Non-line-of-sight Visible Light Positioning System based on Deep Learning

【Objective】Visible Light Positioning (VLP) technology has gained increasing attention due to its potential for providing low-cost, high-precision indoor location services. However, traditional VLP systems rely on Line-of-Sight (LOS) paths and cannot function properly when obstructed by obstacles.【Me...

Full description

Saved in:
Bibliographic Details
Main Authors: HUANG Weijie, LIN Bangjiang, DING Yongqi, LUO Jiabin, HUANG Tianming
Format: Article
Language:zho
Published: 《光通信研究》编辑部 2024-12-01
Series:Guangtongxin yanjiu
Subjects:
Online Access:http://www.gtxyj.com.cn/thesisDetails#10.13756/j.gtxyj.2024.230091
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850073021812834304
author HUANG Weijie
LIN Bangjiang
DING Yongqi
LUO Jiabin
HUANG Tianming
author_facet HUANG Weijie
LIN Bangjiang
DING Yongqi
LUO Jiabin
HUANG Tianming
author_sort HUANG Weijie
collection DOAJ
description 【Objective】Visible Light Positioning (VLP) technology has gained increasing attention due to its potential for providing low-cost, high-precision indoor location services. However, traditional VLP systems rely on Line-of-Sight (LOS) paths and cannot function properly when obstructed by obstacles.【Methods】To address this issue, we propose a novel Non-Line-of-Sight (NLOS) VLP system based on deep learning. This system utilizes reflected light for VLP, overcoming the challenge of LOS obstruction and enhancing the robustness of the VLP system. Considering the low signal-to-noise ratio of the reflected light, the accuracy and adaptability of conventional image detection methods for extracting Light Emitting Diode (LED) spots are limited, resulting in reduced positioning accuracy for NLOS VLP. Therefore, the proposed system employs the deep learning model U-shaped Network (U-Net) to detect LED spots, which demonstrates high accuracy and adaptability after being trained on datasets collected from various environments, thereby improving the system performance. In the simulation, the system estimates the Three-Dimensional (3D) position of the receiver using the Perspective-Three-Point (P3P) algorithm.【Results】This paper constructed a 1.84 m×1.84 m ×1.96 m 3D space simulating an indoor environment for indoor positioning experiments. The experimental results show that under NLOS paths, the system's 3D mean error and Root Mean Square Error (RMSE) are 16.09 and 17.18 cm, respectively. The Two-Dimensional (2D) positioning error has a 90% confidence level at less than 21 cm, and the 3D positioning error has a 90% confidence level at less than 24 cm.【Conclusion】The proposed system has high positioning accuracy and robustness, which can meet the positioning requirements of most indoor applications.
format Article
id doaj-art-7736498af50c4351af047c80b31d0ad8
institution DOAJ
issn 1005-8788
language zho
publishDate 2024-12-01
publisher 《光通信研究》编辑部
record_format Article
series Guangtongxin yanjiu
spelling doaj-art-7736498af50c4351af047c80b31d0ad82025-08-20T02:46:57Zzho《光通信研究》编辑部Guangtongxin yanjiu1005-87882024-12-01230091012300910778025199Non-line-of-sight Visible Light Positioning System based on Deep LearningHUANG WeijieLIN BangjiangDING YongqiLUO JiabinHUANG Tianming【Objective】Visible Light Positioning (VLP) technology has gained increasing attention due to its potential for providing low-cost, high-precision indoor location services. However, traditional VLP systems rely on Line-of-Sight (LOS) paths and cannot function properly when obstructed by obstacles.【Methods】To address this issue, we propose a novel Non-Line-of-Sight (NLOS) VLP system based on deep learning. This system utilizes reflected light for VLP, overcoming the challenge of LOS obstruction and enhancing the robustness of the VLP system. Considering the low signal-to-noise ratio of the reflected light, the accuracy and adaptability of conventional image detection methods for extracting Light Emitting Diode (LED) spots are limited, resulting in reduced positioning accuracy for NLOS VLP. Therefore, the proposed system employs the deep learning model U-shaped Network (U-Net) to detect LED spots, which demonstrates high accuracy and adaptability after being trained on datasets collected from various environments, thereby improving the system performance. In the simulation, the system estimates the Three-Dimensional (3D) position of the receiver using the Perspective-Three-Point (P3P) algorithm.【Results】This paper constructed a 1.84 m×1.84 m ×1.96 m 3D space simulating an indoor environment for indoor positioning experiments. The experimental results show that under NLOS paths, the system's 3D mean error and Root Mean Square Error (RMSE) are 16.09 and 17.18 cm, respectively. The Two-Dimensional (2D) positioning error has a 90% confidence level at less than 21 cm, and the 3D positioning error has a 90% confidence level at less than 24 cm.【Conclusion】The proposed system has high positioning accuracy and robustness, which can meet the positioning requirements of most indoor applications.http://www.gtxyj.com.cn/thesisDetails#10.13756/j.gtxyj.2024.230091VLPNLOSdeep learningP3P algorithm
spellingShingle HUANG Weijie
LIN Bangjiang
DING Yongqi
LUO Jiabin
HUANG Tianming
Non-line-of-sight Visible Light Positioning System based on Deep Learning
Guangtongxin yanjiu
VLP
NLOS
deep learning
P3P algorithm
title Non-line-of-sight Visible Light Positioning System based on Deep Learning
title_full Non-line-of-sight Visible Light Positioning System based on Deep Learning
title_fullStr Non-line-of-sight Visible Light Positioning System based on Deep Learning
title_full_unstemmed Non-line-of-sight Visible Light Positioning System based on Deep Learning
title_short Non-line-of-sight Visible Light Positioning System based on Deep Learning
title_sort non line of sight visible light positioning system based on deep learning
topic VLP
NLOS
deep learning
P3P algorithm
url http://www.gtxyj.com.cn/thesisDetails#10.13756/j.gtxyj.2024.230091
work_keys_str_mv AT huangweijie nonlineofsightvisiblelightpositioningsystembasedondeeplearning
AT linbangjiang nonlineofsightvisiblelightpositioningsystembasedondeeplearning
AT dingyongqi nonlineofsightvisiblelightpositioningsystembasedondeeplearning
AT luojiabin nonlineofsightvisiblelightpositioningsystembasedondeeplearning
AT huangtianming nonlineofsightvisiblelightpositioningsystembasedondeeplearning