Schedule

Below, you can find an overview of all lectures and tutorials with topics and required reading. An overview of the schedule with rooms can also be found on TimeEdit.

Lecture & Tutorial Plan

Session Time Topics Reading
Lecture 1 September 01, 13–15
  • Introduction and overview of the course.
  • Model-based versus data-driven inference.
  • Introduction to detection theory.
  • Orthodox approach to detection: Neyman-Pearson theorem, probability of detection and false alarm, and receiver operating characteristic (ROC).
  • Bayesian approach to detection: Bayesian cost. Probability of error.
  • Kay-II Chapter 1, Sections 3.1-3.7
Lecture 2 September 03, 08–10
  • Detection of deterministic vector signals in white Gaussian noise.
  • Dealing with colored noise. Pre-whitening.
  • Matched filter.
  • Kay-II Sections 4.1-4.4, 4.6
  • Supplementary material
Lecture 3 September 05, 13–15
  • Detection of random Gaussian signals in Gaussian noise.
  • Optimal detector and its performance.
  • Special cases and examples.
  • Kay-II Sections 5.1-5.4, 5.6-5.7
  • Supplementary material
Lecture 4 September 08, 13–15
  • Contrasting the orthodox and Bayesian approach.
  • Detection with M>2 hypotheses.
  • Deterministic vector signals in white Gaussian noise.
  • Special cases and examples.
  • Kay-II Sections 3.8, 4.5, 4.7
Tutorial 1 September 08, 15–17
  • Walk-through by the instructor: Monte-Carlo simulation + Kay-II problem 3.6
  • Problems, from Kay-II: 1.2, 3.4, 3.14, 4.6, 4.8, 4.15, 4.16, 4.19
  • Problems, from the "additional problems" document: #1
Lecture 5 September 10, 08–10
  • Introduction to estimation theory.
  • Orthodox versus Bayesian approach.
  • Performance metrics: bias, variance, MSE, BMSE.
  • Cramer-Rao bound (CRB) and efficiency.
  • Kay-I Chapters 1-2, Sections 3.1-3.5
Lecture 6 September 12, 13–15
  • CRB for vector parameters.
  • Slepian-Bang's formula.
  • Nuisance parameters and decoupling.
  • Application example: source localization using time-of-arrival measurements.
  • CRB for the linear model with Gaussian noise.
  • Kay-I Sections 3.6-3.9, 3.11
Tutorial 2 September 17, 08–10
  • Walk-through by the instructor: Kay-I problem 1.4
  • Problems, from Kay-II: 5.2, 5.3, 5.10, 5.14, 5.18
  • Problems, from Kay-I: 1.1, 1.5, 2.9
  • Problems, from the "additional problems" document: #2
Tutorial 3 September 19, 13–15
  • Walk-through by the instructor: Kay-I: 3.15, 6.15
  • Problems, from Kay-I: 3.1, 3.19, 3.9, 4.11, 6.1, 6.3, 6.16
  • Problems, from the "additional problems" document: #3
Lecture 7 September 22, 13–15
  • MVU estimator for linear model with Gaussian noise.
  • BLUE estimator for linear model with arbitrary noise.
  • Application example: tapped-delay line identification.
  • Application example: two-tone model with sinusoids in noise.
  • Kay-I Chapters 4, 6
Lecture 8 September 26, 13–15
  • Maximum-likelihood estimation.
  • Asymptotic efficiency.
  • Parameter transformations.
  • Kay-I Sections 7.1-7.8, 7.10
Lecture 9 September 29, 13–15
  • Linear and non-linear least-squares.
  • Separable models.
  • Method of moments.
  • First-order approximations.
  • Kay-I Sections 8.1-8.4, 8.9, 9.1-9.5
Tutorial 4 October 01, 08–10
  • Walk-through by the instructor: periodogram example + Kay-I problem 7.3
  • Problems, from Kay-I: 7.1, 7.10, 7.20
  • Problems, from the "additional problems" document: #4
  • Backup time.
Lecture 10 October 03, 13–15
  • Bayesian estimation.
  • MMSE and LMMSE estimators.
  • Nuisance parameters in the Bayesian and orthodox paradigms.
  • Kay-I Sections 10.1-10.7, Chapter 11, Sections 12.1-12.3, 12.5
Lecture 11 October 06, 13–15
  • Bayesian estimation.
  • MMSE and LMMSE estimators.
  • Nuisance parameters in the Bayesian and orthodox paradigms.
  • Kay-I Sections 10.1-10.7, Chapter 11, Sections 12.1-12.3, 12.5
Tutorial 5 October 06, 15–17
  • Walk-through by the instructor: Kay-I problem 7.9
  • Problems, from Kay-I: 8.1, 8.3, 8.5, 8.7, 9.1, 9.7
Tutorial 6 October 08, 08–10
  • Walk-through by the instructor: Kay-I problem 11.3
  • Problems, from Kay-I: 10.6, 10.9, 10.11, 11.9, 11.16, 12.2
  • Problems, from the "additional problems" document: #5
Lecture 12 October 10, 13–15
  • Detection of signals with unknown parameters.
  • Generalized likelihood ratio test (GLRT).
  • GLRT for linear model with Gaussian noise.
  • Kay-II Sections 6.1-6.4, 7.1-7.6
Tutorial 7 October 13, 13–15
  • Walk-through by the instructor: 7.2
  • Problems, from Kay-II: 6.6, 6.10, 7.1, 7.3, 7.9, 7.7, 7.10
Tutorial 8 October 20, 13–15
  • Walk-through by the instructor: 8.7
  • Problems, from Kay-II: 7.23, 7.25, 8.8, 9.4, 9.11, 9.13
  • Problems, from the "additional problems" document: #6