Powder diffraction indexing as a pattern recognition problem: a new approach for determining unit cell parameters based on an artificial neural network

An artificial neural network, in combination with local optimization, is shown to be an effective approach for determining unit cell parameters directly from powder diffraction data. The viability of this new approach is initially demonstrated using simulated powder diffraction data. Subsequently, the successful application of the method to determine unit cell parameters is illustrated for two materials using experimental powder X-ray diffraction data recorded on a standard laboratory diffractometer.