AdVLP: unsupervised visible light positioning by adversarial deep learning

Abstract

Visible light positioning (VLP) is a promising technique to bring location-based service for numerous Internet of Things applications. Recent advances in VLP have shown that machine learning (ML)-based positioning algorithms show satisfying performance in physical environments under highly noisy and interference-rich conditions. With so many ML methods proposed, one major concern is that trained models could fail due to environmental heterogeneity. In this paper, we propose AdVLP, a novel adversarial training method for VLP based on deep neural networks, to address the issue of the vulnerability of data-driven approaches, which happens when channel parameters change. Our proposed method, which is inspired by generative adversarial networks, manages to adapt the two domains of the source dataset and the target dataset. The channel parameters, e.g. the Lambertian order of light-emitting diode (LED) lights and a photodiode, the LED power as well as the location error of the lights, were discussed. We observed in experiments that the accuracy of a neural network method would decrease as the bias of the parameters became larger. Secondly, a higher dimension of the parameter change would make the neural network method more vulnerable. However, our proposed approach could achieve substantial improvement in positioning accuracy.

Artificial Intelligence in Visible Light Positioning for Indoor IoT: A Methodological Review

“…A novel approach, namely AdVLP, is proposed in [80], to tackle both the challenge of estimating the 3D location of a PD receiver and the problem of data-driven methods being vulnerable to changes in channel parameters. The method is based on deep neural networks and uses adversarial training.…”

Visible Light Positioning as a Next-Generation Indoor Positioning Technology: A Tutorial

“…While some works investigate time-based (e.g., deep learning for estimating phase errors between LEDs [411]) or AoA-based VLP (e.g., [412]), by far the largest part of the research focuses on reducing the positioning errors in RSS-based VLP, induced by deterministic system errors, e.g., in the assumptions of eq. ( 19) regarding LED transmit power [413], Lambertian factor [413], transmitter tilt [414] and location [413], etc. These parameters of a system under test are implicitly extensively calibrated by training an ML model, e.g., via supervised learning [415], reinforcement learning [416], or deep learning [417], achieving fast calculation and even cm-level accuracies [418].…”

Recent advances in ubiquitous positioning systems for mobility applications

“…A novel contribution here is toward LBS for internet of things applications. In Hua et al (2021) the authors develop a novel adversarial training method suited for visible light positioning that is based on deep neural networks.…”