Installing and Testing Mosquitto MQTT Broker on Raspberry Pi for IoT Communication

MQTT is a protocol used to send and receive messages over internet. We previously used this protocol in Iot Electricity meter and Raspberry Pi Alexa to publish the data on internet. In this tutorial we will get to know more about the MQTT and the terms related to it. Here we will use the Raspberry Pi as local MQTT broker and control an LED connected to NodeMCU ESP12E through MQTT application dashboard. A DHT11 sensor is also connected to NodeMCU so we get the reading of temperature and humidity on MQTT dashboard, by again using the Raspberry Pi as local MQTT broker.

So, let’s start by understating the MQTT and the terms related to it.

 

What is MQTT?

MQTT stands for Message Queue Telemetry Transport which is designed by IBM. This protocol is simple and lightweight which is used to send and receive messages over internet and it is designed for the devices which have low bandwidth uses. Nowadays this protocol is frequently used in the IoT devices to send and receive the sensors data. Also, In IoT based home automation systems, this protocol can easily be used without using much of internet data.

There are few terms which are used frequently in MQTT:

1. Subscribe and Publish
2. Message
3. Topic
4. Broker 

 

1. Subscribe and Publish: Subscribe means to get the data from other device, and publish means to send the data to other device.

When device1 sends the data to device2 then it is known as Publisher and another is Subscriber and vice-versa.

 

2. Message: Messages are the information that we are sending and receiving. It can be a data or any type of command. For example, if we are publishing the temperature data to the cloud then this temperature data is known as Message.

 

3. Topic: This is the way you register interest for incoming messages or how you specify where you want to publish the message. Topics are represented with strings separated by a forward slash. Data is published on the topics using MQTT and then MQTT device subscribe to the topic to get the data.

 

4. MQTT Broker: This thing is responsible for receiving all messages from publishers, filters the messages and then publishing the messages to the subscribers who are interested in them.

 

When this broker is hosted on the cloud then it is called MQTT cloud. There are many cloud based MQTT services like Adafruit IO, MQTT.IO, IBM bluemix, Microsoft Azure, etc. MQTT can also be used with popular Amazon AWS cloud, which we have explained in Getting Started with Amazon AWS tutorial.

 

We can make our own MQTT broker using the Raspberry Pi. This will be the local MQTT broker i.e. you can send and receive the data on your local network only not from anywhere. So here we will install Mosquitto MQTT broker in Raspberry Pi to make it local MQTT broker and will send the temperature data from the NodeMCU to MQTT dashboard application. Also, we will control an LED connected to NodeMCU using the broker.

 

Installing Mosquitto MQTT Broker on Raspberry Pi

Open the terminal in your Raspberry pi and type the following commands to install the broker

sudo apt update
sudo apt install -y mosquitto mosquitto-clients

 

Wait for the installation to finish. Now, to start the broker on startup of raspberry pi, Type the following command

sudo systemctl enable mosquitto.service

 

That’s it, we are all set to launch our MQTT broker. To check it is installed properly, enter the following command

mosquitto -v

This command will give you the version of your MQTT broker. It should be 1.4.x or above.

 

Testing the Raspberry Pi Mosquitto Broker

1. Run the Mosquitto  broker in background using the below command

mosquitto -d

 

2. Now, we will subscribe a topic in exampleTopic using following command

mosquitto_sub -d -t exampleTopic

 

3. Now, we will publish some message to exampleTopic

mosquitto_pub -d -t exampleTopic -m "Hello world!"

 

You will receive the Hello world! Message in the subscriber terminal.

 

Now, it’s time to control and getting the data from another device in our case we are using NodeMCU and MQTT dashboard application.

• First we will control an LED by sending command using App so in this case NodeMCU behave as a subscriber and App as a publisher.
• Then ESP12E also has DHT11 sensor connected to it and it sends this temperature reading to Mobile MQTT application so in this case mobile will be the subscriber and NodeMCU will be the publisher. And to forward these messages on the respective Topics, the Mosquitto MQTT broker is used.

