Modeling Temperature Dependent Nonlinearity in a Laser Diode Using Fuzzy Logic

Modeling of non-linear behavior of a laser diode is carried out in this paper. A combined approach of fuzzy logic and neural network is used. An Adaptive Neuro-Fuzzy Inference System (ANFIS) based model is trained using experimental data. Theoretical analysis of temperature dependent nonlinearity becomes complex as it requires simultaneous consideration of mechanisms like Auger recombination, Inter Valence Band Absorption (IVBA), thermionic carrier emission out of the active region, lateral carrier spreading, passive layer absorption, spontaneous recombination within passive layers and optical gain reductions. An ANFIS based model presented in this paper accounts for this temperature dependent nonlinearity. Once the model is trained it computes the Total Harmonic Distortion (THD) and the power output for various drive currents and operating temperatures. The validation of the model using experimental data yields a Mean Squared Error (MSE) of 2.35x10 -4