Real time estimation of particle size distribution has a great importance for advanced control strategies that can allow diesel engines to comply with future emission standards. Moreover, knowledge of real time particulate size distribution allows the optimization of the functioning of after-treatment systems. The aim of this paper is to present a Neural Network model able to provide real time information about the characteristics of particulate emissions from a Diesel engine. The model has as inputs some engine parameters such as engine speed, engine load, EGR ratio, etc., and, as output, the particle size distribution. Preliminary results indicated that the model shows, for every engine operating condition, a satisfactory capability of estimating the concentrations of particulate particles with prefixed diameters.