- ✓The Internet of Things refers to the network of physical devices embedded with sensors, software and connectivity that enables them to collect and exchange data with other devices and systems over the internet.
- ✓IoT architectures typically consist of four layers: the perception layer (sensors and actuators that interact with the physical world), the network layer (connectivity infrastructure), the processing layer (data management and analysis) and the application layer (user-facing services).
- ✓Communication protocols used in IoT include MQTT, CoAP, Zigbee, Z-Wave and LoRaWAN, each optimised for different combinations of bandwidth, range, power consumption and device complexity.
- ✓Security is a critical concern in IoT systems: devices often have limited computational resources for security processing, are deployed in physically accessible locations and may remain in service for many years after their software becomes outdated.
- ✓The scale of IoT deployment creates significant data management challenges: a single industrial IoT deployment can generate terabytes of data per day, requiring carefully designed data pipelines, storage architectures and real-time processing capabilities.
Listen to the full episode inside the course. Enrol to access all 80 episodes, plus assignments, tutor support and Student Finance funding.
Start learning →Alex: Welcome back to HTQ Digital Technologies: The Study Podcast. Today Sam and I are starting Unit 10 at Level 5, which is the Internet of Things. Sam, this is one of those technologies that has moved from futuristic concept to pervasive reality remarkably quickly.
Sam: It really has. The idea of connecting physical devices to the internet for monitoring and control has been around for decades, but the combination of cheap, capable hardware, ubiquitous wireless connectivity and cloud computing platforms has made it economically viable at scale in a way it simply wasn't before. The result is an explosion in connected devices that is reshaping whole industries.
Alex: Let's start with a clear definition. What actually constitutes an IoT system?
Sam: An IoT system connects physical devices equipped with sensors and actuators to software systems over a network, typically the internet. The sensors gather data from the physical world, temperature, pressure, location, motion, sound, light and many others. The actuators allow the system to act on the physical world, turning things on and off, opening and closing valves, adjusting speed or position. The data flows to processing systems that interpret it and may trigger actions in response.
Alex: What are the architectural layers?
Sam: The standard way of describing IoT architecture uses four layers. The perception layer is the devices themselves: sensors, actuators and the embedded systems that connect them. The network layer is the connectivity infrastructure that carries data between devices and processing systems: this might be WiFi, cellular, Bluetooth, Zigbee, LoRaWAN or other protocols depending on the application's range, bandwidth and power requirements. The processing layer handles data aggregation, analysis and decision logic, which might happen in the cloud, at the edge close to the devices, or both. And the application layer is the user-facing software that presents information and allows users to control and configure the system.
Alex: Security is a particular concern with IoT, isn't it?
Sam: Critical and often underaddressed. IoT devices frequently have limited computational resources, which constrains the security measures that can be run on them. They may be deployed in physically accessible or even publicly accessible locations where they can be tampered with. They may run for years without software updates, meaning vulnerabilities remain unpatched long after they're discovered. And a compromised IoT device can be used as a foothold for attacking the wider network it's connected to. The Mirai botnet, which recruited hundreds of thousands of poorly secured IoT devices to launch massive distributed denial of service attacks, was a vivid demonstration of these risks at scale.
Alex: And data management at scale is a challenge too?
Sam: Enormous challenge. A single industrial deployment might have thousands of sensors each generating readings every few seconds. The data volumes quickly exceed what can be transmitted and processed in the cloud without significant cost and latency. This is driving the growth of edge computing, where data is processed close to its source before only relevant information or summaries are sent to the cloud. Getting the balance between edge and cloud processing right is one of the key design challenges in modern IoT systems.
Alex: Great introduction to this unit. Thanks, Sam. We'll look at planning an IoT application next.