Multipopulation genetic algorithm-optimized LED layout in a visible light communication system

Abstract

Visible light communication (VLC) systems provide illumination and power, and the distributed, specifically made light-emitting diodes (LEDs) are their fundamental sources of light and signal. We propose an arbitrarily distributed LED lamp layout based on a multipopulation genetic algorithm (MPGA) to solve the unevenness of distributions of optical illuminance and power. Twenty LED lamps were taken as an example, and the position coordinates and half-power angles were optimized under the fitness function related to the variance of received power through the coevolution of multipopulations. The simulation results on MATLAB R2016a showed that the distributions of illuminance and power were more uniform, with a variance of power of 0.3978 dBm and a uniformity ratio of illuminance of 87.42%. Moreover, SNR distribution was excellent, with a QSNR of 0.2007. An average root mean square delay spread of 2.22 ns was obtained, with a variance of 5.6250 × 10 − 20. The proposed method is better than the optimization of lamp arrangement, configuration, and genetic algorithm (GA). Furthermore, from iteration curves, the MPGA acts better than the GA in finding the solution. The fitness function value is better than the GA because the latter easily falls into the locally optimal solution. We provide references to design the LED layout and contribute to making a more comfortable communication environment.