IoT Microcontrollers and SBCs
The brain of any IoT device is either a microcontroller or a single-board computer. These two types of hardware handle the processing, run the code, and coordinate all the other components. Choosing the right one depends on what your IoT application needs to do.
Microcontrollers vs. Single-Board Computers
Think of a microcontroller (MCU) as a bicycle — lightweight, efficient, and good for a specific purpose. A single-board computer (SBC) is like a small car — more powerful, more versatile, but uses more energy and costs more.
+-----------------------------+------------------------------+ | Microcontroller (MCU) | Single-Board Computer (SBC) | +-----------------------------+------------------------------+ | Simple, low-power chip | Full mini-computer | | No operating system needed | Runs Linux or similar OS | | Best for repetitive tasks | Best for complex processing | | Runs one program at a time | Runs multiple programs | | Very low power (milliwatts) | Higher power (watts) | | Low cost ($2–$15) | Moderate cost ($10–$80) | | Example: Arduino, ESP32 | Example: Raspberry Pi | +-----------------------------+------------------------------+
Popular Microcontrollers in IoT
Arduino Uno
Arduino is the most beginner-friendly platform for learning IoT hardware. The Arduino Uno uses an ATmega328P chip running at 16 MHz with 32 KB of memory. You write code in a simplified version of C++, upload it, and the board runs that single program forever until you change it.
Best for: learning, simple sensor reading, controlling LEDs and motors, hobby projects.
Limitation: no built-in Wi-Fi or Bluetooth — you must add extra modules.
ESP8266
The ESP8266 brought affordable Wi-Fi to microcontrollers. It costs under $3 and connects to Wi-Fi natively. An IoT project that needs to send data to a cloud server can do so directly from this chip without any extra hardware.
Best for: Wi-Fi-connected sensors, simple home automation devices, low-cost cloud reporting.
ESP32
The ESP32 is the upgraded version of the ESP8266. It has dual-core processing, built-in Wi-Fi, built-in Bluetooth, and more input/output pins. It handles more complex tasks while still using very little power — making it the most popular choice for IoT product development today.
Best for: smart home devices, BLE beacons, industrial sensors, wearables, battery-powered IoT products.
STM32 Series
STM32 microcontrollers come from STMicroelectronics and are widely used in professional and industrial IoT products. They offer high processing speed, precise analog-to-digital conversion, and rich peripheral support — features that matter for medical devices and industrial automation.
Best for: industrial sensors, medical IoT, automotive applications, precision measurement.
nRF52 Series (Nordic Semiconductor)
The nRF52 chips specialize in Bluetooth Low Energy (BLE) and are the go-to choice for wearables and health monitors. They draw almost no power from a battery, making them ideal for devices that need to run for months or years on a coin cell.
Best for: fitness trackers, medical wearables, asset tracking tags, BLE beacons.
How a Microcontroller Runs IoT Code
A microcontroller runs a continuous loop. The code performs three jobs over and over:
SETUP (runs once at startup) | v LOOP (repeats forever) | +--> Read sensor | +--> Process the reading | +--> Send data / trigger actuator | +--> Wait (delay to save power) | +---> (back to top of LOOP)
This simple structure handles most IoT tasks. The microcontroller wakes up, reads a sensor, decides whether to act or transmit, then goes back to waiting. Keeping the wait time as long as possible saves battery life — a critical concern in field-deployed sensors.
Popular Single-Board Computers in IoT
Raspberry Pi
The Raspberry Pi is the most recognized SBC worldwide. It runs a full Linux-based operating system (Raspberry Pi OS) and can run Python scripts, web servers, databases, and computer vision applications. A Raspberry Pi 4 has 1–8 GB of RAM and a quad-core 1.5 GHz processor — similar to a budget laptop from a decade ago, but the size of a credit card.
Best for: IoT gateways, video surveillance, machine learning at the edge, multi-sensor hubs, prototyping complex systems.
BeagleBone Black
BeagleBone Black targets industrial IoT and real-time control applications. It has a dedicated real-time processing unit (PRU) that can handle time-critical tasks with microsecond precision — important for controlling industrial machinery.
Best for: industrial IoT, real-time sensor control, robotics.
NVIDIA Jetson Nano
The Jetson Nano is an SBC with a GPU (graphics processing unit) built in. This allows it to run deep learning and computer vision models locally — without sending data to the cloud. It is used in applications where AI decisions must happen immediately at the edge.
Best for: AI-powered cameras, object detection, autonomous robots, edge AI systems.
Choosing Between MCU and SBC
Decision Flowchart:
Does the task require an operating system, complex software,
or running multiple programs simultaneously?
|
YES | NO
|
Use SBC Does the device run on battery power for a long time?
|
YES | NO
|
Use MCU Does the device need Wi-Fi or BLE without extra modules?
|
YES | NO
|
ESP32 Arduino / STM32
MCU and SBC Working Together
Many real-world IoT systems use both. A microcontroller handles the low-level sensor reading and actuator control — because it never crashes and uses little power. A single-board computer sits nearby as the local gateway or edge processor — because it handles complex logic, connects to the cloud, and runs data analysis.
Example: A smart building system uses 30 ESP32 microcontrollers installed throughout the building. Each one reads temperature, humidity, and occupancy sensors every minute. They send data over Wi-Fi to a Raspberry Pi gateway on each floor. The Raspberry Pi collects all the sensor data, runs occupancy analysis, and pushes summaries to the cloud every 10 minutes.
Summary
Microcontrollers are efficient, low-power chips best for repetitive sensing and control tasks. Single-board computers are compact but powerful mini-computers that run full operating systems and handle complex processing. The ESP32 and Arduino are the most common MCUs for IoT. The Raspberry Pi is the most common SBC for IoT gateways and edge processing. Many systems combine both types to get the benefits of each.