English

If sufficient demand

The course gives an overview of modern methods for scheduling and hardware/software co-design for real-time and embedded systems. There is a special focus on implementation of machine learning algorithms in resource-constrained systems.

Introduction to real-time and embedded systems;
Servers for handling aperiodic workloads;
Limited preemptive and elastic scheduling;
Multicore scheduling;
Introduction to machine learning and artificial intelligence;
Introduction to HW/SW codesign and partitioning;
Neuromorphic computing with spiking neural networks.

explain the mechanisms of several scheduling algorithms and resource synchronization protocols.
know the principles of scheduling servers for aperiodic workloads.
describe how to handle CPU overloads.
understand the basic principles of multicore scheduling.

conceptually implement neural networks in resource-constrained embedded systems.
perform hardware/software co-design and partitioning for small examples.

Lectures

Written exam

Failed, pass

Basic knowledge on real-time systems from, e.g., FRTN01 Real-Time Systems.

Buttazzo, G.: Hard Real-Time Computing Systems. Springer, 2011.

A reading list of research papers will be provided in class.

The course is given in the form of six half-day lectures and has a final written exam.

FRT160F

 -02-09

Professor Thomas Johansson