This is a course of the “Embedded System” track of the Master in Computer Engineering, held in the 1st term of the 2nd year. This course studies the issues related to the modeling, design and simulation of resource-constrained embedded systems used in the Internet-of-things (IoT) world, where constraints concerns especially energy and computational power. The emphasis of the course will be on the hardware architecture side of the problem using model- and simulation-based approaches.

- Understanding of the energetic issues of IoT devices and systems; - Ability in the analysis of the energy consumption sources; - Ability in the characterization of workloads for the dynamic analysis of IoT systems; - Skills in the design of energy-efficient solutions. - Ability in the quantitative evaluation of the effectiveness of the design solutions; - Understanding of the fundamental features of non-fixed power supply sources such as energy storage devices or power generation sources. - Skills in the design of power supply systems (battery sizing, lifetime of a system).

The course requires the knowledge of C/C++ programming (data structures and algorithms). A basic knowledge of the Matlab/Simulink environment, as well as of calculus, statistics, digital electronics and digital design, and computer architecture can be useful. No specific skills in hardware design (e.g., VHDL/Verilog) is required.

1. Energy Management [2.5 CFU] o Technological and architectural trends and relative energetic implications o Characterization of the various sources of power consumption and their interaction with other metrics o Dynamic power management (DPM); concepts and implementations: shutdown, voltage/frequency scaling, threshold voltage scaling and their relative quantitative analysis; o Application of dynamic power management to the various sub-components of an embedded system and their relative peculiarities (computational units, memories, peripherals); o Other non-DPM based optimizations: information compression and coding; o Quality/energy tradeoff in IoT systems: approximate and error-resilient computations; solutions for non-computing components (e.g., display and other interfaces) 2. Energy Generation and Storage: [1.5 CFU] o Storage: types of energy storage devices (batteries, fuel cell, photovoltaic cells) and their relation with power management o Generation: energy scavenging solutions and their energetic implications o Conversion: types of converters and their efficiency. o Simulation and design of the energy distribution sub-systems in an embedded device.

Lab classes will consist of the implementation of the techniques shown in class using Matlab/Simulink and SystemC. 3 lab sessions with as many deliverables are planned (2 CFU). Labs will be held during class time using the students’ computers.

There is no official textbook. Class handouts will be made available on the course webpage. Additional material such as papers, links to websites, software and manuals, will be also made available on the course webpage.

Modalità di esame: Prova scritta (in aula); Elaborato scritto prodotto in gruppo;

Exam: Written test; Group essay;

... Expected learning outcomes: - Understanding of the topics covered in the course - Skills in the design of the power supply system of an embedded or mobile device based on its functionalities - Skills in the design of power management solutions and their quantitative evaluation Exam criteria and rules: The exam aims at assessing the skills and knowledge described above by means of a two-part evaluation: The first part consists of a closed-book written test including both numerical exercises and open-answer questions. The time allowed for the test is 2 hours and the maximum score is 24 points. The second part consists of the evaluation of the reports on the lab exercises. Reports have to be delivered by the end of the winter exam session. Each lab will get a different score depending on its quality. The total maximum score for the lab reports is 9 points. The sum of the test score and the lab evaluations will yield the final score (maximum 33 points corresponding to 30 cum laude). Student can also integrate the final evaluation up to a maximum of 6 points with an individual project proposed by the instructor to be delivered by the end of the exam session. The project represents an additional element to assess the student's skills in solving problems that cover the entire context of energy-constrained embedded systems, unlike the more specific cases covered in the labs; moreover, projects allow the instructor to assess the autonomy of the student in building the knowledge base required to carry out the project itself.

Gli studenti e le studentesse con disabilità o con Disturbi Specifici di Apprendimento (DSA), oltre alla segnalazione tramite procedura informatizzata, sono invitati a comunicare anche direttamente al/la docente titolare dell'insegnamento, con un preavviso non inferiore ad una settimana dall'avvio della sessione d'esame, gli strumenti compensativi concordati con l'Unità Special Needs, al fine di permettere al/la docente la declinazione più idonea in riferimento alla specifica tipologia di esame.