Auralization of road traffic:a study of finite length noise barriers

Abstract

Road-traffic noise is one of the dominating noise sources in urban areas and consequently creates problems for health and wellbeing. Engineers and architects often use mathematical models (auralization) to simulate sound during the planning and design process of buildings e.g., simulate sound fields in virtual spaces indoors. At the division of Applied Acoustics at Chalmers a toolbox to auralize sound outdoors has been developed, with which it is possible to simulate sounds of a vehicle passing by with or without a noise barrier. A further study of noise barriers is conducted in this work. The aim was to investigate the Harmonoise model for noise barriers and to evaluate the perception of the sounds created with the model. The Harmonoise model was added into the existing toolbox and auralized sounds for different receiver locations, both with and without noise barrier, were evaluated in a listening test together with real recorded sounds without noise barrier. The participants in the listening test noticed a difference in plausibility between the recorded and the auralized sounds. The auralized sounds with and without noise barrier gave a similar plausibility at 27 km/h and 45 km/h. The sounds with noise barrier had higher average plausibility than the sounds without noise barrier, when the receiver was placed nearest to the edge of the noise barrier. The participants perceived that the sounds with noise barrier were most plausible when the distance between the receiver and the road was 7.5 m whereas the plausibility decreased when the distances increased to 15 m and 30 m. Whereas the average plausibility decreased when a large portion of the sound clip consisted of sound screened by the noise barrier, a higher average plausibility was obtained when a shorter time of the sound clip consisted of sound screened by the noise barrier.