Courses Currently Offered in English

Course Version

Course Code:EASP
Version Number:1
Course Name:Adaptive Signal Processing
Credit Units:4
ECTS:6
Cost Units:30
Examination type (E-with exam):E
Grading threshold:5
Initiation semester:09L
Person responsible:dr hab. inż. Konrad Jędrzejewski
Description:ICT programme taught in English

Hours per week

Class typeHours
W2
L1
P1

Class types: W - lecture, C - tutorial, L - laboratory, P - project

Prerequisites

---

Prerequitise types: W - required, Z - recommended

Similar Courses

---

Last Course Instances

Semester codeInstance codeLecturerInstituteMax. number of students
18ZAdr hab. inż. Konrad JędrzejewskiPE30

Thematic Classification

Class CodeClass name (in Polish)
ANGLAll Courses in English (A)

Conspectus

Summary (in Polish)The purpose of the course is to present modern methods of digital adaptive signal processing. The course covers foundations of discrete random signal analysis, estimation theory, optimal signal processing and adaptive signal processing algorithms. The practical applications of adaptive signal processing techniques and algorithms are presented and discussed. Students implement selected adaptive algorithms and investigate their properties during lab exercises.
Lecture contents
  • Introduction to adaptive signal processing. Basic ideas and general principles of adaptation in signal processing. Typical applications of adaptive signal processing: identification, inverse modelling, interference and noise cancellation, echo cancellation. (4h)
  • Review of methods of discrete random signal analysis. Spectral representations of discrete random signals. Response of linear systems to random signals. Discrete stochastic signals modelling, AR, MA and ARMA processes, Wold?s theorem. Stationarity and stability of the modelling system. (4h)
  • Foundations of the estimation theory. General properties of estimators. Classical and Bayesian estimation. Basic estimation methods under different assumptions on prior information of signal, maximum a posteriori (MAP) estimation, maximum likelihood (ML) estimation, minimum mean-squared error (MMSE) estimation. Cramer-Rao lower bound. (4h)
  • MMSE estimation. Estimation of signals. Optimal linear filter design. Normal equations. Solution of the normal equations. Optimal linear prediction. Application of linear prediction to identification of AR model parameters. (4h)
  • Adaptive filters and filtering systems. LMS adaptive algorithm. Performance of the LMS algorithm, stability and convergence. Examples of application. (6h)
  • Adaptive gradient algorithms. LMS-Newton algorithm. Recursive least-squares (RLS) algorithm. Exponentially weighted RLS algorithm. Computational complexity of RLS algorithms. RLS versus LMS. Examples of application. (6h)
  • Lattice structure adaptive filters. Forward and backward prediction. Properties of the lattice filter. Estimation of reflection coefficients. Gradient adaptive lattice (GAL) algorithm. (2h)
Laboratory contents
  1. Modeling and spectrum analysis of discrete random signals.
  2. Implementation and performance analysis of LMS algorithm.
  3. Implementation and performance analysis of RLS algorithm.
  4. Adaptive cancellation of noises and interferences.
  5. Selected applications of adaptive algorithms in telecommunication.
Project contentsStudents implement adaptive signal processing algorithms and investigate or compare their performance and properties.
ReferencesBasic
  1. P. M. Clarkson, Optimal and adaptive signal processing, CRC Press, 1993.
Optional
  1. S. Haykin, Adaptive Filter Theory, Third Edition, Prentice Hall, 1996.
  2. B. Widrow, S. D. Stearns, Adaptive signal processing, Prentice-Hall, Englewood Cliffs, 1985.
  3. M.L Honig, D.G. Messerschmitt, Adaptive Filters Structures Algorithms and Application, Kluwer Academic Publishers, 1984.
SummaryCelem kursu jest zapoznanie studentów z nowoczesnymi cyfrowymi metodami adaptacyjnego przetwarzania sygnałów. W ramach wykładu omawiane są podstawy analizy sygnałów stochastycznych, teorii estymacji, optymalnego przetwarzania sygnałów oraz wynikające z nich algorytmy adaptacyjnego przetwarzania sygnałów. Przedstawiane i omawiane są przykłady zastosowań praktycznych metod i algorytmów adaptacyjnego przetwarzania sygnałów. W ramach zajęć laboratoryjnych studenci samodzielnie implementują wybrane algorytmy i badają symulacyjnie ich właściwości.