SIGNAL PROCESSING for MULTIMEDIA APPLICATION
Academic Year 2025/2026 - Teacher: Giovanni SCHEMBRAExpected Learning Outcomes
Knowledge and understanding of the main elements concerning digital signal processing for multimedia applications
Acquire and understand the fundamental tools for handling digital signals and systems, including their characterization in the Fourier and Z domains.
Development of skills for the analysis and design of digital filters.
Ability to acquire and process aerial images from drones.
Applied knowledge and understanding of state-of-the-art techniques in digital communication systems, also aimed at practical applications in contexts different from the usual ones
Development of the necessary skills to analyze a discrete signal using spectral estimation techniques and to design a linear time-invariant system through its frequency response, in order to gain mastery of their use even in the future and in contexts other than those addressed during the course.
Independent judgment on what has been learned
Development of an adequate degree of autonomy in identifying the characteristics of linear time-invariant systems and the tools that can be used not only for designing simple digital filters, such as those presented in lectures, but also for designing more complex systems such as those for audio and video signal manipulation, including those generated by aerial cameras mounted on drones, for which a deeper understanding of the subject matter is required.
Communication skills for conveying information to heterogeneous audiences
Development of the ability to communicate effectively, using appropriate technical language, on topics related to multimedia signal generation, processing, and reproduction.
Ability to learn autonomously about developments related to the topics covered in class
Development of the ability to stay updated on scientific and technological advances in the field of digital signal processing, in order to independently explore new techniques for filtering and manipulating digital signals that will emerge in the future, with particular reference to emerging technologies for digital signal processing in the field of the “Internet of Sound” (IoS).
Course Structure
The course is composed of a part of theory, and a part of practice and laboratory activities.
In the case lectures will be partially or fully realized remotely by a video-communications platform, what declared above could undergo some changes, in order to achieve the objectives targeted in this syllabus.
Required Prerequisites
In-depth knowledge of signal theory, in particular Fourier analysis, sampling of an analog signal, and frequency response of a linear time-invariant system. Characterization of an analog linear system.
Attendance of Lessons
Detailed Course Content
The program is organized in seven Elementary Teaching Units (ETU):
|
Topic |
Textbook Information |
|
ETU 0. Introduction to the course
|
|
|
Presentation of the course syllabus and examination methods. |
Slides presented during lectures |
|
ETU 1. Signal Digitization
|
|
|
Sampling and quantization. Sampling theorem and aliasing. Quantization noise and non-uniform quantization. Signal reconstruction. Ideal reconstruction through cardinal interpolation. Sampling under non-ideal conditions: band-unlimited signals and anti-aliasing filters; non-instantaneous sampling and compensation. Reconstruction under non-ideal conditions: reconstruction with non-instantaneous pulses and compensation. |
Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 1, §§1.1–1.5, and slides presented during lectures |
|
ETU 2. Fourier Transform for Sequences and Z-Transform
|
|
|
Canonical sequences: unit impulse, unit step, exponential sequence, sinusoidal sequence. Operations on sequences. Discrete-Time Fourier Transform (DTFT): definition, inverse DTFT, DTFT of canonical sequences. Z-Transform for sequences: definition, relation to Fourier Transform, region of convergence, computation of Z-transform for notable sequences. Inverse Z-transform of rational functions. Convergence of Fourier and Z-transforms. |
Ch. 1, §§1.6–1.10, Ch. 2, and slides presented during lectures |
|
ETU 3. Discrete-Time Systems
|
|
|
Linear Time-Invariant (LTI) systems. Computation of output sequence. Causality and stability. Representations of an LTI system through impulse response, difference equation, transfer function, and frequency response. Phase delay and group delay. Distortionless systems. All-pass systems. Minimum-phase systems. Symmetric FIR filters: main properties (linear phase, constant phase delay, zero locations). Elementary discrete-time systems: delay, moving average, accumulator (or numerical integrator), numerical differentiator, decimator (downsampler), interpolator (upsampler), sequence multiplexer. Half-band FIR filters. Discrete analytic signal generator. |
Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 3, and slides presented during lectures |
|
ETU 4. Discrete Fourier Transform (DFT) and Fast Fourier Transform (FFT) |
