Influence of Lighting LED Design Parameters on the Dynamic Nonlinear Response

In study, we investigated the influence of lighting light emitting diodes (LED) chip parameters, including the doping concentration, ABC parameters, and bias current, on the kernels of the Volterra representation of the nonlinear dynamic response. The Volterra kernels were obtained from the LED rate equations and transformed from the time domain to the frequency domain using the harmonic input method. An analysis of kernels up to third order for a specific type of lighting LED (the Luxeon Rabel blue LED) demonstrated the dynamic character of LED nonlinearity. We further validated our approach by application to a simulated discrete multitone (DMT) modulation experiment and obtained excellent agreement between the theoretical and simulation results. We also proposed two measures of LED nonlinearity based on the first- and second-order kernels of the frequency Volterra representation. These measures facilitated analysis of the device’s nonlinear behavior for other advanced modulation formats, such as pulse amplitude modulation (PAM), and made subsequent investigation of the impact of the LED chip parameters on the 1st-order frequency response and dynamic nonlinearity easier. The latter analysis demonstrated that with increasing doping concentration, the 3-dB modulation bandwidth increased, whereas the nonlinear distortions were diminished. Similarly, increasing the bias current boosted the modulation bandwidth, as well as the nonlinear distortions (unlike to the trend obtained by increasing the doping concentration). The obtained results may be useful for designing LED chips and selecting a suitable LED modulation format.