EPF: TP IoT

This hands-on aims at setting up a mini IoT system in 3 hours, using Raspberry Pi and sensors available.

Prerequisite

Materials

Every group disposes following materials:

1 Raspberry Pi 2 Model B+ with 1 power adapter
1 screen
1 keyboard
1 mouse
1 Ethernet cable
1 HDMI cable
1 DHT11 temperature & humidity sensor
1 micro SD card
1 breadboard
Electronic jumpers
Plus: Wifi keys (if you guys want to play with the wifi)

Technologies

In order to succeed this Hands-on, you’ll need:

a CloudMQTT (https://www.cloudmqtt.com/) account
a MongoLab (https://mlab.com/home) account
basic knowledge of Node.js
basic knowledge of Python

Step 1: Prepare your Raspberry Pi

Step 1.1 Install OS for your Pi

Step 1.2 Set up development environment

Update and Upgrade you PI firmware:

$ sudo apt-get update
$ sudo apt-get upgrade

Install python environment:

$ sudo apt-get install build-essential python-dev

Step 2: Read data from the sensor

Step 2.1 Get to know your Raspberry PI GPIOs

Every group possesses a digital temperature and humidity sensor. Before starting wiring the jumbers, get to know the Raspberry Pi's GPIO (General Purpose Input/Output) pins or with some more interactive Pinout map.

Step 2.2 Make your electronic schema

Once you are ready, plug the sensor to your Raspberry Pi:

VCC to physical pin 2
GND to physical pin 6
Input to GPIO23 (physical pin 16)

Step 2.3 Read the values

The sensor we provide is a DHT22 type of digital temperature & humidity sensor. Install the Adafruit Python DHT library to read the value of the sensor.

You can either start coding directly on the raspberry pi or use ssh connection to edit the script from your own computer (configure your Pi to enable ssh connection). Refer to /embedded/sensor.py for reference:

import Adafruit_DHT

# Sensor should be set to Adafruit_DHT.DHT22
sensor = Adafruit_DHT.DHT22
sensor_id = 1

# Example using a Raspberry Pi with DHT sensor
# connected to GPIO23.
sensor_pin = 23

# Create a loop to read the sensor data
while True:
  # Try to grab a sensor reading.  Use the read_retry method which will retry up
  # to 15 times to get a sensor reading (waiting 2 seconds between each retry).
  humidity, temperature = Adafruit_DHT.read_retry(sensor, sensor_pin)
  if humidity is not None and temperature is not None:
    temp_fmt = round(temperature, 1)
    hum_fmt = round(humidity, 1)
    print 'Temp=%s*C  Humidity=%s%%' % (temp_fmt, hum_fmt)
  else:
    print 'Failed to get reading. Try again!'

  time.sleep(30)

Launch your sensor script by

$ sudo python sensor.py

Step 3: Build the pipeline

This part focus on explaining how we use different cloud services to build the data pipeline.

3.1 Publish to MQTT

One of the most popular open source MQTT broker is Mosquitto. The service CloudMQTT we choose for this hands-on are managed Mosquitto servers in the cloud.

Before pushbling data to the MQTT broker, you have to define:

the topic name
the data structure

Paho

As an MQTT client, you can use Paho which is a good MQTT client.

Install the library:

$ sudo pip install paho-mqtt

Initialize the MQTT client and publish data on the define topic:

import paho.mqtt.client as mqtt
import urlparse

# Init MQTT client
mqttClient = mqtt.Client()
# e.g. mqtt://username:password@m20.cloudmqtt.com:port
mqttConnectionString = ""
url = urlparse.urlparse(mqttConnectionString)
mqttClient.username_pw_set(url.username, url.password)
mqttClient.connect(url.hostname, url.port)

# TODO publish data

3.2 Subscribe to MQTT

To verify the data is well received by the MQTT broker. Create a simple node.js application using mqtt.js to subscribe to the topic that you previously defined and listen to the message:

var mqtt = require('mqtt');
var mqttClient = mqtt.connect("");
mqttClient.on('connect', function () {
  console.log("MQTT connected.")
  // TODO subscribe to the topic
});
 
mqttClient.on('message', function (topic, message) {
  console.log("Topic: " + topic);
  console.log("Message: " + message.toString());
});

Step 4: Store data

A very important part of IoT system is to collect data for future analysis. So now it's time to store them somewhere. In the example, we choose MongoDB to finish the simple data insertion and query:

var mongoClient = require('mongodb').MongoClient;
mongoClient.connect(mongodbUrl, function(err, db) {
  if(!err) {
    console.log('MongoDB connected');
    // TODO create collection & retrieve the db object
  }
});

mqttClient.on('message', function (topic, message) {
  console.log("Topic: " + topic);
  console.log("Message: " + message.toString());
  // TODO store data to database
});

var insertSensorData = function(db, data, cb) {
	// TODO insert data to db
};
var findLastSensorDataItem = function(db, cb) {
	// TODO fetch the last inserted data item
};

Step 5: Display the data

For the data visualization part, we prepare a simple REST service that allows any type of dashboard to fetch data and display, using the Express framework.

var express = require('express');
var app = express();

// ....

// REST API
app.get('/sensor', function (request, response) {
  response.header('Access-Control-Allow-Origin', '*');
  response.header('Access-Control-Allow-Methods', 'GET,PUT,POST,DELETE');
  response.header('Access-Control-Allow-Headers', 'Content-Type');
  response.header('Content-Type', 'application/json');
  // TODO fetch data and send as response
});

app.listen(3000, function () {
  console.log('App listening on port 3000!');
});

samehBF / iot-hands-on