Robust Visible Light Communication Networks: System Analysis and Implementation

Wireless networks are dominating our daily lives. By 2023, over 70% of the global population will have mobile connectivity and there will be 3.6 networked devices per capita. To accommodate this increasing density in wireless devices, further advancements in communication technologies are required. Despite the continuous efforts spent on Radio Frequency (RF) technologies, they lack the sufficient RF spectrum to meet this ongoing and ever-expanding traffic demand. To tackle this, Visible Light Communication (VLC) has recently attracted significant attention on complementing RF networks to provide robust communication in ultra-dense scenarios. In VLC, data is transmitted by rapidly modulating the light intensity of Light Emitting Diodes (LEDs), with receivers able to decode the information at data rates up to Gbit/s. Nonetheless, VLC brings extra challenges as the propagation is predominantly line-of-sight (LOS), the channel is highly subject to the geometric properties of the environment, the small confinement of light limits user mobility and overlapping light sources require tight interference coordination. The goal of this thesis is to provide deeper insights in overcoming these challenges and making VLC networks robust in all circumstances. To achieve this, the thesis starts from a dense transmitter infrastructure and aims to optimize several representing and important performance metrics. The outcome of this thesis can be summarized in five main contributions. As a first contribution, I design and build a proof-of-concept VLC testbed consisting of 36 densely deployed transmitters and 4 mobile receivers. By supporting dynamic adaptation of the transmitter (known as beamsteering) and receiver orientations, the VLC channel can be precisely characterized in a wide range of experiments. This testbed functions as a fundamental building block in this thesis for the experimental validation of the proposed systems and algorithm designs. Relying on this testbed, the second contribution includes the design of a multi-user cell-free VLC network. Unlike conventional cellular networks, there is no partitioning in cells and all users are simultaneously served by all transmitters within the receiver's field-of-view (FOV). I optimize the power efficiency in the network and conclude from the evaluation results that a dense cell-free network is a suitable candidate for VLC. As the LOS link can easily be blocked by the surroundings, the third contribution investigates the impact of blockage on the performance and proposes two techniques to mitigate it: 1) transmitter densification and 2) a user-in-the-loop system that optimizes the rotation of users holding the receivers. Since the VLC channel is very sensitive to the geometry of the transmitters and receivers, misalignment can cause severe connectivity outages. To avoid this, as a fourth contribution, I optimize their orientation in a low-complexity manner. Rotating the receiver can greatly increase the lowest signal strengths and thus boosting the robustness. At the transmitter's side, steering the beams allows to generate bright beamspots that follow the users. I show that a blended architecture, consisting of a mixture of fixed and steerable light sources, can successfully boost the data rate while also supporting user mobility. Lastly, as a fifth contribution, I propose a low-complexity yet powerful interference coordination technique for VLC. To accomplish this, I develop a mathematical framework to decide when interference is harmful. Relying on this, I design a time division contention-free scheduling protocol. The protocol is semi-distributed, executing the coarse-grained interference control centrally and fine-grained control distributively on the transmitters. By taking this divide-and-conquer approach, I can significantly reduce the complexity compared to the state of the art without sacrificing performance. Despite these advances in VLC communication, next-generation networks are not expected to count on a single technology. That is, realizing ultra-robust wireless communication will require seamless integration of multiple technologies, including RF. Nevertheless, this work takes a big step forward in analyzing and improving the robustness of VLC networks, paving the way towards ultra-robust wireless communication with visible light.

Publication year:2020

Accessibility:Open