Internet of Things
Introduction
2024-10-21
Prof. Dr. Oliver Hahm
- Study of Computer Science at
Freie Universität Berlin - Software Developer for ScatterWeb and Zühlke Engineering
- Research on IoT and Operating Systems
Contact
E-mail: oliver.hahm@fb2.fra-uas.de
Appointments: via e-mail, room 1-212
Join the RIOT!
RIOT is the friendly operating system for the IoT!
You’re interested in …
- …programming the IoT?
- …collaborate with hundreds of people from all over the world?
- …contribute to a big FLOSS project?
Get in touch and do some hacking at the All RIOT event at the university! Usually every second Wednesday at 2pm in room 1-237.
Or look at https://allriot.dahahm.de
Organizational
- This seminar is for students of Allgemeine Informatik only
- There are separate seminars for students of High Integrity Systems (HIS)
- It takes place as a face-to-face course, i.e., participation is mandatory
- All course material can be found at https://teaching.dahahm.de/teaching/iot_ws24.html
Organizational
- Individual work
- Each student selects a research field from a given list
- Get an overview about the selected field
- Present this overview to the rest of the course
- Identify a relevant research question in this field
- Submit a report in the end
- Summarize a randomly selected topic
Hands-on Experience
In order to gain a solid understanding of IoT technologies, it is inevitable to gather practical experience: Four times this semester we will conduct some hands-on sessions.
campUAS
Enrolment Key:
HahmIoT
Assessment
- 30% for the report
- Content
- Structure
- Presentation and format
- Literature
- 20% practical work
- 40% for the presentation
- 10% summary on a randomly selected topic
A Brief History of the Internet of Things
- A Coca-Cola vending machine was connected to the Internet at Carnegie Mellon University
- The Smart Dust research proposal at Berkeley kick-started research on Wireless Sensor Networks (WSNs)
- Kevin Ashton (P&G) coined the term Internet of Things
- Cisco identified the birth of IoT by the tipping point “when more ‘things or objects’ were connected to the Internet than people”.
Connecting Smart Objects at Internet Scale
- From 3.5 billion Users to 50 billion Devices on the Internet
- Transforming Things into Smart Objects
- Enabling Interconnected Smart Services
Use Cases
Mobile Health
Building & Home Automation
Industrial Automation
Micro & Nano Satellites
Constraints and Requirements
Low-end IoT Devices: Limited Resources (RFC7228)
Arduino Due
Raspberry Pi Pico
nRF52840 Dongle
- Memory < 1 Mb
- CPU < 100 MHz
- Energy < 10 Wh
Application Requirements
- Interoperability
- Energy Efficiency
- Reliability
- Latency
- Low Cost Factor
- Autonomy
- Security
- Scalability
- Sustainability
- Privacy
- Safety
Evolution of low-power hardware
- Small microcontrollers are getting more and more powerful, energy-efficient, secure, and/or cheaper
- What are the latest developments?
- Which impact does this have on software design?
Programming low-end IoT devices
- Requirements and constraints of low-end IoT devices influences the choice of the programming language
- Available compiler (+ toolchain)
- Tooling (IDE, debugger etc.)
- Size of resulting binaries
- Access to hardware
- Safety and security concerns
- Learning curve
- Feature set
Operating systems for low-end IoT devices
- The particular challenges of IoT applications mandate for new operating systems
- Typical candidates are:
- RIOT
- Zephyr
- mbedOS
- FreeRTOS
- mynewt
- Contiki
- Linux
- What about standards like POSIX for these OS?
Energy-efficient wireless protocols
- (Wireless) Communication is typically one of the biggest energy consumer for low-end IoT devices
- Hence, efficient technologies are required
- Examples
- IEEE 802.15.4
- BLE
- IEEE 802.11ah
The IoT Network Stack
- Integration of constrained-node networks into the Internet
- Adaptation of traditional Internet Protocols
- Standardization and interoperability efforts
Traditional Internet
HTTP
TCP
OSPF
IPv6
IEEE 802.11, Ethernet
Layer
IoT
CoAP
UDP
RPL
6LoWPAN
IEEE 802.15.4, BLE
Routing protocols for constrained networks
- The constraints and requirements of (low-power) IoT networks pose new challenges on the routing protocols to be used inside and between local IoT networks.
- Survey the evolution of WSN routing protocols.
- RPL, its flavors, and what else?
- MANET protocols?
Transport layer issues for constrained-node networks
- Many (low-end) IoT solutions are in favor of UDP on the transport layer because of its low complexity and lightweight
- However, many traditional backend solutions (like MQTT or HTTP) are based on TCP
- Most recently a new Internet transport layer has evolved: QUIC
- What is the current state and what are the perspectives?
- What are challenges and what are the opportunities?
Lightweight integrity and confidentiality
- How to encrypt and sign data in IoT networks?
- Channel security vs. object security
- Cryptography for constrained devices
Cloud solutions for IoT applications
- The backend of an IoT application is typically hosted in the cloud
- IoT cloud providers offer various services like providing endpoints, data processing, device management, or software update services
- Multiple commercial cloud providers exist
- Azure IoT Hub
- Google Cloud
- AWS IoT
Software updates for IoT systems
- Software updates for IoT systems is a crucial cornerstone of security.
- SUIT and what else?
- Problems, challenges, approaches
Survey on IoT applications
- Is IoT still missing a killer app?
- Many application scenarios: Home automation, building automation, industry automation, mobile health, connected cars
- What are the (economical, egological …) benefits from connected devices?
Any Questions?
