Numerical Methods in Acoustics

Lecturer

Wolfgang Kreuzer (wolfgang.kreuzer(at)oeaw.ac.at)

Tuesday 13:00 - 14:30, Zoomlink

Content

Acoustics is concerned with the generation, propagation and perception of sound and combines research from different scientific disciplines like theoretical and applied mathematics, physics, mechanics, biology, psychology and many more.
In the course we want to give an overview over some methods and models applied in acoustics. We will give an overview of the theoretical background as well as applications of these models. The course includes an introduction into basic concepts of acoustics, time frequency analysis of signals and numerical methods to solve the Helmholtz equation. These methods can be used to model and calculate for example the resonances of the human vocal tract, the efficiency on noise barriers or filter functions based on the geometry of the human ear (pinna) that are important for 3D audio.

Additional Material

  • A short introduction to acoustics (in german)
  • There is some sort of technical report (in German), that covers the topics of the lecture in a bit more detail. But please be careful, it is very likely that there are still some errors in it, that I haven't found yet: Numerische Methoden in der Akustik. The notes have been updated in December, here are the changes.
  • From time to time I will also use some Octave/Matlab scripts that could be interesting to experiment with:
    • Missing Fundamental: Demoscript for the topic Missing Fundamental and Tartini tone
    • Gibbs Effekt: Demo regarding Gibbs Effekts. A square wave is approximated by its Fourier series with different lenghts. Listen how the different approximations sound.
    • Filterdemo1: Demo regarding cutting and filtering of a signal. We take a combination of two sin tones with noninteger frequencies and cut them in the frequency domain and do a filtering.
    • Filterdemo2: Demo regardigin cutting and Filtering of  (white) noise.
    • Chirp: Demo eines Chirp Signals, das mit einer zu geringen Samplingfrequenz abgetastet wird