Receiver Power Optimization in Indoor Visible Light Systems Using Machine Learning Method

As one of the indoor wireless communication technologies, visible light communication (VLC) has gradually become a hot research direction because of its high security, good confidentiality, large bandwidth, and so on. One or more light emitting diodes (LEDs) as light sources can support low energy consumption lighting and high-speed wireless information access. In this paper, by building the indoor model, according to Lambert radiation model, particle swarm optimization (PSO) and ant colony algorithm (ACA) are used to analyze LEDs layout. The optimal layout is achieved in a
$\mathbf{5}\mathbf{m}\times \mathbf{5}\mathbf{m}\times \mathbf{3}\mathbf{m}$
room. The direct current (DC) gain is
$\mathbf{1.1847}\times \mathbf{10}^{-\mathbf{6}}$
under normal layout,
$\mathbf{2.3189}\times \mathbf{10}^{-\mathbf{6}}$
using PSO algorithm and
$\mathbf{2.0891}\times \mathbf{10}^{-\mathbf{6}}$
using ACA.