|Fees||Look up fees|
Topics include: State estimation algorithms (Kalman and Wiener filtering); parameter estimation algorithms (Least Squares, Maximum Likelihood, Maximum a Posteriori) and their adaptive versions.
Other topics to be selected from: system identification, spectral analysis, nonlinear filtering; hidden Markov model signal processing; expectation maximization algorithm; distributed detection and estimation; information-theoretic aspects of estimation and detection (Cramer Rao bound, Divergence measures).
Intended learning outcomes
Intended Learning Outcomes (ILOs)
The aim of this subject is to give students a rigorous introduction to the mathematical tools commonly employed in statistical signal processing.
Having completed this subject it is expected that the student be able to:
1. Use the principle of orthogonality to derive least squares system identification and minimum mean square error state estimation algorithms
2. Use probability theory to analyse properties of system identification and filtering algorithms
3. Formulate and solve optimal system identification and filtering problems
- Ability to apply knowledge of basic science and engineering fundamentals;
- In-depth technical competence in at least one engineering discipline;
- Ability to undertake problem identification, formulation and solution;
- Ability to utilise a systems approach to design and operational performance;
- Expectation of the need to undertake lifelong learning, capacity to do so;
- Capacity for independent critical thought, rational inquiry and self-directed learning;
- Intellectual curiosity and creativity, including understanding of the philosophical and methodological bases of research activity;
- Openness to new ideas and unconventional critiques of received wisdom;
- Profound respect for truth and intellectual integrity, and for the ethics of scholarship.
Last updated: 2 December 2019