with Feature Concatenation [C ] // 2023 Photonics Global Conference (PGC) . Stockholm, Sweden : IEEE , 2023 : 10343792 .

Lu X , Liu Y , Chen J , et al . Patterns Quantization with Noise Using Gaussian Features and Bayesian Learning in VLC Systems [J ] . Journal of Lightwave Technology , 2022 , 40 ( 12 ): 3777 - 3785 .

Ma J , He J , Shi J , et al . Nonlinear Compensation based on K -Means Clustering Algorithm for Nyquist PAM-4 VLC System [J ] . IEEE Photonics Technology Letters , 2019 , 31 ( 12 ): 935 - 938 .

任亚浩 , 李建锋 , 刘晓爽 , 等 . 基于高斯混合模型的非线性可见光通信系统解调研究 [J ] . 激光与光电子学进展 , 2023 , 60 ( 15 ): 1506008 .

Ren Y H , Li J F , Liu X S , et al . Demodulation of Nonlinear Visible Light Communication based on Gaussian Mixture Model [J ] . Laser & Optoelectronics Progress , 2023 , 60 ( 15 ): 1506008 .

Zhao Y , Wang N . Parameter Selection of Gaussian Kernel for Cost-Sensitive Support Vector Machines in Imbalanced Data Classification [C ] // 2023 4th International Conference on Computer Vision, Image and Deep Learning (CVIDL) . Zhuhai, China : IEEE , 2023 : 10166009 .

赵长春 , 姜晓爱 , 金英汉 . 非线性回归支持向量机的SMO算法改进 [J ] . 北京航空航天大学学报 , 2014 , 40 ( 1 ): 125 - 130 .

Zhao C C , Jiang X A , Jin Y H . Improved SMO Algorithm of Nonlinear Regression Support Vector Machine [J ] . Journal of Beijing University of Aeronautics and Astronautics , 2014 , 40 ( 1 ): 125 - 130 .

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于OFDM系统中双相位共轭孪生波的仿真研究

基于OFDM和Nyquist单载波调制的可见光性能比较

基于∑△调制器的非线性可见光通信系统研究

PLC与VLC融合通信系统中的信道估计技术分析

基于SSD的OFDM技术在可见光通信中的应用