Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Intelligent monitoring systems can effectively predict or detect anomalies and issues in smart working systems and ecosystems and implement the proper countermeasures.
In their effective and efficient use, attributes like responsiveness, performance, quality and trustworthiness should be appropriately tested and assessed before integrating the monitoring system into an ecosystem.
At the same time, predicting security and trust vulnerabilities is crucial for IoT interconnected systems and ecosystems, especially when integrating new, third-party, or open-source components.
The proposed course will overview the challenges and primary functional and non-functional properties assessment methodologies. It will also provide guidelines for successful monitoring and testing activities and illustrate the commonly adopted tools and techniques. An introduction to Complex Event Processing technologies will also offered.

Course Contents in brief:

1. Event-Driven monitoring concepts
   1. pros/cons
   2. event
   3. rules
   4. sources and probes

2. Introduction of functional and non-functional properties

   1. Property definition and selection

   2. Who should be involved?

3. Monitoring of complex systems

   1. Representing the behavioral model

   2.  Business Processes (BPMN) for monitoring industrial systems

   3.  Instrumentation of a system

   4.  Data acquisition and timing

   5.  Analysis and Vulnerability detection

4. Knowledge representation and management

   1. Using Ontologies

   2.  Using Generative AI

   3.  Complex Event Processing language

5. Monitoring for Cybersecurity

   1. Integrating access control

   2.  Malicious behavior detection

   3.  Mitigation strategies

   4.  pros and cons

6. Intelligent predictions and predictive failure detection

   1. Exploiting Digital Twin

   2.  Exploiting failure detection model

   3.  Exploiting AI

   4.  Prediction and time constraints

7. Assessing monitoring systems

   1. Main testing approaches

   2.  Testbed definition

   3.  Performance evaluation

Schedule:

1. April 1, 2025 – 09:00-12:00→Monitor basic concepts and functional and non-functional properties
2. April 3, 2025 – 09:00-12:00 → Monitoring of complex systems
3. April 7, 2025 – 14:00-17:00 → Knowledge representation and management
4. April 8, 2025 – 09:00-12:00 → Monitoring for Cybersecurity  and smart predictions
5. April 10, 2025 – 09:00-13:00 → Monitoring the monitoring system and Practical examples.

Hours:
12 hours (3 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

The aim of the course is to provide a comprehensive understanding of the most commonly used optimization methods and algorithms in the field of information engineering research. The course will include practical exercises using MATLAB.

Course Contents in brief:

1. Mathematical optimization
2. LS and linear programming
3. Combinatorial optimization
4. Convex optimization
5. Geometric optimization
6. Approximation algorithms for NP-hard problems

Schedule:

1. March 11, 2025, h. 14.30-17.30, Mathematical optimization and LS and linear programming (3 hours).
2. March 12, 2025, h. 14.30-17.30, Combinatorial optimization (3 hours)
3. March 13, 2025, h. 14.30-17.30, Convex and geometric optimization (3 hour)
4. March 14, 2025, h. 14.30-17.30, Approximation algorithms for NP-hard problems (3 hours)

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor
Aula Riunioni del Dipartimento di Ingegneria dell'Informazione, Largo Lucio Lazzarino 2, Piano 6
Lezione online il 20/03/2025

To register to the course, click here

Short Abstract:

This course explores how LLMs transform academic writing and presentations, offering high English accuracy, title suggestions, and support across academic tasks, including social aspects. Participants will learn to critically assess and effectively prompt LLM outputs.
Key challenges, such as unintended text changes, lack of readability and relevance flags, and overly verbose email structures, are discussed.
The course emphasizes manuscript skills (clarity, conciseness, and readability), managing referees, and effective presentations, including co-creating slides with LLMs, selecting inclusive visuals, and improving pronunciation.

Course Contents in brief:

Large Language Models

LLMs are transforming how academic papers and presentations are written. They have some major advantages:

1. they generate texts that are 99.9% accurate from an English point of view (though not from a factual point of view); they can correct texts already in English with at least 95% accuracy
2. they can give suggestions on titles for papers, on how to interact with referees, and on some elements regarding the quality and content of the writing
3. they can help with practically every aspect of academic life - not just studying for also socializing

The key is in learning how to prompt and to judge the bot's output using critical thinking skills.
However LLMs have major drawbacks:

1. when correcting and paraphrasing they may make serious undesired modifications
2. they do NOT flag some major issues that guarantee that a paper will be published: readability, conciseness, relevance, inclusion of all aspects that referees expect to find (e.g. limitations of work)
3. they generate emails that always follow the same verbose style and structure

Manuscript writing skills

Focus on key areas of a paper where special human attention is required – highlighting key findings, differentiate own work from that of others, discussing limitations; short clear simple sentences and paragraphs.
Effective titles and abstracts that will attract attention and thus be read.
Readability.
Dealing with referees and editors.

