Course detail

Digital Signal Processing

FEKT-BCZSAcad. year: 2011/2012

One-dimensional and two-dimensional discrete signals and systems. Z-transform. Discrete Fourier transform, FFT. State-space canonic structures, serial and parallel forms. Design of type FIR and IIR digital filters. Multi-rate systems. Filter banks. Homomorphous processing of filters. Complex and real cepstrums. Application of cepstrums to speech and image processing. Signal quantization in discrete systems. Realization of digital filters and FFT in digital signal procesors.

Language of instruction

Czech

Number of ECTS credits

5

Mode of study

Not applicable.

Guarantor

prof. Ing. Jiří Mišurec, CSc.

Department

Department of Telecommunications (UTKO)

Learning outcomes of the course unit

The student will have a good grasp of algorithms for digital signal processing, s/he will be able to apply them and model them by means of the MATLAB program. S/he will be knowledgeable regarding algorithm realization on microprocessors and digital signal processors, inclusive of all the problems involved in the realization.

Prerequisites

The subject knowledge on the secondary school level is required.

Co-requisites

Not applicable.

Planned learning activities and teaching methods

Teaching methods depend on the type of course unit as specified in the article 7 of BUT Rules for Studies and Examinations.

Assesment methods and criteria linked to learning outcomes

Requirements for completion of a course are specified by a regulation issued by the lecturer responsible for the course and updated for every.

Course curriculum

1. Discrete signals - basic discrete signals, classification of one dimensional discrete signals.
2. Discrete signals - multi dimensional discrete signals, correlation of discrete signals.
3. Discrete systems - initial conditions, discrete systems as block diagrams.
4. Discrete systems - classification of discrete systems, linear time invariant system, combination of discrete time invariant systems, causallity and stability of time invariant systems, FIR and IIR systems.
5. State diagram of linear time invariant system.
6. Z- transform and using.
7. Frequency analysis of discrete signals - time discrete Fourier line, spectral power, FT of discrete aperiodic signal, feature of FT, cepstrum.
8. Frequency characteristics of linear time invariant system, frequency filters.
9. Discrete FT definition, features, vector form.
10. Inverse systems and deconvolution - reciprocal disrete system.

Work placements

Not applicable.

Aims

The aim of the course is to provide comprehensive explanation of the basic theory of digital signal processing, with emphasis on applications in microprocessor technology. Extra emphasis is laid on methods of spectral analysis and digital linear and non-linear filtering. The course is concluded with a discussion of quantizing effects and of the realization of algorithms in microprocessors and digital signal processors.

Specification of controlled education, way of implementation and compensation for absences

The content and forms of instruction in the evaluated course are specified by a regulation issued by the lecturer responsible for the course and updated for every academic year.

Recommended optional programme components

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

MITRA,S.K., Digital Signal Processing-A Computer-Based Approach. The McGraw-Hill Companies, Inc. New York 1998 (EN)
OPPENHEIM, A.L., SCHAFER, R.W., Digital Signal Processing, Prentice-Hall, Inc. New Jersey, 1995. (EN)
SMÉKAL,Z., VÍCH,R., Zpracování signálů pomocí signálových procesorů. Radix spol.s.r.o., Praha 1998. (CS)

Recommended reading

Not applicable.

Classification of course in study plans

  • Programme EECC Bc. Bachelor's

    branch B-MET , 2 year of study, summer semester, elective specialised
    branch B-TLI , 2 year of study, summer semester, elective specialised

  • Programme EEKR-CZV lifelong learning

    branch EE-FLE , 1 year of study, summer semester, elective specialised

Type of course unit

 

Lecture

26 hod., optionally

Teacher / Lecturer

prof. Ing. Jiří Mišurec, CSc.

Syllabus

Discrete signals and systems. Discrete signals - sequences. Linear time-invariant discrete system. Stability and causality. Frequency representation. Sampling of continuous signals, aliasing. Two-dimensional signals and systems.
Z-transform, convergence region and properties. Inverse z-transform and its calculation by means of the residue theorem. Solution of difference eqations using the z-transform.
Transfer function of the pole-zero plot, frequency response and its geometrical interpretation. Two-dimensional z-transform.
Discrete Fourier transform and its features. Circular (periodic) convolution and its calculation by means of DFT. Calculation of discrete convolution, method of overlap-add and overlap-save. Two-dimensional DFT.
Fast Fourier transform. Calculation of two real sequences, calculation of double-length real sequence. Fast convolution and correlation.Calculation of inverse DFT by means of direct DFT.
Representation of discrete systems using matrices and signal flow graphs. Mason's rule. State-space canonic structures, serial and parallel forms. Solution of state-space difference equations.
Design of type FIR digital filters, linear phase. Method of windowing, method of frequency response sampling. Optimum uniform rippled filters. Remez algorithm.
Design of type IIR digital filters. Making use of analog prototypes. Frequency transformation. Methods of signal invariance and bilinear transformation.
Multirate systems. Undersampling (decimation) and interpolation. Change in sampling frequency in the form of rational fraction. Filter banks.
Homomorphous processing of signals. Complex and real cepstrums. Application of cepstrums in speech and image processing.
Signal quantization in in discrete systems. Fixed- and floating-point representation of numbers, quantization and rounding. Quantization of transfer function coefficients. Quantization of intermediate results, limit cycles, scaling to reduce arithmetic overflow. Quantization of continuous signal.
Hardware and architecture of microprocessor circuits for digital signal processing. Survey of demands on processing signals from various regions. Harvard architecture. Definition of digital signal processor, classification of digital signal processors by generations, properties of individual generations. Common properties of various types of digital signal processor.
Realization of digital filters and FFT processor in digital signal processors. Development tools, on-chip emulation (DSPlus, DSP56002EVM).

Exercise in computer lab

26 hod., compulsory

Teacher / Lecturer

prof. Ing. Jiří Mišurec, CSc.

Syllabus

Basic operations in Matlab, generation and representation of discrete signals.
Spectral representation of discrete periodic and non-periodic signals.
Discrete Fourier series and transform and its connection with Fourier series and transform. Fast Fourier transform (FFT).
Discrete linear and periodic convulution and correlation. Calculation using the FFT.
Test No 1.
Modelsof discrete systems, external and state-space description. Transfer function, impulse response, pole-zero plot.
Design of type FIR digital filters, windowing method, Remez algorithm
Desaign of type IIR digital filters.Bilinear transformation and impulse invariance methods.
Test No 2.
Multirate systems, decimation and interpolation.
Complex and real cepstrums. Unwrapping of phase.
Quantization effects in discrete systems. Implementation of algorithms on microprocessors.
Test No 3.