Duration
25h Th, 20h Pr, 20h Proj.
Number of credits
| Master of Science (MSc) in Electrical Engineering | 5 crédits |
Lecturer
Language(s) of instruction
English language
Organisation and examination
Teaching in the second semester
Schedule
Units courses prerequisite and corequisite
Prerequisite or corequisite units are presented within each program
Learning unit contents
This course introduces the fundamental electronic, biological, and mathematical principles for designing neuromorphic signal processors, that is, analog signal processing units inspired by how single neurons, neuronal circuits, and entire nervous systems process signals. The focus is mostly on the elementary components (the "building blocks") of neuronal circuits and systems: single neurons and synapses. The final part of the course presents and explains the detailed functioning of event-based cameras, a prototypical example of a real-world application of neuromorphic signal processing.
It is highly recommended to take this course together with the Brain Inspired Computing course as the two courses are thought to complement each other.
The course will cover the following table of contents:
0. Introduction to neuromorphic engineering: why should we go neuromorphic.
1. Review of suprathreshold and subthreshold MOSFET characteristics, with an emphasis on subthreshold.
2. Basic static subthreshold circuits and transconductance amplifiers.
3. Translinear circuits and current-mode circuits.
4. Linear filters: transconductance-based design, current-mode design.
5. Review of biological neuronal behaviors. Modeling biological neurons and synapses.
6. Neuromorphic synapses and basic synaptic plasticity.
7. Neuromorphic neurons: translinear mode and current mode. Limitations of present designs and looking to the future.
8. Elements of mixed-mode (analog-digital) VLSI design and learning in VLSI circuits.
9. Photoreceptors and event-based cameras. Generalization to other senses. This theme will involve hands-on-session on neuromorphic hardware (DAVIS event-based camera) to be developed in the Department Neuromorphic Lab.
Learning outcomes of the learning unit
Through practical classes and exercises developed in the SPICE environment, and by following the principles introduced in the theoretical classes, students will be able to:
- Design linear and nonlinear, static and dynamic, subthreshold MOSFET circuits.
- Understand how single biological neurons and biological neuronal circuits and systems work, and how we can model them.
- Translate mathematical models of neurons and synapses into subthreshold MOSFET designs.
- Understand the basic principles for designing VLSI neuromorphic systems with learning capabilities.
- Understand and design event-based cameras and other kinds of neuromorphic sensors.
- Design neuromorphic brain-inspired visual computation layers to be used downstream event-based cameras.
Prerequisite knowledge and skills
Basic linear and nonlinear dynamical and control systems; electronic circuits and SPICE modeling.
Planned learning activities and teaching methods
The course includes both face-to-face lectures, exercise sessions, hardware hands-on-sessions in the Department Neuromorphic Lab, and projects. For project development, students will be able to access hardware and experimental data from the Department Neuromorphic Lab.
Mode of delivery (face to face, distance learning, hybrid learning)
Face-to-face course
Recommended or required readings
Shih-Chii Liu, Jorg Kramer, Giacomo Indiveri, Tobias Delbruck, Rodney Douglas - Analog VLSI Circuits and Principles - The MIT Press (2002)
Exam(s) in session
Any session
- In-person
written exam AND oral exam
Written work / report
Additional information:
Homeworks; project development and oral presentation/exam.
Work placement(s)
Organisational remarks and main changes to the course
Contacts
Alessio Franci.
https://sites.google.com/site/francialessioac/