Guide in PDF format
Logo

Estimation and Detection Methods

Code: 45640
Credits: 6
2026/2027
Degree programme Type Course
Telecommunication Engineering OB 1

Contact lecturer

Name :
José A. Lopez Salcedo
Email :
jose.salcedo@uab.cat

Teaching staff

Gonzalo Seco Granados

Teaching staff (external to UAB)

Xing Liu

Group languages

You can consult this information at the end of the document.

Prerequisites

Basic knowledge of statistical signal processing is recommended.

Objectives

This course provides a overview of detection and estimation theory, including both classical and modern data-driven learning frameworks. The course first introduces the topic of detection theory, spanning fundamental performance criteria through sequential and classification-based methods. The course then covers classical estimation methods, including performance bounds and recursive least squares, before progressing to Bayesian estimation techniques such as the Kalman filter. Building on this foundation, the course introduces sparsity-aware estimation and compressed sensing, followed by Bayesian learning techniques and probabilistic graphical models. Together, these topics equip students with the theoretical tools and algorithmic techniques needed to design and analyze estimation and detection systems for modern signal processing applications.


Learning outcomes

  • (CA04) Implement digital receivers for use in telecommunication applications, both fixed and mobile.
  • (CA05) Apply signal processing techniques in interdisciplinary contexts, such as bioengineering, finance, or industrial processes.
  • (KA04) Compare classical estimation methods and Bayesian estimation methods, identifying their differential characteristics.
  • (KA05) List the main signal detection techniques and their application to problems with full or partial knowledge of the parameters involved.
  • (SA07) Apply classical and Bayesian estimation methods, as well as detection methods, to solve problems involving signal processing in digital receivers.
  • (SA08) Design signal processing techniques in radio navigation, positioning and radar systems.
  • (SA09) Evaluate the performance of estimation and detection methods through theoretical analysis and simulation.

Contents

1. Detection methods

  • Principles of detection theory
  • Detection performance.
  • Detection criteria for completely known statistics.
  • Detection criteria in the presence of unknown parameters.
  • Sequential detection.
  • Classification


2. Classical estimation methods

  • Principles of estimation theory.
  • Estimation in the presence of nuisance parameters.
  • Recursive least squares (RLS).
  • Performance lower bounds.

3. Bayesian estimation methods

  • Bayesian estimators (general and linear).
  • Kalman filter.
  • Bayesian bounds.


4. Sparsity-aware estimation and learning

  • Sparse signal representation
  • Characterization of l0, l1, l2 norms
  • Algorithms: LASSO, greedy algorithms, iterative shinkrage/thresholding algorithms
  • Compressed sensing


5. Bayesian learning

  • EM algorithm
  • Sparse Bayesian Learning (SBL)
  • Gaussian mixture models
  • Particle filtering


6. Probabilistic graphical models

  • Graphical models
  • Bayesian networks
  • Factor graphs

Learning activities and methodology

Title Hours ECTS Learning outcomes
Type: Autonomous
Student individual work: preparation of practical sessions 20 0.8 CA04, CA05, KA04, SA07, SA08, SA09
Student individual work: study and problem resolution 70 2.8 CA04, CA05, KA04, KA05, SA07, SA08
Type: Guided
Laboratory sessions 15 0.6 CA04, KA04, SA07, SA08, SA09
Theory lectures 30 1.2 CA05, KA04, KA05, SA07, SA08, SA09
Type: Supervised
Tutoring 9 0.36 CA05, KA05, SA07, SA08

Classroom activities

  • Theory classes: presentation of theoretical content.
  • Laboratory sessions: application of the techniques presented in the theory classes to different real systems and implementation with different simulation software.
  • Partial and final exams.


Autonomous activities

  • Study of the theoretical and practical contents of the subject.
  • Problem solving and preparation of assignments with solutions to some sets of problems.
  • Exam preparation.
  • Practical work: carrying out and deepening of laboratory practices.


Annotation: within the schedule set by the centre or degree programme, 15 minutes of one class will be reserved for students to evaluate their lecturers and their courses or modules through questionnaires.

Assessment

Continuous assessment activities

Title Weight Hours ECTS Learning outcomes
Exam 1 35% 2 0.08 CA05, KA04, KA05, SA07, SA08, SA09
Exam 2 35% 2 0.08 CA05, KA04, SA08, SA09
Laboratory 30% 2 0.08 CA04, CA05, KA04, SA07, SA08, SA09

The final grade for the subject is obtained as the weighted average of the grades obtained in the continuous assessment activities.


Second-chance exam

The student has the option of taking a second-chance exam, which will be held once the face-to-face classes have ended, during the retake exam period set by the school. In the second-chance exam, the student may retake the part corresponding to exam 1, the part corresponding to exam 2, or both at the same time. In any case, the grade for the second-chance exam, whether corresponding to the retake of exam 1 or exam 2, will replace the grade that the student has obtained in the exam that is being retaken.


Once the grade for the second-chance exam has been replaced, the final grade for the subject is also obtained as the average of the grades for the exams (retaken, if applicable) and the laboratory grade.


Academic integrity and verification of authorship

The teaching staff reserves the right to call for an interview or an individual oral test of contrast when there are indications of copying or lack of authorship in an assessable activity. The impossibility of satisfactorily accrediting authorship may have effects on their qualification, in accordance with current academic regulations.


Consideration of "Not Assessable"

The final grade will be "Not Assessable" only when the student does not take any exam, neither those of the continuous assessment nor the retake.


Consideration in the case of copying or plagiarism

Without prejudice to other disciplinary measures that may be deemed appropriate, and in accordance with current academic regulations, tests or reports where the student has committed irregularities, such as plagiarism, cheating, copying, allowing copying, etc., which could lead to a variation in the qualification, will be graded with a zero.

Bibliography

  • S. M. Kay, Fundamentals of statistical signal processing. Estimation theory, vol. I, Prentice-Hall, 1993.
  • S. M. Kay, Fundamentals of statistical signal processing. Detection theory, vol. II, Prentice-Hall, 1998.
  • S. M. Kay, Fundamentals of statistical signal processing. Practical algorithm development, vol. III, Pearson, 2013.
  • H. L. Van Trees, K. L. Bell, Bayesian bounds for parameter estimation of nonlinear filtering/tracking, IEEE Press, 2007.
  • M. S. Grewal, A. P. Andrews, Kalman filtering: theory and practice using Matlab, John Wiley & Sons, 2001.
  • S. Theodoridis, Machine Learning. From the Classics to Deep Networks, Transformers, and Diffusion Models, 3rd Ed., Academic Press, 2024.

Software

The laboratory sessions will make use of the Matlab software.

Course groups and languages

The information provided is provisional until November 30. After this date, you will be able to consult the language of each group through this link. To access the information, you will need to enter the course CODE

Type of teaching Group Language Semester Shift
(TEmRD) Teoria (màster RD) 1 English first semester afternoon
(PLABmRD) Pràctiques de laboratori (màster RD) 1 English first semester afternoon