Presentation skills

Drafting a skeleton presentation co-created with ChatGPT.
Designing, practising and delivering well-structured  presentation that attracts and holds audience attention.
Drafting a script using simple and short sentences.
Choosing appropriate and inclusive images - highlighting importance of diversity.
Learning from TED and 3MTs.
Improving pronunciation and intonation in English.

Schedule:

1. March 6, 2025: 10:30-13.00 - Via Caruso
2. March 7, 2025: 10:30-13.00 - Via Caruso
3. March 11, 2025: 10:30-13.00 - Via Caruso
4. March 13, 2025: 10:30-13.00 - Via Caruso
5. March 18, 2025: 10:30-13.00 - Via Caruso
6. March 20, 2025: 10:30-13.00 - Lezione Online
7. March 25, 2025: 10:30-13.00 - Via Caruso
8. March 27, 2025: 10:30-13.00 - Largo Lucio Lazzarino

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Nanomaterial and micromachining-enabled soft and wearable electronics hold great promise in enabling telehealth and telemedicine. Wearable electronic devices transducing physical, chemical, and/or physiological responses to electrical signals, have been used in health monitoring, such as real-time detection of blood pressure, respiration rates, body temperature, and human motion. Combined with the ever-improving wireless and radio-frequency (RF) technologies, wearable electronic devices can transmit information in real time and be powered wirelessly in a passive and energy-saving manner. This short course will review the recent progress of nanomaterials and microsystems, as well as their unique electrical, (bio-)chemical, thermal, and mechanical properties that make them suitable for wearable antenna and circuit applications. This short course will also discuss challenges and breakthroughs in wireless sensing and interrogation technologies for wearable IoT devices.

Course Contents in brief:

1. Introduction to wireless health, telemedicine, and telediagnosis (2 hours)
2. Introduction to smart skins and wearable technologies (2 hours)
3. Micro/nano-driven wearable antennas and circuits: material selection, design and fabrication (10 hours)
4. Wireless sensing, communication and energy harvesting for IoT sensors (4 hours)
5. Signal processing and multisensor fusion (2 hours)

Schedule:

1. Day1 – April 8, 2025: 13:30-17:30 Introduction to wireless health, telemedicine, and telediagnosis (2 hours); Introduction to smart skins and wearable technologies (2 hours)
2. Day2– April 9, 2025: 13:30-17:30 Micro/nano-driven wearable antennas and circuits: material selection, design and fabrication (4 hours)
3. Day3 – April 10, 2025: 13:30-17:30 Micro/nano-driven wearable antennas and circuits: material selection, design and fabrication (4 hours)
4. Day4 – April 11, 2025: 9:00-13:00 Micro/nano-driven wearable antennas and circuits: material selection, design and fabrication (2 hours); Wireless sensing, communication and energy harvesting for IoT sensors (2 hours)
5. Day5 – April 11, 2025: 14:00-18:00 Wireless sensing, communication and energy harvesting for IoT sensors (2 hours); Signal processing and multisensor fusion (2 hours)

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

AI and deep learning are improving and revolutionizing machine vision and, while 2D elaboration are almost all camera model agnostic, their 3D application require a deep understanding of the projective geometry and the basic concepts of traditional localization approaches.

This course offers an overview on the matter starting from the basic concepts up to some demos with the application of the most advanced AI powered localization algorithms. 

Course Contents in brief:

1. Camera modeling.
2. Tracking of points and structured markers with mono e stereo cameras.
3. 3D scan elements and their application to tracking.
4. Marker-less model (edge) based tracking.
5. Marker-less DL tracking: detection, instance and model aware approaches.

Schedule:

1. February 20, 2025 -  9:00-13:00  "Principles of geometric optics and camera model.Geometric and algebraic pinhole model of common cameras. Camera calibration algorithms and demo."
2. February 21, 2025 -  9:00-13:00  "PnP based tracking of structured planar markers.Tracking of planar unstructured markers. Tracking accuracy of planar markers."
3. February 25, 2025 -  14:00-18:00  "Multiview tracking.Accuracy of stereo tracking. 3D scan elements and their application to tracking."
4. February 27, 2025 -  14:00-18:00  "Introduction to marker less tracking.Model (edge) based tracking. Marker-less DL tracking: detection, instance and model aware approaches."
5. February 28, 2025 -  9:00-13:00  "Exam". 

