Raspberry Pi: What is it?

What is the Raspberry Pi?

The Raspberry Pi is a credit-card-sized single-board computer (SBC) developed by the Raspberry Pi Foundation in the United Kingdom. Originally designed to democratise computer science education, it quickly became the reference platform for professional IoT projects thanks to its unique combination of computing power, connectivity, and low cost (starting from 35 euros for a Raspberry Pi 4).

The Raspberry Pi 4 (and its successor, the Pi 5) features a quad-core ARM processor, up to 8 GB of RAM, USB 3.0 ports, a Gigabit Ethernet port, built-in Wi-Fi and Bluetooth, and a 40-pin GPIO (General Purpose Input/Output) connector that allows direct connection of sensors, actuators, and expansion modules. It runs a full Linux operating system (Raspberry Pi OS, based on Debian), providing access to the entire Linux software ecosystem: Python, Docker, Node.js, databases, and web servers.

In industrial and professional contexts, the Raspberry Pi is used as an IoT hub (collecting data from multiple sensors), a network gateway (converting between Zigbee/LoRaWAN and IP protocols), an edge computing platform (local data processing before cloud transmission), and an embedded application server (running Python/Django applications directly on-site).

Why the Raspberry Pi matters

The Raspberry Pi has democratised access to embedded computing and profoundly transformed the professional IoT landscape. Its importance rests on several distinctive factors.

• Full computing power: unlike simple microcontrollers (Arduino, ESP32), the Raspberry Pi runs a complete Linux OS, allowing the use of Python, Docker, databases, and web frameworks directly in the field.
• Rich software ecosystem: any software available on Linux ARM is usable on Raspberry Pi, from Django to Mosquitto (MQTT broker) to Zigbee2MQTT, TensorFlow Lite, and PostgreSQL.
• GPIO and extensibility: the 40 GPIO pins allow connection of countless sensors and modules without complex intermediate electronics, via I2C, SPI, UART, and digital GPIO protocols.
• Accessible cost: a Raspberry Pi 4 with case, power supply, and SD card costs less than 100 euros, making IoT projects financially viable even for SMEs with limited budgets.
• Massive community: millions of developers and open-source projects provide a knowledge base, tutorials, and libraries that significantly accelerate development.

How it works

The Raspberry Pi operates like a standard Linux computer but in a compact form factor suited to field deployments. The operating system is installed on a microSD card (or an SSD via USB for greater reliability). On boot, the Pi loads Linux and the configured services: data collection via sensors connected to the GPIO, MQTT service for data transmission, and Python scripts for processing and business logic.

For sensor data acquisition, the Raspberry Pi uses its hardware interfaces. The I2C bus connects digital sensors (temperature, humidity, pressure, air quality) on just two wires, with the ability to chain multiple sensors on the same bus. The SPI bus offers higher throughput for LoRa modules, displays, and analogue sensors via ADC converters. Digital GPIO pins handle simple input/output (motion detectors, relays, status LEDs).

Collected data is transmitted to the central backend via the MQTT protocol. An MQTT client (paho-mqtt in Python) publishes measurements on structured topics (e.g., building/floor3/room301/temperature). The Django backend, hosted on a central server or in the cloud, subscribes to these topics and processes data in real time: PostgreSQL storage, alert triggering, and dashboard updates. The Raspberry Pi can also perform local pre-processing (edge computing): filtering, aggregation, and anomaly detection, reducing the volume of transmitted data and the latency of critical alerts.

Concrete example

At KERN-IT, the Raspberry Pi is the cornerstone of our IoT solutions. For a Belgian telecom operator, we deployed Raspberry Pi devices in hundreds of technical cabinets across the territory. Each Pi collects data from temperature, humidity, and vibration sensors connected via I2C, transmits it via MQTT to our Python backend (Flask or Django depending on the project), and the results are visualised in real time on KERN MAP, our interactive mapping platform.

The Raspberry Pi locally runs a Python script implementing edge computing: it analyses temperature trends and triggers an immediate local alert (LED + buzzer) if a critical threshold is reached, without waiting for the central server's response. Simultaneously, the alert is published on MQTT for email notification and dashboard update. Docker is used on each Pi to containerise services (collection, MQTT, monitoring), facilitating remote updates via SSH and an automated deployment tool based on Fabric.

Implementation

1. Choose the model: Raspberry Pi 4 (4 GB) for most IoT projects, Pi 5 for intensive edge computing, Pi Zero 2 W for compact deployments requiring Wi-Fi.
2. Prepare the system: flash Raspberry Pi OS Lite (no GUI) onto an industrial-grade microSD card. Configure SSH, networking, and a watchdog for automatic restart in case of freeze.
3. Connect sensors: wire sensors to GPIO pins respecting voltage levels (3.3V). Use the adafruit-circuitpython Python libraries for I2C/SPI sensor interaction.
4. Install the MQTT client: deploy the paho-mqtt client with a Python script that publishes measurements to your central MQTT broker (Mosquitto or HiveMQ).
5. Containerise with Docker: package your services (collection, processing, MQTT) in Docker containers for reproducible deployment and remote updates.
6. Secure the deployment: configure TLS encryption for MQTT, change the default password, enable the firewall (ufw), and disable unnecessary services.

Associated technologies and tools

• Raspberry Pi OS: official Debian-based operating system, optimised for Raspberry Pi hardware.
• Python (paho-mqtt, adafruit-circuitpython): primary language for IoT programming on Raspberry Pi, with rich libraries for sensors and MQTT.
• Docker: IoT service containerisation for reproducible deployment and simplified updates.
• MQTT (Mosquitto): messaging protocol and broker for communication between Raspberry Pi and the central backend.
• Fabric: Python deployment tool for automating remote updates on Raspberry Pi fleets via SSH.
• Grafana: data visualisation platform, compatible with metrics collected by Raspberry Pi devices.

Conclusion

The Raspberry Pi is far more than an educational gadget: it is a professional IoT platform combining the power of a Linux computer with the flexibility of a microcontroller. Its ability to run Python, Docker, and complex network services makes it the ideal choice for connected sensor projects requiring local processing and rich connectivity. At KERN-IT, the Raspberry Pi is at the heart of our IoT stack, connecting field sensors to our Python backend (Flask or Django depending on the project) and KERN MAP to deliver comprehensive, reliable, and scalable monitoring solutions to our Belgian clients.