Optimization of multi-receiver SNRs for indoor visible light communication based on modified evolutionary algorithm.

Based on indoor visible light communication (VLC) technology, the large-scale LED array can be used for the integration system of communication, positioning and illumination. However, due to the different propagation time between LEDs and receivers, the signal to noise ratio (SNR) is affected by the inter symbol interference (ISI). This paper proposes a multi-receiver SNR optimization method by selecting communicating LEDs in the LED array. The proposed method uses the modified evolutionary algorithm (EA) to obtain the optimal SNR. In the processes of initialization, crossover and mutation operations, instead of general stochastic generation and transformation, the heuristic gene initialization, maximum contributing gene reservation and the minimum contributing gene elimination, which are specially designed for VLC LED selection problem, are proposed to improve the EA. The delay features of different LEDs to receivers are quantitatively analyzed and used as the criteria to gene initialization, crossover and mutation in order to attenuate the ISI. Simulation results reveal that the proposed method converges faster and obtains larger SNRs than the general and the hyper heuristic EA. [ABSTRACT FROM AUTHOR]