Hours:
24 hours (6 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Multi-physics and physics-based modelling represent important tools to support quantitative analyses and parametric design for different engineering problems, including the structural evaluation of complex mechanical parts, the functional assessment of power supply systems, transducers and smart materials (e.g., piezoresistive, magnetostrictive), the implementation of electromagnetic field sources and the optimization of culture conditions in engineered cellular systems.

The course aims at introducing the students to physics- and multi-physics-based modelling in the field of information engineering using the finite element method (FEM), with a particular focus on biomedical applications. 

The first part of the course will provide them with the basics for developing, solving and handling results of FEM models in the COMSOL Multiphysics environment. Specifically, simple dynamical systems involving (bio)physical phenomena concerning transport and reaction of chemical species, heat transfer and solid mechanics will be presented. Then, the second part of the course will be devoted to hands-on sessions on the software, identifying applications relevant to the PhD activities of the students. 

Course Contents in brief:

1. Brief introduction to physics- and multi-physics-based modelling using FEM
2. COMSOL Multiphysics: an easy-to-use interface for FEM modelling in information engineering
3. Less is more: geometrical and physical symmetries and definition of the space dimension 
4. Design of easy and complex geometries and meshing (predefined and customized)
5. Study design: solver choice, steady-state and time-dependent models, parametric sweep
6. Data handling: result post-processing and visualization
7. Using and combining specific modules (“Fluid flow”, “Transport and reaction of chemical species”, “Heat transfer” and “Structural mechanics”) for modelling simple multi-physics systems
8. Identification and implementation of specific models relevant to the PhD field of the students

Schedule:

1. March 4th, Tuesday – 4h (9 – 13)
2. March 5th, Wednesday – 4h (14 – 18)
3. March 6th, Thursday – 4h (14 – 18)
4. March 7th, Friday – 4h (14 – 18)  
5. March 14th, Friday – 4h (9 – 13)
6. March 21st, Friday – 4h (9 – 13)

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

In recent years, the role of sensor systems as key components of complex artificial apparatuses has been continuously gaining importance. The effectiveness of future devices, such as robots, autonomous vehicles and wearable medical systems, in improving the life of human beings critically depends on the availability of specialized sensor systems allowing the acquisition of precise information about the surrounding environment. In the last three decades, challenging requirements of size, cost and reliability drove the research towards the present generation of sensors, which leverage on the progress in micro-technologies. On the other hand, the lack of sensor systems with adequate performances is still limiting exciting application that could significantly contribute to the safety of humans in everyday life. For the success of present and future sensor systems, the crucial role of electronics is undisputable. 

This course is meant to analyze the structure of a modern sensor systems, classifying the different components according to their specific function and proposing an original hierarchical overview based on modularity. Within this general view, the focus is placed on the intimate relationships between physical components, i.e. the transducers, and the electronic units, highlighting the challenging requirements that the former posed to the latter. By a in depth analysis of significant case studies, it will be shown how the function performed by the electronic circuits is essential for the transduction mechanism. Furthermore, the fundamental design approaches and technologies that allowed the integration of complex sensor systems on a single chip (System on a Chip) or in single packages (System in Package) will be discussed. Finally, the trend toward versatile universal sensor interfaces will be exposed, focusing on a research product being developed by the research group of the author of this course [1-4]. Theoretical lectures will be alternated with experimental demonstrations performed by means of purposely built development boards controlled with the Python language. 

Course Contents in brief:

1. Overview of sensor system: classification, characteristics & trend.
2. Beyond the standards: design strategies for non-intrusive sensor systems with challenging constraints.
3. Let’s practice: experimental demonstrations of sensor systems by means of purposely built development boards

Schedule:

1. Day1 – February 10, 2025 - 14:00 /18:00, Lecture #1, (3 h) + Exercitation (1 h)
2. Day2 – February 11, 2025 - 14:00 /18:00, Lecture #2, (3 h) + Exercitation (1 h)
3. Day3 – February 12, 2025 - 14:00 /18:00, Lecture #3, (3 h) + Exercitation (1 h)
4. Day4 – February 13, 2025 -14:00 /18:00, Lecture #4, (3 h) + Final Exam (1 h)

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

With the advent of modern machine learning and deep learning techniques, the field of mathematical modelling has become increasingly saturated with powerful and effective models. However, most deep learning models have two major problems: the need for very large datasets (thousands to millions data) and the tendency to produce overconfident solutions. Probabilistic deep learning is thus emerging as a powerful support tool in the development of machine learning models, allowing the user to interpret results more realistically and providing an assessment of the risk of error related to the models' output (Trustworthy AI). This aspect is particularly important in many fields of application, such as medicine, biology, law, engineering, astronomy, etc.
This PhD course explores the most popular state-of-the-art models of probabilistic deep learning. In particular, Bayesian networks, probabilistic Ensembles and post-hoc techniques for integrating a pre-existing model with reliability estimates at near-zero computational cost are introduced.

