EEE515Z: Tentative course schedule

24 periods in total: 20 Lectures, 4 tutorial-style sessions. 
Course breakdown on a per-lecture basis 
Introduction, elements of matrix algebra (quadratic forms, Gram determinant) 
Elements of matrix algebra (differentiation, orthogonal matrices, QR factorisation, SVD, projections and rotations 
Optimisation, multivariate methods, principal components 
Hypothesis testing: Neyman-Pearson detectors 
Framework, decision rules, classifying tests, testing of binary hypotheses 
Neyman-Pearson lemma, ROC curves, sufficiency in hypothesis testing 
Composite binary hypotheses, UMP tests, Karlin-Rubin theorem 
Invariance, UMP invariant tests 
Matched filters, CFAR matched filters, locally most powerful tests 
Hypothesis testing: Bayes detectors 
10  Risk, Simple binary hypothesis Bayes detector 
11  General formulation, likelihood ratios and posterior probabilities, continuous-time hypotheses 
Minimum variance unbiased estimation/Maximum likelihood estimation 
12  MVUB estimators, BLU estimators, Cramer-Rao lower bound 
13  Efficient estimators, ML estimation, asymptotic properties, sufficiency, invariance 
Bayes estimators 
14  Bayes risk, minimax estimators, computing Bayes estimators, Bayes sufficiency and conjugate priors 
15  MVN model, Gauss-Markov theorem, linear statistical model, sequential Bayes 
Minimum mean-squared error estimation 
16  Conditional expectation and orthogonality, MMSE and LMMSE estimators, linear prediction 
17  Kalman filtering 
Least Squares 
18  Linear model, least squares solution, performance, weighted LS 
19  Constrained LS, underdetermined LS, structured correlation matrices 
20  Overview of principles and techniques 

BackBack to SSP Home Page
Last modified: Tue Feb 3 15:07:07 GMT 1998