Abstract
We put forward and demonstrate a angle-of-arrival (AOA) based visible-light-positioning (VLP) system using quadrant-solar-cell (QSC) and third-order ridge regression machine learning (RRML) to improve the positioning accuracy.
Highlights
Visible light communication (VLC) is an emerging technology for the generation wireless communications [1,2,3]
We compare the experimental results of the QSC based AOA visible light positioning (VLP) system without ML, with the proposed 3rd order regression ML, and with the proposed 3rd order regression machine learning (RRML)
The RMS of average position errors are reduced from 7.2177 cm to 3.2025 cm, and to 3.0881 cm when using the 3rd order regression ML, and the proposed 3rd order RRML respectively; showing the improvements of 55.6% and 57.2% respectively
Summary
Visible light communication (VLC) is an emerging technology for the generation wireless communications [1,2,3]. Fingerprinting based VLP was proposed in which specific signal detected was compared to the existing record for positioning [6]. It required stable received signals over time and a large data record. Trilateration based VLP using received-signal-strength (RSS) was proposed [8]; it needed multiple Txs. Recently, angle-of-arrival (AOA) schemes were proposed based on image sensor [9] or aperture-based Rx [10]. Angle-of-arrival (AOA) schemes were proposed based on image sensor [9] or aperture-based Rx [10] Another AOA VLP system using a quadrant photodiode angular diversity aperture receiver (QADA) is reported [11]. Experimental results show that root-mean-square (RMS) of average position error is reduced by 57.2% by using the RRML
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
