Due to the evolution of digital technology more and more signal processing tasks are accomplished digitally using microprocessors or dedicated digital hardware. The most important advantage of digital signal processing is flexibility. A digital signal processor (DSP) may serve as an universal device for many different applications. Even in the case of non-stationary signals, signals whose statistical properties vary in time, a digital system may adjust its algorithms automatically according to a pre-defined error criteria. This behaviour is known as adaptive signal processing.

Many basic signal processing techniques are already applicable to a wide range of problems. These general techniques are the subject of this course. Thus our course is designed to provide an introduction to the field of digital signal processing.

Besides the theory several examples are given to support the understanding of how the algorithms work. MATLAB software is used for a straightforward notation of the algorithms and to run the examples.

The lecture starts with an introduction providing some mathematical background and notation of digital signals and systems. Furthermore the sampling theorem, the z-transform and also the MATLAB software is refreshed. A further chapter is on spectral and coherence estimation with applications to random time-series analysis. We will go on with filter design, decimation and interpolation (sampling rate conversion), least squares modelling and finally we discuss adaptive signal processing and Wiener filters.