 

Circuit Diagram

 

Connect the circuit as shown in the diagram. Here the DHT11 is used for the temperature readings but a LM35 Temperature sensor can also be used. We already used DHT11 sensor in many of our projects including with NodeMCU to build a weather station.

 

Lets start writing the code for the NodeMCU to subscribe and publish the data.

 

Code and Explanation

For this download the Adafruit MQTT library from Sketch -> Include Library -> Manage Libraries. Search for Adafruit MQTT and install it. After installing the library. Go to examples -> Adafruit mqtt library -> mqtt_esp8266 

 

Then edit this code according to our Raspberry Pi IP address and Wi-Fi credentials.

Include all the libraries for ESP8266WIFI and Adafruit MQTT.

#include <ESP8266WiFi.h>
#include "Adafruit_MQTT.h"
#include "Adafruit_MQTT_Client.h"
#include "DHT.h"

 

Then define the SSID and Password for your Wi-Fi, from which you want to connect your ESP-12e. Make sure your RPi and NodeMCU connect with the same network.

#define WLAN_SSID       "xxxxxxxx"
#define WLAN_PASS       "xxxxxxxxxxx"

 

This section defines the Adafruit server, in this case the IP address of your Raspberry Pi and server port.

#define AIO_SERVER      "ip address of your Pi"
#define AIO_SERVERPORT  1883                  

 

Below fields will remain empty because we are not using the Adafruit cloud.

#define AIO_USERNAME    ""
#define AIO_KEY         "" 

 

Then create an ESP8266 WiFiClient class to connect to the MQTT server.

WiFiClient client;

 

Setup the MQTT client class by passing in the WiFi client and MQTT server and login details.

Adafruit_MQTT_Client mqtt(&client, AIO_SERVER, AIO_SERVERPORT, AIO_USERNAME, AIO_KEY);

 

Setup a feed called 'Temperature' for publishing the temperature.

Adafruit_MQTT_Publish Temperature = Adafruit_MQTT_Publish(&mqtt, AIO_USERNAME "/feeds/temperature");

 

Setup a feed called 'led1' for subscribing to changes.

Adafruit_MQTT_Subscribe led1 = Adafruit_MQTT_Subscribe(&mqtt, AIO_USERNAME "/feeds/led");

 

In setup function, we declare PIN of NodeMCU on which you want to get output. Then, connect NodeMCU to Wi-fi access point.

void setup() {
  Serial.begin(115200);
  delay(10);
  pinMode(LED, OUTPUT);
  Serial.println(F("Adafruit MQTT demo"));
  // Connect to WiFi access point.
  Serial.println(); Serial.println();
  Serial.print("Connecting to ");
  Serial.println(WLAN_SSID);
  WiFi.begin(WLAN_SSID, WLAN_PASS);
  while (WiFi.status() != WL_CONNECTED) {
….
….
…
Setup MQTT subscription for led feed.
  mqtt.subscribe(&led1);
}

 

In loop function, we will ensure that the connection to the MQTT server is alive using MQTT_connect(); function.

void loop() {
  MQTT_connect();

 

Now, subscribe ‘led’ feed and get the string from MQTT broker and convert this string to number using atoi(); function and write this number to LED pin using digitalWrite(); function.

Adafruit_MQTT_Subscribe *subscription;

  while ((subscription = mqtt.readSubscription(20000))) {
    if (subscription == &led1) {
      Serial.print(F("Got: "));
      Serial.println((char *)led1.lastread);
      int led1_State = atoi((char *)led1.lastread);
      digitalWrite(LED, led1_State);
    } 

 

Now, get the temperature in a variable and publish this value using Temperature.publish(t) function.

float t = dht.readTemperature();
…
..

  if (! Temperature.publish(t)) {
    Serial.println(F("Failed"));
  } else {
    Serial.println(F("OK!"));
  }

 

Full code with a demonstration video is given at the end of this tutorial. Upload the code to NodeMCU board and open the MQTT dashboard app that you have downloaded in the Smartphone.

You can also control Raspberry Pi GPIO from anywhere in world using MQTT cloud like Adafruit IO, MQTT.IO etc, which we will learn in next tutorial.