A practical probabilistic prediction of road traffic noise from a filtered poisson process model with a simplified elementary time pattern of triangular type

Abstract

To date, many theoretical methods for predicting the statistics of road traffic noise have been proposed involving the introduction of vehicle distribution models, such as an equally spaced model, an exponentially distributed model, or an Erlang distribution type model. In such cases, very often, the sound propagation characteristic was first restricted to an idealized case like a free sound field, and then sometimes extended to an arbitrary sound propagation environment. Needless to say, however, it was too difficult to predict systematically the probability distribution of road traffic noise fluctuation at an observation point under the actual sound propagation environment with the complex diffraction and/or attenuation effects. Thus, this paper is devoted above all to consideration of a practical probabilistic method of prediction of the statistics of road traffic noise by use of a filtered Poisson process model with a simplified elementary time pattern. The effectiveness of the proposed simple method is experimentally verified too, by applying it to the actual road traffic noise data observed in a large city.