High-precision approach to localization scheme of visible light communication based on artificial neural networks and modified genetic algorithms

Weipeng Guan; Yuxiang Wu; Canyu Xie; Hao Chen; Ye Cai; Yingcong Chen

doi:10.1117/1.OE.56.10.106103

10 October 2017 High-precision approach to localization scheme of visible light communication based on artificial neural networks and modified genetic algorithms

Weipeng Guan, Yuxiang Wu, Canyu Xie, Hao Chen, Ye Cai, Yingcong Chen

Author Affiliations +

Funded by: National Undergraduate Innovative Entrepreneurial Training Program, National Undergraduate Innovative and Entrepreneurial Training Program, Special Funds for the Cultivation of Guangdong College Students' Scientific and Technological Innovation (“Climbing Program” Special Funds), Special Funds for the Cultivation of Guangdong College Students’ Scientific and Technological Innovation (“Climbing Program” Special Funds), Cultivation of Guangdong College Students’ Scientific and Technological Innovation (“Climbing Program” Special Funds), Special Funds for the Cultivation of Guangdong College Students’ Scientific and Technological Innovation, Guangdong Science and Technology Plan Project, Guangdong Provincial Science and Technology Department, Guangzhou Science and Technology Planning Project, China, Guangdong Yangfan Plans to Introduce Special Team of Innovation and Entrepreneurship, Sail Plan Special Innovative Entrepreneurial Teams in Guangdong Province, Guangzhou Science and Technology Program Key Projects

Optical Engineering, Vol. 56, Issue 10, 106103 (October 2017). https://doi.org/10.1117/1.OE.56.10.106103

Abstract

An indoor positioning algorithm based on visible light communication (VLC) is presented. This algorithm is used to calculate a three-dimensional (3-D) coordinate of an indoor optical wireless environment, which includes sufficient orders of multipath reflections from reflecting surfaces of the room. Leveraging the global optimization ability of the genetic algorithm (GA), an innovative framework for 3-D position estimation based on a modified genetic algorithm is proposed. Unlike other techniques using VLC for positioning, the proposed system can achieve indoor 3-D localization without making assumptions about the height or acquiring the orientation angle of the mobile terminal. Simulation results show that an average localization error of less than 1.02 cm can be achieved. In addition, in most VLC-positioning systems, the effect of reflection is always neglected and its performance is limited by reflection, which makes the results not so accurate for a real scenario and the positioning errors at the corners are relatively larger than other places. So, we take the first-order reflection into consideration and use artificial neural network to match the model of a nonlinear channel. The studies show that under the nonlinear matching of direct and reflected channels the average positioning errors of four corners decrease from 11.94 to 0.95 cm. The employed algorithm is emerged as an effective and practical method for indoor localization and outperform other existing indoor wireless localization approaches.

Citation Download Citation

Weipeng Guan, Yuxiang Wu, Canyu Xie, Hao Chen, Ye Cai, and Yingcong Chen "High-precision approach to localization scheme of visible light communication based on artificial neural networks and modified genetic algorithms," Optical Engineering 56(10), 106103 (10 October 2017). https://doi.org/10.1117/1.OE.56.10.106103

Received: 19 June 2017; Accepted: 6 September 2017; Published: 10 October 2017

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE