Embedded Systems for IoT Notes

Embedded System

An embedded system is a computer system designed to perform a fixed function inside a larger electrical or mechanical system. Unlike a desktop computer, it is usually hidden inside the product and is optimized for reliability, cost, power, size, and response time.

• Input: buttons, sensors, switches, camera, microphone, network data.
• Processing: microcontroller, microprocessor, SoC, memory, firmware.
• Output: display, motor, relay, buzzer, LED, actuator, cloud message.
• Control: the system repeatedly reads inputs, applies logic, and produces output.

Applications

Embedded systems are everywhere because modern products need sensing, control, communication, and automation.

Embedded Software

Embedded software is the program stored in the device. It may be bare-metal firmware, an RTOS based application, or a Linux based program. It controls hardware registers, reads sensors, handles interrupts, performs calculations, and communicates data.

loop forever:
  read sensor value
  compare value with limit
  if value is unsafe:
    turn on alarm
    send alert message
  wait for next sample time

Embedded System Design Steps

1. Define requirements and constraints.
2. Select processor, memory, sensors, and communication modules.
3. Design hardware circuit and power supply.
4. Write firmware and driver code.
5. Test timing, safety, accuracy, and power consumption.
6. Deploy, maintain, and update when required.

System on Chip (SoC)

A System on Chip is an integrated circuit that contains several computer components on one chip. A typical SoC may include CPU cores, GPU, memory controller, I/O controllers, timers, ADC, security blocks, communication interfaces, and power management.

• Reduces board size because many functions are inside one chip.
• Reduces power consumption by optimizing internal communication.
• Improves speed because internal buses are faster than external wiring.
• Used in mobile phones, Raspberry Pi boards, routers, cameras, and IoT gateways.

SoC Internal Blocks

Embedded Hardware Design

Embedded hardware design means selecting electronic components and connecting them so that the system meets functional, timing, power, size, and cost requirements.

1. Choose MCU, MPU, or SoC based on processing need.
2. Select sensors, actuators, display, and communication interface.
3. Design power supply, voltage regulation, and protection.
4. Create schematic and PCB layout.
5. Prototype, measure signals, and correct faults.

Embedded Hardware Development

Hardware development is iterative. Engineers build prototypes, test with instruments, update circuits, and validate final boards before mass production.

Introduction to Raspberry Pi

Raspberry Pi is a small, affordable single-board computer. It runs Linux, supports USB, HDMI, Wi-Fi, Ethernet on many models, and provides GPIO pins for connecting sensors and actuators. It is useful when an IoT project needs more processing power than a microcontroller.

• Runs a full operating system such as Raspberry Pi OS.
• Can host a web server, database, dashboard, and automation tools.
• Supports Python, C, Node.js, PHP, shell scripts, and many IoT libraries.
• Can connect to external circuits through GPIO, I2C, SPI, UART, PWM, and USB.

When to Use Raspberry Pi

On-board Components

• SoC: main processor and internal controllers.
• RAM: temporary memory for programs and OS.
• MicroSD slot: stores operating system and files.
• USB ports: keyboard, mouse, storage, camera, adapters.
• HDMI: connects display or monitor.
• Ethernet and Wi-Fi: network connection for IoT data.
• GPIO header: connects external electronics.

Useful Commands

# update package list
sudo apt update

# check GPIO command availability
pinout

# show network address
hostname -I

# enable common interfaces
sudo raspi-config

Temperature and Humidity Sensor: DHT11

DHT11 is a low-cost digital sensor that measures temperature and relative humidity. It uses one data pin. It is slower and less accurate than DHT22, but it is common in beginner IoT labs.

import time
import board
import adafruit_dht

dht = adafruit_dht.DHT11(board.D4)

while True:
    temperature = dht.temperature
    humidity = dht.humidity
    print(temperature, humidity)
    time.sleep(2)

Motion Sensor: PIR

PIR means Passive Infrared sensor. It detects motion by sensing changes in infrared radiation from warm objects such as humans. It is widely used in security lights, automatic doors, and room occupancy detection.

• VCC: usually 5 V on common PIR modules.
• OUT: digital signal, HIGH when motion is detected.
• GND: ground connection.
• Delay and sensitivity: adjustable on many modules.

from gpiozero import MotionSensor
from signal import pause

pir = MotionSensor(17)

pir.when_motion = lambda: print("Motion detected")
pir.when_no_motion = lambda: print("No motion")

pause()

Obstacle Detection Using Ultrasonic Sensor

HC-SR04 ultrasonic sensor measures distance by sending a sound pulse and measuring the echo time. Raspberry Pi GPIO is 3.3 V, so the echo pin often needs a voltage divider because many HC-SR04 modules output 5 V.

from gpiozero import DistanceSensor
from time import sleep

sensor = DistanceSensor(echo=24, trigger=23)

while True:
    print("Distance:", round(sensor.distance * 100, 2), "cm")
    sleep(1)

Communicating Using GPIO Interfacing

GPIO pins can work as digital input or output. They can also support special functions.

Accessing On-board Wi-Fi

On-board Wi-Fi connects Raspberry Pi to a router or hotspot. After connection, the Pi can send sensor data to a server, cloud dashboard, or local web page.

# scan visible Wi-Fi networks
nmcli dev wifi list

# connect to Wi-Fi
sudo nmcli dev wifi connect "NetworkName" password "Password"

# test internet connectivity
ping -c 4 ashishvegan.com

Connecting Database with Raspberry Pi

A Raspberry Pi can store sensor readings in SQLite, MySQL, MariaDB, or a simple JSON file. SQLite is useful for small local projects because it stores data in one file and does not need a separate database server.

import sqlite3
from datetime import datetime

db = sqlite3.connect("sensor_data.db")
db.execute("""CREATE TABLE IF NOT EXISTS readings (
  id INTEGER PRIMARY KEY AUTOINCREMENT,
  created_at TEXT,
  temperature REAL,
  humidity REAL
)""")

db.execute(
  "INSERT INTO readings(created_at, temperature, humidity) VALUES (?, ?, ?)",
  (datetime.now().isoformat(), 30.5, 62.0)
)
db.commit()
db.close()

LAMP Web Server

LAMP means Linux, Apache, MySQL or MariaDB, and PHP. On Raspberry Pi, LAMP allows students to create local web dashboards that read data from sensors and display it in a browser.

sudo apt update
sudo apt install apache2 php mariadb-server
sudo systemctl enable apache2
sudo systemctl start apache2

GPIO Control Over Web Browser

A browser button can send a request to a server script. The script can run safe GPIO code to switch a relay, LED, or appliance. For real projects, add authentication and electrical isolation before controlling high voltage loads.

<button onclick="fetch('/relay.php?state=on')">ON</button>
<button onclick="fetch('/relay.php?state=off')">OFF</button>

Communicating Data Using On-board Module

Raspberry Pi can use on-board Wi-Fi, Ethernet, and Bluetooth for data communication. Common protocols are HTTP for web APIs, MQTT for lightweight messaging, and WebSocket for live dashboards.

Home Automation Using Raspberry Pi

Home automation uses sensors, relays, and software rules to control electrical devices. Raspberry Pi can provide schedules, web control, voice assistant integration, and remote monitoring.

• Use relay modules with opto-isolation for mains devices.
• Separate low-voltage electronics from high-voltage wiring.
• Add manual override and emergency shutdown.
• Use authentication for web control pages.

MQTT Protocol

MQTT is a lightweight messaging protocol based on publish and subscribe. Devices publish data to topics. Other devices or dashboards subscribe to topics and receive messages through a broker.

import paho.mqtt.publish as publish

publish.single(
  "home/room1/temperature",
  payload="30.5",
  hostname="localhost"
)

Using Node-RED Visual Editor

1. Install or open Node-RED on Raspberry Pi.
2. Open the editor in a browser using port 1880.
3. Drag input, function, output, MQTT, and dashboard nodes.
4. Connect nodes to create a flow.
5. Deploy the flow and test messages.

# start Node-RED
node-red-start

# open in browser
http://raspberrypi.local:1880