Open AccessLow spatial complexity adaptive artificial neural network post-equalization algorithms in MIMO visible light communication systems
Author(s) -
Yiheng Zhao,
Peng Zou,
Zhixue He,
Ziwei Li
Publication year - 2021
Publication title -
optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.394
H-Index - 271
ISSN - 1094-4087
DOI - 10.1364/oe.440155
Subject(s) - mimo , visible light communication , quadrature amplitude modulation , computer science , artificial neural network , equalization (audio) , bit error rate , algorithm , superposition principle , adaptive equalizer , communications system , electronic engineering , telecommunications , decoding methods , mathematics , optics , artificial intelligence , engineering , beamforming , physics , mathematical analysis , light emitting diode
In this paper, we experimentally propose a feasible and low spatial complexity adaptive artificial neural network (AANN) post-equalization algorithm in MIMO visible light communication (VLC) systems. By introducing the power ratio and the MIMO least mean square (MIMO-LMS) post-equalization algorithm into the structure design process of the artificial neural network (ANN) post-equalization algorithm, we reduced the spatial complexity of the post-ANN equalization algorithm to less than 10%. At the same time, the bit error rate (BER) performance of AANNs did not decrease. Finally, we achieved a data rate of 2.1Gbps in the AANN equalized 16QAM superposition coding modulation (SCM) and carrier-less amplitude-phase (CAP) single-receiver MIMO (SR-MIMO) VLC system.