BSc:DigitalSignalProcessing

Digital Signal Processing

• Course name: Digital Signal Processing
• Course number: XYZ
• Knowledge area: Math, Computational Science

Administrative details

• Faculty: Computer Science and Engineering
• Year of instruction: 3rd year of BS
• Semester of instruction: 2nd semester
• No. of Credits: 4 ECTS
• Total workload on average: 144 hours overall
• Frontal lecture hours: 2 per week
• Frontal tutorial hours: 0 per week
• Lab hours: 2 per week
• Individual lab hours: 0
• Frequency: weekly throughout the semester
• Grading mode: letters: A, B, C, D

Prerequisites

• Mathematical Analysis I
• Mathematical Analysis II
• Analytic Geometry and Linear Algebra I
• Analytic Geometry and Linear Algebra II
• Discrete Math and Logic

Course outline

The goal of the course is to present rigorous mathematical foundations of signal processing tools using intuitive geometric point of view. The course is designed to provide basic mathematic knowledge needed for further studies of applied signal processing and digital signal processing from engineering as well as from mathematical perspective.

Expected learning outcomes

The course will provide an opportunity for participants to:

• know basics of the mathematical theory of
  • Discrete-time Fourier Transformation (DTFT),
  • Discrete Fourier Transformation (DTFT),
  • Fast Discrete Fourier Transformation (FDFT),
  • Fourie series;
• understand relations between
  • analog and digital signals,
  • continuous and discrete signals,
  • discrete signals and their convolution,
  • discrete systems and filters,
  • impulse and frequency domains;
• get practical experience with mathematical package SciLab/Octave in digital signal processing.

Programming related learning outcomes

Course includes 6 programming laboratory assignments:

1. Basics of Octave: realize computation algorithms.
2. DSP for audio signals: echo generation/cancellation, auto-correlation.
3. Convolution reverb and room impulse response.
4. DFT/FFT implementation, verification of the Convolution theorem.
5. Developing spectrogram function from scratch.
6. Fourier series approximation, Gibbs phenomenon.

Special attention will be paid for the quality and the OO-style of the implementation.

Expected acquired core competences

• Vector spaces
• Hilbert spaces
• complex numbers
• continuous(-time) and discrete(-time) signals
• Discrete(-time) Fourier transform
• Z-transform
• Fourier series
• Dirac delta function.

Required computer resources

Alternatives:

• SciLab is a free and open-source, cross-platform numerical computational package available for download from https://www.SciLab/Octave.org;
• GNU Octave is software featuring a high-level programming language, primarily intended for numerical computations (mostly compatible with MATLAB) available for download from gnu.org/software/octave/.

Evaluation

• Grading items:
  • In-class participation 1 point per week for each individual contribution in class (i.e. 14 points overall).
  • Final overall in-course participation/contribution (“instructors’ gratitude”) 6 points.
  • Each in-class theory or practice tests — up to 10 points (i.e. 20 points overall).
  • Mid-term exam up to 20 points, final examination up to 20 points (i.e. 40 points overall).
  • Each lab 15 points (i.e. 90 points overall).
• Final grade scale:
  • A: 136..170 points (i.e. 80% at least);
  • B: 102..135 points (i.e. 60% at least);
  • C: 68..101 points (i.e. 40% at least);
  • D: less than 68 points.