¹û¶³Ó°Ôº

IoT is a growing field encompassing digital and physical technologies with applications across different industrial sectors, and spanning manufacturing, consumer electronics, security, mining, design, transport, exploration and healthcare.

Traditional computer science areas focus primarily on digital information processing. However, creating IoT systems involves specific challenges. For instance, each individual sensor system (edge node) needs to deal with the boundaries between the digital and physical worlds, with strict real-time requirements (e.g., response times) or with limited access to resources (e.g., computing power, energy consumption). Each of these nodes also needs to operate as part of a heterogeneous network of interconnected individual systems.

This module provides you with the knowledge and experience required to create of such edge nodes. You will acquire joint understanding of the dynamics of computers, software and physical processes (e.g., sensors, micro-controllers), to create systems dealing with the physical-digital boundary. You will also acquire concepts related to networks and concurrency, as to enable interoperation with other IoT nodes.

We will use formal models, to capture the temporal and data dynamics of software and networks, but also the concurrent and continuous nature of the physical world. We will then develop a critical understanding of the usual tools available (e.g., sensors, actuators, embedded processors) and the techniques required to translate our models into physical realizations. We will apply such knowledge in the lab, creating and testing actual edge node systems.

Aims:

The aims of this module are to:

• Provide students with the tools required to design and implement edge nodes, particularly addressing their sensing actuation requirements and their interoperation as part of a larger IoT system.
• Describe formal approaches to design edge nodes, addressing both the need for continuous and discrete models.
• Develop students’ critical understanding of the usual tools available (e.g., sensors, actuators, embedded processors) and the techniques required to implement them, translating their models into physical realizations.
• Allow students to apply the knowledge acquired to create specific edge nodes, combining theory (i.e., hybrid models) and practice (i.e., actual implementations).

Intended learning outcomes:

On successful completion of the module, a student will be able to:

1. Use formal models, to capture the temporal and data dynamics of software and networks and the concurrent and continuous nature of the physical world.
2. Understand a comprehensive set of techniques used to design and analyse of computational systems that interact with physical processes.
3. Demonstrate a critical awareness of the properties of the technologies involved (i.e., sensors, embedded processor, communication interface, output modalities), their low-level properties (bandwidth, range, noise) and the techniques allowing practical use (e.g., sampling and filtering).
4. Apply knowledge and practical understanding to the implementation of actual edge nodes.

Indicative content:

The following are indicative of the topics the module will typically cover:

Theory (lectures):

Content will be specifically tailored to the aims of the module, using relevant literature (i.e., textbooks, papers) from related disciplines in cyberphysical systems and digital signal processing. More specifically, the program will cover continuous, discrete and hybrid modelling of edge nodes; sensors/actuators and required filtering and control techniques; and techniques for multi-tasking, concurrent and network communication. Delivery will be reinforced by exercises in class, as well as weekly exercises to be completed during the module.

Practicals (lab sessions):

The prior theoretical aspects will be put in practice during our lab sessions, using a variety of computing platforms (micro-controllers, DSP processors), sensors and actuators, applied to a series of specific problems (e.g., sensor calibration, filtering, and data processing; robot control and hill climbing).

Requisites:

To be eligible to select this module as an optional or elective, a student must be registered on a programme and year of study for which it is formally available.