|
|
DFT and IDFT. Circular shift. Circular and linear convolution. Application to the computation of the output of an LTI system. FFT. Radix-2 decimation-in-time algorithm. Inverse FFT. Spectral estimation: spectral resolution, periodogram, window selection and duration. Short-Time Fourier Transform (STFT). Sequence correlation. Fast linear convolution between two sequences and its application to sequence filtering. “Overlap & Add” and “Overlap & Save” algorithms. |
Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 5, and slides presented during lectures |
|
ETU 5. Design and Implementation of Digital Filters |
|
|
Digital filter design. Design specifications. Design of basic FIR filters. FIR filters: use of linear-phase FIR filters. Design by the window method. Low-pass FIR design. Band-pass FIR design, FIR differentiators, Hilbert FIR filters. Least-squares method. Frequency-sampling method. Equiripple method and Chebyshev criterion, Parks–McClellan approach. IIR filters: direct and indirect design methods. Bilinear transformation and impulse invariance. Design of high-pass, band-pass, and band-stop IIR filters. |
Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 6, Ch. 7, and slides presented during lectures |
|
ETU 6. Processing of Aerial Images from Drones |
|
|
Lossless and lossy coding. Overview of coding techniques for speech signals and image coding (JPEG and MPEG). Acquisition of images from drones using RGB, thermal, multispectral, and hyperspectral cameras. 3D reconstruction via photogrammetry with LiDAR and point cloud processing.. |
Notes provided by the professor and slides presented during lectures |
Textbook Information
[Smi] The Scientist and Engineer’s Guide to Digital Signal Processing, Second Edition, Steven W. Smith, California Technical Publishing, San Diego, California.
[Opp] Alan V. Oppenheim, Ronald W. Schafer, Digital Signal Processing, Prentice Hall Digital Processing Series
[Pro] J. G. Proakis, D. G. Manolakis, Digital signal processing.
Course Planning
| Subjects | Text References | |
|---|---|---|
| 1 | Introduction to the course | Slides presented during lectures |
| 2 | Signal Digitization | Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 1, §§1.1–1.5, and slides presented during lectures |
| 3 | Fourier Transform for Sequences and Z-Transform | Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 1, §§1.6–1.10, Ch. 2, and slides presented during lectures |
| 4 | Discrete-Time Systems | Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 3, and slides presented during lectures |
| 5 | Discrete Fourier Transform (DFT) and Fast Fourier Transform (FFT) | Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 5, and slides presented during lectures |
| 6 | Design and Implementation of Digital Filters | Textbook: F. Argenti, L. Mucchi, E. Del Re, “Elaborazione numerica dei segnali”, Ch. 6, Ch. 7, and slides presented during lectures |
| 7 | Processing of Aerial Images from Drones | Notes provided by the professor and slides presented during lectures |
Learning Assessment
Learning Assessment Procedures
The final assessment consists of an oral examination. Students who regularly attend the course may opt to take a midterm assessment covering ETUs 1–5 and complete a project related to ETU 6. Successful completion of both components allows the student to take the oral examination on a reduced syllabus.
Examples of frequently asked questions and / or exercises
- Analog-to-digital conversion: sampling, quantization, and encoding; guard band; ideal sampling, spectrum of the sampled signal, sampling theorem, aliasing and spectrum folding; quantization noise and SNR calculation; SNR for sinusoidal and Gaussian signals; non-uniform quantization.
- Definitions: Fourier Transform (FT) for sequences, normalized frequency and angular frequency, FT periodicity, inverse FT; notable sequences (unit impulse, rectangular pulse, unilateral exponential, sinusoidal sequence) and their FT.
- In MATLAB: sampling and reconstruction of an analog signal; plot of the original analog signal, plot of the sampled sequence, plot of the FT of the sampled sequence.
- Representations of an LTI system: impulse response, finite-difference equation, transfer function, pole-zero representation, frequency response. Conversion from one representation to another. Differences between FIR and IIR filters in the various representations.
- Given an LTI system described by a finite-difference equation: discuss a possible stable and causal implementation and compute the output for given input sequences. Plot the location of poles and zeros on the complex plane.
- Notch (band-stop), comb, and all-pass filters, with verification of the frequency response in MATLAB.
- Spectral estimation: spectral resolution, choice of window and its length, periodogram. Fens Laboratory – spectral estimation experiments. Correlation between sequences.
- MATLAB design of a low-pass FIR filter using the window method.