A theoretical model for parallel noise barriers

The conventional approach for modeling parallel noise barriers has been mostly based on using the single barrier theory with image sources to model the reflected fields. This approach does not take into account diffraction in the field between the two barriers. The conventional method also tends to predict higher insertion loss degradation than field measurements have shown. A theoretical model is proposed to include diffraction in the part of the field due to single or multiple reflections. For the reflected field between the barriers, a barrier is assumed to act as a baffled sound source whose pressure amplitude and phase are given by that of the reflected wave. The source field and the barrier shape are unrestricted. In this case, only the reflected portion of the incident wave propagates back to the opposite barrier whereas in the image source method the whole incident wave is reflected. The angular spectrum method is used to calculate the field from the various reflections. The new model is compared to various other calculations and published field measurements.