Signal Processing 2
Course No. 389.170
2017W, VU, 3.0h, 4.5EC
- Welcome to Signal Processing 2 (389.170), winter term 2018/19.
- The course starts on October 4, 2018, at 11:00 in lecture hall EI4.
- Check the TISS news for further announcements.
Lecturer: Gerald MATZ, Ao.Univ.Prof. Dipl.-Ing. Dr.techn.
Assistant: Thomas DITTRICH, Univ.Ass. Dipl.-Ing.
Course type: VU – lectures and exercises
Credits: 4.5 ECTS
Course language: English
Time and place: Thursday 11:00-12:00 (primarily exercises, otherwise lectures) and Friday 10:30-12:00 (lectures), both in lecture hall EI4
Lecture notes: Detailed lecture notes will be made available
Corresponding TISS page: TISS – 389.170
1. One random variable:
Cumulative distribution function (cdf) and probability density function (pdf), discrete random variables, transformation of random variables, conditional cdf and pdf, moments, characteristic function, inequalities, conditional expectation, special distributions
2. Two random variables:
Joint cdf and pdf, discrete random variables, transformation of random variables, conditional cdf and pdf, moments, correlation, covariance, statistical independence, orthogonality and uncorrelatedness, characteristic function, conditional expectation, special distributions, complex random variables, circular symmetry
3. Random vectors:
cdf and pdf, discrete random vectors, transformation of random vectors, conditional cdf and pdf, mean, correlation matrix, covariance matrix, statistical independence, orthogonality and uncorrelatedness, characteristic function, conditional expectation, special distributions, complex random vectors, Karhunen-Loeve decomposition, whitening transformation, innovations representation, MMSE estimation, LMMSE estimation (Wiener filter), ML estimation
4. Random signals (stochastic processes):
pdf, stationarity, second-order description (mean, autocorrelation function), cyclostationarity, power spectral density, cross-correlation function und cross-power spectral density, effects of linear systems, time averages und ergodicity, discrete-time random signals, special random signals, complex random signals and circular symmetry, whitening filter, innovations filter, Wold decomposition, AR, MA and ARMA processes, LMMSE estimation (Wiener filter), linear prediction
The theory presented in the lectures is illustrated by problems in exercise classes.
The problem sets will be available online in TISS.
Students who are admitted may register for the written exam via TISS. There are typically four exam dates per year (see examination dates). Please de-register in TISS or via email in case you cannot attend an exam you registered for.
The written exam consists of four problems and lasts three hours. A copy of the lecture notes will be provided during the exam. Additionally, a calculator may be used. The use of other documents or electronic devices is not permitted.
The result of the written exam will be sent to your generic student address (e<matriculation_number>@student.tuwien.ac.at) in a signed email.
Specific questions regarding exam problems and exercise problems can be discussed during office hours, which are scheduled as follows:
- October – January: Monday from 14:00 to 14:30 in room CG EG14 (except on holidays; further exceptions will be announced via TISS)
- February – September: by appointment
Additional exercise problems (available in german only): Download (PDF)
Formulary (included in the lecture notes since winter term 2010/11): Download (PDF)