Table of Contents
Fetching ...

Toward Hybrid COTS-based LiFi/WiFi Networks with QoS Requirements in Mobile Environments

Emilio Ancillotti, Loreto Pescosolido, Andrea Passarella

TL;DR

This paper tackles QoS-aware vertical handover in hybrid LiFi/WiFi networks built from commercial off-the-shelf devices, focusing on mobile indoor scenarios where LiFi performance may degrade before a full connectivity loss. It introduces two proactive handover mechanisms—EWMA-based monitoring of signal power and CRC-based packet failure ratio—to anticipate QoS degradation and trigger a switch to WiFi before QoS falls below a target of $20$ Mbps. Evaluated on a lab conveyor-belt testbed with realistic mobility, the CRC-based method with a confirmation check achieves QoS outages under $1$ second and maintains a stable WiFi takeover distance around $120$ cm across speeds, outperforming the baseline which triggers handover only on complete link loss. The work demonstrates practical, real-time QoS-aware handovers using Libteam with COTS hardware and lays groundwork for parameter optimization and multi-user extensions.

Abstract

We consider a hybrid LiFi/WiFi network consisting of commercially available equipment, for mobile scenarios, where WiFi backs up communications, through vertical handovers, in case of insufficient LiFi QoS. When QoS requirements in terms of goodput are defined, tools are needed to anticipate the vertical handover relative to what is possible with standard basic mechanisms, which are only based on a complete loss of connectivity. We introduce two such mechanisms, based on signal power level readings and CRC-based packet failure ratio, and evaluate their performance in terms of QoS-outage duration, considering as a benchmark an existing baseline solution based on the detection of a connectivity loss. In doing this, we provide insights into the interplay between such mechanisms and the LiFi protocol channel adaptation capabilities. Our experimental results are obtained using a lab-scale testbed equipped with a conveyor belt, which allows us to accurately replicate experiments with devices in motion. With the proposed methods, we achieve QoS outages below one second for a QoS level of 20 Mbps, compared to outage durations of a few seconds obtained with the baseline solution.

Toward Hybrid COTS-based LiFi/WiFi Networks with QoS Requirements in Mobile Environments

TL;DR

This paper tackles QoS-aware vertical handover in hybrid LiFi/WiFi networks built from commercial off-the-shelf devices, focusing on mobile indoor scenarios where LiFi performance may degrade before a full connectivity loss. It introduces two proactive handover mechanisms—EWMA-based monitoring of signal power and CRC-based packet failure ratio—to anticipate QoS degradation and trigger a switch to WiFi before QoS falls below a target of Mbps. Evaluated on a lab conveyor-belt testbed with realistic mobility, the CRC-based method with a confirmation check achieves QoS outages under second and maintains a stable WiFi takeover distance around cm across speeds, outperforming the baseline which triggers handover only on complete link loss. The work demonstrates practical, real-time QoS-aware handovers using Libteam with COTS hardware and lays groundwork for parameter optimization and multi-user extensions.

Abstract

We consider a hybrid LiFi/WiFi network consisting of commercially available equipment, for mobile scenarios, where WiFi backs up communications, through vertical handovers, in case of insufficient LiFi QoS. When QoS requirements in terms of goodput are defined, tools are needed to anticipate the vertical handover relative to what is possible with standard basic mechanisms, which are only based on a complete loss of connectivity. We introduce two such mechanisms, based on signal power level readings and CRC-based packet failure ratio, and evaluate their performance in terms of QoS-outage duration, considering as a benchmark an existing baseline solution based on the detection of a connectivity loss. In doing this, we provide insights into the interplay between such mechanisms and the LiFi protocol channel adaptation capabilities. Our experimental results are obtained using a lab-scale testbed equipped with a conveyor belt, which allows us to accurately replicate experiments with devices in motion. With the proposed methods, we achieve QoS outages below one second for a QoS level of 20 Mbps, compared to outage durations of a few seconds obtained with the baseline solution.

Paper Structure

This paper contains 8 sections, 1 equation, 9 figures.

Figures (9)

  • Figure 1: Hybrid LiFi/WiFi network architecture
  • Figure 2: LED lamp and IR sensor on the ceiling
  • Figure 3: Conveyor Belt
  • Figure 4: Goodput measured at different distances under various speed conditions
  • Figure 5: Goodput averages at different speeds
  • ...and 4 more figures