Digital Source Coding (3V, 0Ü)

This lecture covers the description of principles for data reduction. For understanding the fundamentals of data reduction it is important to know characteristics of sources and as well as the principles of audio-visual perception. Typical examples for audio and video codecs suggested for MPEG-2 are developed and explained using the introduced models and coding principles.

Markoff-source, AR-process
Entropy coding
Lineare and nonlinear quantization
predictive coding, transform coding
Data compression for audio und video
Current technology trends: object based coding (MPEG-4), JPEG 2000

Event / Lecturer	Location	Date	Start
Lecture Dr. Wolfgang Endemann	P1-03-316	Thursday, 10:00 - 12:15	Expected start: Thursday, 16.04.2026
Note: Registration for the event is done through the LSF (Link).

Course content

1. Fundamentals of source coding: sources, sinks, decorrelation, quantization, coding
2. Decorrelation techniques: techniques in the time and frequency domain
3. Functional blocks of modern source coding methods: hybrid DCT, wavelets, vector quantization, algebraic coding, object-oriented coding
4. Structure of codecs for audio coding (speech codecs and generic codecs), still image coding, and video coding
5. Selected system examples of image coding (JPEG, MPEG-2, MPEG-4 AVC, HEVC) and audio coding (MPEG audio, GSM codecs)

Module type

Elective compulsory module in the Master’s program “Electrical Engineering and Information Technology,” specializations “Nano and Quantum Technology” and “Computer Engineering.” Elective compulsory module in the Master’s program in Industrial Engineering and Management, recommended specialization “Information Technology,” reference module number: MB-319.
Can only be taken as an additional module if you are enrolled in the Master’s program in Electrical Engineering and Information Technology.