Wavelet transform-based artificial neural networks (WT-ANN) in PM10 pollution level estimation, based on circular variables

Introduction In this paper, a novel method in the estimation and prediction of PM10 is introduced using wavelet transform-based artificial neural networks (WT-ANN). Discussion First, the application of wavelet transform, selected for its temporal shift properties and multiresolution analysis characteristics enabling it to reduce disturbing perturbations in input training set data, is presented. Afterward, the circular statistical indices which are used in this method are formally introduced in order to investigate the relation between PM10 levels and circular meteorological variables. Then, the results of the simulation of PM10 based on WT-ANN by use of MATLAB software are discussed. The results of the above-mentioned simulation show an enhanced accuracy and speed in PM10 estimation/prediction and a high degree of robustness compared with traditional ANN models.