A Deep Learning Approach to Universal Binary Visible Light Communication Transceiver

被引:16
作者
Lee, Hoon [1 ]
Quek, Tony Q. S. [2 ,3 ]
Lee, Sang Hyun [4 ]
机构
[1] Pukyong Natl Univ, Dept Informat & Commun Engn, Busan 48513, South Korea
[2] Singapore Univ Technol & Design, Singapore 487372, Singapore
[3] Kyung Hee Univ, Dept Elect Engn, Yongin 17104, South Korea
[4] Korea Univ, Sch Elect Engn, Seoul 02841, South Korea
基金
新加坡国家研究基金会;
关键词
Training; Transceivers; Optical transmitters; Optical pulses; Light emitting diodes; Receivers; Neural networks; Visible light communication; deep learning; dimming support; primal-dual method; DESIGN; NONLINEARITY; MITIGATION; SCHEME;
D O I
10.1109/TWC.2019.2950026
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper studies a deep learning (DL) framework for the design of binary modulated visible light communication (VLC) transceiver with universal dimming support. The dimming control for the optical binary signal boils down to a combinatorial codebook design so that the average Hamming weight of binary codewords matches with arbitrary dimming target. An unsupervised DL technique is employed for obtaining a neural network to replace the encoder-decoder pair that recovers the message from the optically transmitted signal. In such a task, a novel stochastic binarization method is developed to generate the set of binary codewords from continuous-valued neural network outputs. For universal support of arbitrary dimming target, the DL-based VLC transceiver is trained with multiple dimming constraints, which turns out to be a constrained training optimization that is very challenging to handle with existing DL methods. We develop a new training algorithm that addresses the dimming constraints through a dual formulation of the optimization. Based on the developed algorithm, the resulting VLC transceiver can be optimized via the end-to-end training procedure. Numerical results verify that the proposed codebook outperforms theoretically best constant weight codebooks under various VLC setups.
引用
收藏
页码:956 / 969
页数:14
相关论文
共 44 条
  • [1] Color Intensity Modulation for Multicolored Visible Light Communications
    Ahn, Kang-Il
    Kwon, Jae Kyun
    [J]. IEEE PHOTONICS TECHNOLOGY LETTERS, 2012, 24 (24) : 2254 - 2257
  • [2] [Anonymous], KINGBRIGHT LED DATAS
  • [3] [Anonymous], P IEEE GLOB WORKSH
  • [4] [Anonymous], 2003, TECH REP
  • [5] [Anonymous], 2013, CoRR abs/1308.3432
  • [6] Model-Free Training of End-to-End Communication Systems
    Aoudia, Faycal Ait
    Hoydis, Jakob
    [J]. IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2019, 37 (11) : 2503 - 2516
  • [7] Learning Deep Architectures for AI
    Bengio, Yoshua
    [J]. FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2009, 2 (01): : 1 - 127
  • [8] Boyd S., 2010, Found. Trends Mach. Learn, V1, P122, DOI DOI 10.1561/2200000016
  • [9] Boyd S., 2004, CONVEX OPTIMIZATION
  • [10] Mitigating LED Nonlinearity to Enhance Visible Light Communications
    Deng, Xiong
    Mardanikorani, Shokoufeh
    Wu, Yan
    Arulandu, Kumar
    Chen, Bin
    Khalid, Amir M.
    Linnartz, Jean-Paul M. G.
    [J]. IEEE TRANSACTIONS ON COMMUNICATIONS, 2018, 66 (11) : 5593 - 5607