Signal Processing and Optimization Methods in Data Mining
Signalverarbeitung und Optimierungsmethoden für Data Mining
6 CPWinterENno longer offered
Prerequisites
Recommended, as written
Courses on Algebra and Analysis
Contents
The first part of the course introduces foundations of signal processing methods in electrical power engineering. Introducing lectures provide ideas of discrete representations of continuous signal with basic functions and signal orthogonal projection. Next stage of the course concerns continuous representations of deterministic signals. Introduced definitions allow passing into nonparametric time-frequency representations. Then, follow the time-scale representations with wavelets. Backgrounds of statistical signal processing are introduced with outlined parameter estimation, optimal filtering, linear modeling and estimation. Finally, various methods of parametric spectrum estimation are outlined. Examples of application in electrical engineering are presented throughout the course.
Learning outcomes
Students will get detailed knowledge about signal processing issues and notions of optimization methods, with practical examples and exercises. In the optimization part students will learn foundations of optimization methods and their numerical application in engineering. They will be skilled to optimize various types of practical industrial problems.
Assessment
• Written examination, 90 minutes
Literature (2)
- S. Haykin, B. Van Veen – Signals and Systems, John Wiley & Sons, Inc., 1999, C.W. Therrien – Discrete Random Signals and Statistical Signal Processing, Englewood Cliffs, Prentice Hall, 1992, web based lecture script
- K.P. Chong, S.H. Zak: An Introduction to Optimization, 2nd edition, New York, John Wiley, 2001, J.F. Bonnans: Numerical optimization: theoretical and practical aspects, Springer-Verlag, 2003, M. Asghar Bhatti: Practical Optimization Methods, Berlin, Springer-Verlag 2000
Remarks
The lecture will be not offered in winter semester 2017/18.
Source: BTU's module description · retrieved 19 Oct 2026Original