Course syllabus

# Optimal och adaptiv signalbehandling

Optimum and Adaptive Signal Processing

## EITN60, 7,5 credits, A (Second Cycle)

## General Information

## Aim

## Learning outcomes

## Contents

## Examination details

## Admission

## Reading list

## Contact and other information

Optimum and Adaptive Signal Processing

Valid for: 2019/20

Decided by: PLED BME

Date of Decision: 2019-03-28

Elective for: BME4-sbh, C4, D4-ssr, E4-ss, E4-bg, F5, F5-ss, MWIR2, Pi4-ssr

Language of instruction: The course will be given in English

The course provides basic knowledge in statistical signal processing and the theory of optimal methods and how they can be applied. The course presents signal processing methodology and solutions to problems where digital systems tune in automatically and adapt to the environment. The student is given enough theoretical and practical knowledge to independently be able to formulate the mathematical problem, solve it and implement the solution for use with real-life signals.

Knowledge and understanding

For a passing grade the student must

- have knowledge about and understand the main concepts in optimum and adaptive filter theory
- be able to apply the most commonly used methods to real problems and real-life signals (Matlab-level)
- be able to formulate mathematical problems based on described situations

Competences and skills

For a passing grade the student must

- be able to explain the main principles behind the most common adaptive methods (LMS and RLS)
- be able to explain/calculate the convergence and stability properties for these methods
- be able to sketch the most common block diagrams/structures used for adaptive filters and their properties
- be able to set parameters needed to make the algorithms work
- be able to foresee the consequences for the algorithms when implemented in fixed-point arithmetic
- be able to implement adaptive filters

Judgement and approach

For a passing grade the student must

- have the ability to analyze, evaluate and implement adaptive algorithms, and be able to interpret and describe the principles which they are based on.
- have the insight that many different technical problems can be solved using the same methods.

*Optimum filtering*

- Wiener filters
- Linear prediciton
- The Levinson-Durbin algorithm

*Basics about adaptive filters*

- From optimal to adaptive filters
- Cost functions, minimization problems and iterative procedures
- Convergence and tracking capability, implementation aspects
- Strategies for how to connect adaptive filters

*The LMS family*

- Principle and derivation
- Convergence analysis and parameter settings
- Variants including Normalized LMS, Leaky LMS, Fast LMS and Sign LMS
- Matlab implementation
- LMS in fixed-point arithmetic
- Principle and derivation
- Parameter settings

*The RLS family*

- Aspects when used
- Matlab implementation
- Numerical properties

Grading scale: TH - (U,3,4,5) - (Fail, Three, Four, Five)

Assessment: The grade is based on the exam in the end of the course.

The examiner, in consultation with Disability Support Services, may deviate from the regular form of examination in order to provide a permanently disabled student with a form of examination equivalent to that of a student without a disability.

Parts

Code: 0119. Name: Written Exam.

Credits: 6. Grading scale: TH. Assessment: Written Examination.

Code: 0219. Name: Project.

Credits: 1,5. Grading scale: UG. Assessment: Project Report.

Required prior knowledge: ESS040, EITF75 Digital Signal Processing or ETI265, EITA50 Signal Processing in Multimedia or EITF15 Signal processing - theory and applications.

The number of participants is limited to: No

The course overlaps following course/s: ETTN05, ETT042

- Haykin S: Adaptive Filter Theory, Fifth Edition. Pearson, 2014, ISBN: 0-273-76408-X.

Course coordinator: Frida Sandberg, frida.sandberg@bme.lth.se

Course homepage: http://www.bme.lth.se/course-pages/optimal-och-adaptiv-signalbehandling/optimum-and-adaptive-signal-processing/

Further information: Exercises 14 h, computer exercises 14 h and laboratory work 2 x 4 h. The course might be given in English.