LAUSR

A light-emitting diode (LED) acts as a transmitter in a visible-light communication (VLC) system. However, the nonlinear characteristics of LED limit the performance of the VLC system by degrading the quality of a transmitted signal. In recent years, several forms of predistorters have been proposed to mitigate the effects of LED nonlinearity; however, none of them have been able to approach the performance of a linear VLC system. In this paper, we propose an adaptive learning architecture (ALA)-based predistortion technique to estimate and compensate for LED nonlinearities in a VLC system. A DC-biased optical orthogonal frequency division multiplexing signal is considered. The performance with and without predistorter is analyzed assuming optical channel. It is shown that degradation due to LED nonlinearity can be compensated by using ALA-based predistortion, and the overall predistorter–LED system is able to approach near-linear performance. Further, the proposed predistorter architecture is also able to track the variations in LED nonlinearity and compensate them. Simulation results based on error vector magnitude, symbol error rate, amplitude distortion (AM/AM) curves and constellation plots validate the performance of our proposed technique.