Course Contents in brief:

1. Introduction to types of uncertainty: epistemic and aleatoric
2. From determinist to probabilistc networks
3. Intrusive approach: Bayesian Neural Networks and Ensemble Learning
4. Non-intrusive approach: MC-Dropount and Trust score
5. From theory to practice: implement probabilistic networks on real world data

Schedule:

1. Day1 – February 4, 2025, h. 9:00-13:00 (4 hours)
2. Day2 – February 6, 2025, h. 9:00-13:00 (4 hours)
3. Day3 – February 11, 2025, h. 9:00-13:00 (4 hours)
4. Day4 – February 13, 2025, h. 9:00-13:00 (4 hours)
5. Day5 – February 18, 2025, h. 9:00-13:00 (4 hours)

Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

The course will cover the main aspects of Edge Computing, from a practical perspective and with the aim to provide useful tool for application development. An initial technology overview will provide the description of MEC and related standards and industry associations working in the field, together with various open-source frameworks available. A particular emphasis will be given to the ETSI MEC standard (Multi-access Edge Computing), by mentioning as well other SDOs and the relationship with 3GPP for the definition of the future 6G systems. A special focus will be given to AI and the recently available frameworks that can help software development. A classroom exercitation will complete the course, where students will be asked to build their own application example exploiting MEC and based on the open-source software introduced during the course.

Course Contents in brief:

1. Edge Computing, Fog computing, Cloud computing
2. ETSI MEC Framework and Reference Architecture
3. MEC in 4G, MEC in 5G, MEC in Wi-Fi networks
4. MEC Services, MEC Management, MEC Mobility
5. Network APIs, CAMARA APIs, MEC APIs
6. Open-Source frameworks: MEC Sandbox, ETSI Forge
7. Artificial Intelligence, tools and Ecosystem
8. Classroom exercise

Schedule: 27-30 Janaury 2025

1. Day 1: 15.00-18.00
2. Day 2: 9.00-12.00, 14.00-18.00
3. Day 3: 9.00-12.00, 14.00-18.00
4. Day 4: 9.00-12.00

Hours:
16 hours (4 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

The increasing complexity of automotive and industrial systems demands innovative methods for designing and verifying control and monitoring algorithms. In this context, the Model-Based Design (MBD) approach emerges as a crucial solution to tackle these challenges. The unprecedented availability of computational power and data—both real and synthetic—opens up new possibilities for improving time-to-market and reducing validation costs. This course offers a comprehensive overview of modern control and monitoring architectures for power converters and drives, highlighting the importance of MBD.

Students will gain practical and theoretical skills to design, verify, and test advanced control and monitoring algorithms. The MBD approach, with its ability to leverage computational power and available data, provides a robust and effective methodology to address the challenges posed by the increasing complexity of modern power electronics systems.

Course Contents in brief:

• Lecture 1: Introduction to advanced control and monitoring architectures, including adaptive and predictive controllers, Kalman filters, and health state estimation algorithms. The theoretical foundations of the MBD approach and its impact on system quality and reliability will also be covered.

• Lecture 2: Utilization of algorithm verification tools through MIL/SIL (Model/Software-In-the-Loop) approaches, computational analysis and optimizations, automatic C/C++ code generation, and timed sequential logic coverage analysis. Special emphasis will be placed on code and model coverage analysis to ensure all parts of the system are adequately tested and verified.

• Lecture 3: Hands-on session using a specific toolbox for the NXP embedded platform (based on Cortex-M4, typical for automotive/automation applications) for SIL and PIL (Processor-In-the-Loop) testing for real-time verification. Techniques for designing test sequences from specifications will be addressed, ensuring the developed algorithms meet the stringent requirements of industrial applications.

• Lecture 4: Demonstration of a simple test bench using the GPIO of the NXP evaluation board to connect an inverter and a low-voltage/power synchronous motor, allowing for prototype HIL (Hardware-In-the-Loop) testing, interacting with runtime simulation. This practical example will show how MBD can be applied to create efficient and reliable control and monitoring solutions.

Schedule:

• Monday, January 20, h. 9:00-13:00
• Tuesday, January 21, h. 9:00-13:00
• Wednesday, January 22, h. 9:00-13:00
• Thursday, January 23, h. 9:00-13:00