Syllabus 2007/2008 ETI270

Syllabus academic year 2007/2008

DIGITAL SIGNAL PROCESSING IN AUDIO/VIDEO

ETI270

Higher education credits: 6. Grading scale: TH. Level: G2 (First level). Language of instruction: The course might be given in English. Optional for: C3, C3sst, D4, D4bg, E3bg, E3ss, Pi4. Course coordinator: Professor Leif Sörnmo, leif.sornmo@es.lth.se, Elektrovetenskap. Prerequisites: ESS040 Digital Signal Processing, ETI265 Signal Processing in Multimedia or ETT080 Signals and Communications. Assessment: Exam at the end of the course. Optional hand-in problem after half-time. Two mandatory laboratory exercises. Home page: http://www.es.lth.se/ugradcourses/digsigav.

Aim
The course provides solutions to problems in signal processing where filter design and efficient algorithms are in demand. The student should obtain sufficient insights on theory and application to independently formulate the mathematical problem, to solve it, and to implement the solution for processing of real signals.

Knowledge and understanding
For a passing grade the student must

have knowledge on different approaches to filter design, derivation of efficient algorithms, e.g., the FFT, and filter banks.
to apply the most common methods on real problems and signals related to audio and video (Matlab level)
to define mathematical models within the area based on various situations.

Skills and abilities
For a passing grade the student must

to understand the principles of different filter design methods (IIR and FIR and variants), and to design such filters in Matlab.
to describe and analyze error sources present in the implementation of algorithms in the DSP environment.
to describe efficient algorithms for the computation of the discrete-time Fourier transform.
to describe and analyze the most common types of filter banks.
to describe and implement algorithms for audio signal processing.

Judgement and approach
For a passing grade the student must

be able to analyze, assess, and implement filters, filter banks, and efficient algorithms, and also to interpret and to describe their inherent principles.
have insight on the fact that seemingly different technical problems can be dealt with using the same methods.

Contents
Filter design

Digital IIR filter design
Bilinear transformation
Digital FIR filter design
Windowing, ideal filters, and the Gibbs phenomenon
Equirippel filters

Implementation

Structure verification
Efficient FFT algorithms
Different types of quantization

Multirate signal processing and filter banks

Up- and downsampling
Decimation and interpolation
Polyphase decomposition
Nyqvist filters
Uniform filter banks
Two-channel QMF
Multirate filters and wavelets

Audio- and video signal processing

Signal processing for sound effects (echo, reverb, and others)
Signal processing for data compression of music (mp3)
Signal processing for image and video compression

Literature
Mitra S K: Digital Signal Processing, A Computer Approach, third edition, McGraw-Hill 2006. ISBN: 0-07-286546-6, and additional material provided by the department.