Noise pollution from air traffic is a major environmental problem affecting many citizens. Aircraft flyover noise represents an extremely complex auditory scenario. Noise annoyance is not only dependent on sound exposure levels. There are also many other acoustic and psycho-acoustic factors (spectral content, time modulations, sharpness, tonality,...) that play an important role.
This paper presents a model-based sound synthesis approach which takes these factors into consideration. The sound synthesis approach is interesting for a number of reasons: (1) it allows to study and assess aircraft noise annoyance in relation to the various acoustic and psycho-acoustic characteristics of the sound, (2) it helps to understand sound quality differences among various types of aircrafts in both landing and take-off conditions and (3) it forms an excellent basis for target sound design.
The presented research is one of our major activities in the EU-project SEFA (Sound Engineering for Aircraft; coordinator: Dornier GmbH, Germany) which aims at developing technology design criteria to make aircrafts more acceptable from a noise signature point of view.
The synthesis of aircraft flyover noise is based on a sound synthesis model that is identified from sound recordings near the ground. A sound synthesis model is a compact and sound-quality-accurate model consisting of Doppler shifted tonal components and of third octave noise bands to characterize the broadband noise. The reflection characteristics of the ground and the time-delay between the direct incident and reflected sound are taken into consideration to characterize the typical ground-interference pattern in the sounds.
The sound synthesis approach was applied to a large database of aircraft sound recordings, covering various types of aircrafts in landing and take-off conditions. The flyover noise synthesis results were validated with a number of listening tests. Impressive results were achieved.
