IMPLEMENTATION OF NEURAL NETWORK METHOD TO INVESTIGATE DEFECT CENTERS IN SEMI-INSULATING MATERIALS
Abstract
A neural network (NN) method has been proposed as a new algorithm for extraction of defect centers parameters in semi-insulating materials from experimental data obtained by photoinduced transient spectroscopy (PITS). The new algorithm is applied to investigate irradiation-induced defect centers in high resistive silicon. The folds on the PITS spectral surface formed due to the presence of defect levels are best fitted with a two-dimensional approximation function with implementation of the NN learning process. As a result, the Arrhenius plots for defect centers are obtained and the parameters of these centers are determined.
You currently do not have access to the full text article. |
---|