Projekte sind ein essenzieller Bestandteil des Curriculums von Mobile Computing. Die Studierenden bekommen die Möglichkeit, das im Zuge ihres Studiums erworbene theoretische Wissen selbst praktisch umzusetzen. Ein sowohl für StudentInnen als auch für Lehrende immer wieder spannendes Unterrichtskonzept, in dem schon erfolgreiche Startups wie z.B. runtastic und Butleroy ihre Anfänge gefunden haben.
Detecting Fundamental Frequency of Drums
This project deals with the basics of drum tuning and methods to do it. Problems of conventional tuning methods are described and practical solutions for those are presented. The goal is to implement a suitable and modern application for Android smartphones which assists the user in tuning the drums by detecting and displaying the fundamental frequency of the drum. However, this part of the project is limited to finding a suitable algorithm for detecting the frequency from pre-recorded audio samples and implementing it in MATLAB.
The used algorithms utilize pre-processing in terms of down-mixing and normalization. Windowing and filtering of the signal is also considered in the implementation.
Pre-recorded audio samples of drums are used for proof of concept. The detection of the fundamental frequency is done in frequency domain by computing the fast Fourier transform of the pre-processed signal. From this, the magnitude spectrum and the normalized autocorrelation function are calculated and plotted.
The simplest way of estimating the fundamental frequency of a sound is by extracting the frequency with the highest magnitude relative to the others. Generally, this approach is considered to be very unreliable and inaccurate although it can still produce decent results in some cases.
Utilizing the normalized autocorrelation function, the distance between the fundamental frequency and its harmonics can be estimated. This distance or lag corresponds to the fundamental frequency because the harmonics are positioned at integer multiples of the fundamental frequency.