Digital Signal Processing
Code  Completion  Credits  Range  Language 

F7AMBCZS  Z,ZK  5  2P+2C  English 
 Garant předmětu:
 Václava Piorecká
 Lecturer:
 Hana Děcká, Vladimír Krajča, Václava Piorecká, Marek Piorecký, Jan Štrobl
 Tutor:
 Hana Děcká, Vladimír Krajča, Václava Piorecká, Marek Piorecký, Jan Štrobl
 Supervisor:
 Department of Biomedical Technology
 Synopsis:

The course deals with the following topics  characteristics of signals, linear time invariant systems (LTI), stationary, nonstationary signals, deterministic, ergodic and stochastic processes, description of signals in continuous and discrete domains, A/D conversions and converters, sampling and quantization problems, aliasing and Nyquist's theorem, noise suppression and data preprocessing, fast and discrete Fourier transforms, efficient FFT estimation methods, other discrete transforms: ztransform, its properties and applications in DSP, inverse transforms, poles and zeros of the system, frequency response, correlation and convolution, introduction to digital filter design, FIR and IIR filters and adaptive filters, spectral analysis and spectrum estimation methods, current methods of analysis in time and frequency domain, coherence and phase characteristics, parametric and nonparametric methods, periodogram and AR spectrum.
 Requirements:

Requirements for credit: Compulsory attendance at the exercise, max. 1 absence excused. Submitted and processed assigned tasks in the given terms, the maximum number of points for submitted tasks is 50.
Assessment of the exam: The exam is implemented by a written test and an oral exam in the form of a discussion over the test results. A total of 50 points are possible.
Without obtaining credit and recording the credit in KOS, it is not possible to implement the exam.
The evaluation is carried out according to the ECTS scale on the basis of the results of the semester and exam test.
 Syllabus of lectures:

1. Introduction to digital signal processing (DSP). Motivation, application areas. Overview of basic operations. Convolution, correlation, digital filtering, discrete transforms. Linear time invariant systems (LTI).
2.Characteristics of random signals and their estimation. Confidence intervals, mean, standard deviation, median.
3. Stochastic processes, ergodic, stationary, nonstationary. AR, MA, ARMA data models.
4.A/D and D/A conversion. Sampling, uniform and nonuniform quantization, oversampling, antialiasing filtering. Nyquist theorem. Conversion errors. Signal conditioning. Aliasing. Analog filtering. Trends. Digital data formats and implications of quantization.
5.Discrete transforms, sequences and systems. Discrete Fourier Transform (DFT). Computational complexity. Gibbs effect.
6. Fast Fourier Transform (FFT). Inverse transform. Properties of the DFT. Decimation in time, decimation in frequency domain. FFT algorithm, „butterfly“. Techniques for increasing the efficiency of FFT computation for real signals.
7. ztransform and its application in DSP. Features. Poles and zeros, complex plane. Frequency response. Stability of linear systems. Applications in filter design.
8.Digital filtering. Finite Impulse Response (FIR) filters. Window method. Remez exchange algorithm.
9. IIR filters (Infinite Impulse Response). Design methods. Quantization errors of coefficients. Examples of EEG signal filtering.
10.Adaptive and median filters.
11. Spectral analysis. Spectral power density. Basic methods. Parametric and nonparametric methods.
12. Periodogram and methods of its calculation. Aliasing, spectral leakage. Spectrum reciprocity, coherence and phase, chordance. Spectral analysis and signal synthesis using FFT. Absolute and relative spectrum. Disadvantages of the periodogram. Windowing.
13.Modern methods of spectrum estimation... Practical problems of spectrum estimation. Parametric models. YuleWalker equations, LDR algorithm. Burg and Marple algorithm. Phase delay estimation.
14. Graphical representation of spectral analysis results. Topographic mapping of brain activity. Compressed spectral analysis (CSA). 3D spherical splines. Bispectrum.
 Syllabus of tutorials:

1. Basic operations: unit jump and impulse response, convolution, correlation, digital filtering, discrete transform (1)
2.Basic operations: unitjump and impulse response, convolution, correlation, digital filtering, discrete transform (2)
3.Characteristics of random signals and their estimation
4.AR, MA and ARMA data models
5.A/D and D/A conversion
6.Discrete transforms (DFT in particular)
7. ztransformation
8.Digital filtering of biosignals
9. FIR filters (Finite Impulse Response)  design and properties
10. IIR filters (Infinite Impulse Response)  design and properties
11.Design and testing of adaptive filters
12. Spectral analysis of biosignals
13.Spectrum estimation methods
14. Graphical display of spectral analysis results
 Study Objective:
 Study materials:

Mandatory:
1.MEDDINS, Bob. Introduction to digital signal processing. Oxford: Newnes, ©2000. ix, 161 s. Essential electronics series. ISBN 0750650486
2.IFEACHOR, Emmanuel C. a Barrie W. JERVIS. Digital signal processing: a practical approach. Harlow: AddisonWesley, c1993. ISBN 020154413X.
3.PROAKIS, John G. a Dimitris G. MANOLAKIS. Digital signal processing. 4th ed. Harlow: Pearson Education Limited, c2014. ISBN 9781292025735.
Recommended::
1.INGLE, Vinay K. a John G. PROAKIS. Essentials of digital signal processing using MATLAB. 3rd ed. s.l.: Cengage Learning, c2012. ISBN 9781111427382.
2.CHAPRA, Steven C. Applied numerical methods with MATLAB: for engineers and scientists. 3rd ed. New York: McGraw  Hill, 2012. McGrawHill international edition. ISBN 9780071086189.
 Note:
 Timetable for winter semester 2022/2023:
 Timetable is not available yet
 Timetable for summer semester 2022/2023:
 Timetable is not available yet
 The course is a part of the following study plans:

 Prospectus  magisterský (!)
 Biomedical and Clinical Engineering (compulsory course)