 |
| Detailed information |
| Original study plan |
Master's programme Electronics and Information Technology 2025W |
| Learning Outcomes |
Competences |
| Students know and understand the fundamental parameter estimation methods, the basics of optimum filters, adaptive filters and Kalman filters, both in qualitative and mathematical terms, and can apply them to in-depth problems and generalize the results obtained.
|
|
| Skills |
Knowledge |
Students are able to
- assess the performance of classical estimation methods (MVU, BLUE, ML, LS), and Bayesian methods (MAP, MMSE, LMMSE) in various applications. K4, K5
- derive and apply these estimators for applications such as, amplitude estimation, frequency estimation, power estimation, signal extraction, system identification, and data estimation. K2, K3, K5, K6
- explain the underlying principles of optimum filters and their applications. K2
- apply and assess Wiener Filters and Least Squares Filters for various applications. K3, K5
- apply and asses the LMS (Least Mean Squares) and the RLS (Recursive Least Squares) algorithm for various applications, like system identification, inverse system identification, noise cancellation, and signal prediction. K3, K4, K5
- explain the principles and limitations of adaptive filtering techniques. K2
- apply and analyze the Kalman filter algorithm for the linear Gauss-Markov model. K3, K4
- apply and analyze the extended Kalman filter algorithm for nonlinear state space models. K3, K4
|
- Parameter Estimation
- Classical Methods: MVU, BLUE, ML, LS
- Bayesian Methods: MAP, MMSE, LMMSE
- Applications: amplitude estimation, frequency estimation, power estimation, signal extraction, system identification, data estimation
- Optimum Filters
- Wiener Filters
- Least Squares Filters
- Applications: system identification (channel estimation), inverse system identification (e.g. for channel equalization), noise reduction, linear prediction (e.g. for voice signals)
- Adaptive Filters
- LMS (Least Mean Squares) algorithm
- RLS (Recursive Least Squares) algorithm
- Kalman Filters
- Standard Kalman Filter
- Extended Kalman Filter
- Applications
|
|
| Criteria for evaluation |
Written or oral exam (depending on the number of subscribed students)
|
| Methods |
Lecture using slides and blackboard, Matlab based presentations
|
| Language |
English and French |
| Study material |
- Lecture Slides
- S. Kay, Fundamentals of Statistical Signal Processing: Estimation Theory, Prentice Hall, Rhode Island 1993.
- D.G. Manolakis, V.K. Ingle, S.M. Kogon, Statistical and Adaptive Signal Processing, Artech House, 2005.
|
| Changing subject? |
No |
| Further information |
This lecture and the accompanying exercise course form an inseparable didactic unit. The presented learning outcomes are achieved through the interaction of the lecture and the exercise course.
|
| Earlier variants |
They also cover the requirements of the curriculum (from - to)
489INTEOASV17: VL Optimum and Adaptive Signal Processing Systems (2017W-2022S)
489WSIVOASV14: VL Optimum and Adaptive Signal Processing Systems (2014W-2017S)
|
|
