For many Internet of Things (IoT) projects a message queuing system like MQTT (Message Queue Telemetry Transport) is all that is needed to connect sensors, devices and graphic interfaces together. If however you require a database, with sorting, queuing and multi-media support then there are some great cloud storage platforms that are available. One that is definitely worth taking a look at is Google Firebase.

Like any IoT solution, Google Firebase can have inputs and sensors sending data directly into it, and a variety of client applications to view the data (Figure 1), but Google Firebase also offers other features such as: file storage, machine learning, messaging, and server side functions. In this article I will look at:

In this article I will look at:

1. Setting up a sample Firebase IoT project
2. Use Python to simulate I/O data
3. Create a Web dashboard with Node-Red to view and write data
4. Create an Android app using AppInventor to view and write data
5. Look at a more complex data monitoring example in Python

Getting Started

Google Firebase [1] may not have the huge variety of options that Amazon Web Services (AWS) has, but I found as an IoT engineer Google Firebase had all the features that I needed to get up and running quickly, and I didn’t need a credit card for my free activation.

To begin log in with your Google account at https://firebase.google.com/, and select “Get Started”. I created a sample project called My-IoT. After the project is created Firebase will give it a unique identifier name, such as: my-iot-4ded7.

Firebase data can be stored in either a Realtime Database, or a Cloud Firestore. The Realtime Database is an unstructured NoSQL format that has been around for about 4 years, so there are lots of 3rd party components. The Cloud Firestore stores the data in structure collections and it’s fairly new so the 3rd party options are a little more limited.

For our test IoT project we will use the  real-time database, and it can be created from the Database -> Data menu option.

The database structure can be built directly from the Firebase web console or imported from a JSON file. For this database 3 grouping are defined, one for Android, Node-Red and Raspberry Pi I/O values. The Firebase console allows for setting, changing and viewing the database values.

The security is configured in the Database -> Rules menu option. To keep things simple for testing read and write security is set to true for public access,  (remember to change this for a production system).

The project’s remote access settings are shown in the Authentication section with the “Web setup” button.

Python and Firebase

There are a number of Python libraries to access Firebase. The pyrebase library has some added functionality such as: queries, sorting, file downloads and streaming support. The pyrebase library only support Python 3, and it is installed by:

sudo pip3 install pyrebase
sudo pip3 install --upgrade google-auth-oauthlib

A Python 3 test program to set two values (pi/pi_value1 and pi/pi_value2) and read a single point and a group of point is shown below. Remember to change the configuration settings to match your project settings.

import pyrebase, random

# define the Firebase as per your settings
config = {
  "apiKey": "AIzaSyCq_WytLdmOy1AIzaSyCq_WytLdmOy1",
  "authDomain": "my-iot-4ded7.firebaseapp.com",
  "databaseURL": "https://my-iot-4ded7.firebaseio.com",
  "storageBucket": "my-iot-4ded7.appspot.com"
}

firebase = pyrebase.initialize_app(config)
db = firebase.database()

# set 2 values with random numbers
db.child("pi").child("pi_value1").set(random.randint(0,100))
db.child("pi").child("pi_value2").set(random.randint(0,100))

# readback a single value
thevalue = db.child("pi").child("pi_value1").get().val()
print ("Pi Value 1: ", thevalue)

# get all android values
all_users = db.child("android").get()
for user in all_users.each():
    print(user.key(), user.val())

The Firebase web console can be used to check the results from the test program.

Node-Red and Firebase

Node-Red is a visual programming environment that allows users to create applications by dragging and dropping nodes on the screen. Logic flows are then created by connecting the different nodes together.

Node-Red has been preinstalled on Raspbian Jesse since the November 2015. Node-Red can also be installed on Windows, Linux and OSX.  To install and run Node-Red on your specific system see https://nodered.org/docs/getting-started/installation.

To install the Firebase components, select the Manage palette option from the right side of the menu bar. Then search for “firebase” and install node-red-contrib-firebase.

For our Node-Red example we will use web gauges to show the values of our two Python simulated test points. We will also send a text comment back to our Firebase database. The complete Node-Red logic for this is done in only 6 nodes!

To read the Pi values two firebase.on() nodes are used. The output from these nodes is connected to two dashboard gauge nodes. Double-clicking on firebase_on() node to configure Firebase database and the item to read from. Double-clicking on the gauge node allows you to edit gauge properties.

 To send a text string to Firebase a text input node is wired to a firebase modify node. Edit the firebase modify node with the correct database address and value to set.

After the logic is complete, hit the Deploy button on the right side of the menu bar to run the logic.  The Node-Red dashboard user interface is accessed by: http://ipaddress:1880/ui, so for example 192.168.1.108:1880/ui.

AppInventor

AppInventor is a Web based Android app creation tool (http://appinventor.mit.edu), that uses a graphical programming environment.

AppInventor has two main screens. The Designer screen is used for the layout or presentation of the Android app and the Blocks screen is used to build the logic. On the right side of the top menu bar, the Designer and Blocks buttons allow you to toggle between these two screens.

On the Designer screen, an app layout is created by dragging a component from the Palette window onto the Viewer window.

We will use AppInventor to create an Android app that will read Pi values from our Firebase IoT database, and write a value back.

For the visuals on this application a Button, Label (2),  ListView, Textbox and FirebaseDB component will be used. After a component is added to the application, the Components window is used to rename or delete that component. The Properties window is used to edit the features on a component. For the FirebaseDB component configure the Token and URL to match with your project settings.

Once the layout design is complete, logic can be added by clicking on the Blocks button on the top menu bar.

Logic is built by selecting an object in the Blocks window, and then click on the specific block that you would like to use.

AppInventor is pretty amazing when it comes to doing some very quick prototyping. Our entire app only uses two main blocks.

The when FirebaseDB1.DataChanged block is executed whenever new data arrives into the Firebase database. The DataChanged block returns a tag and a value variable. The tag variable is the top level item name, (“pi” or “nodered” for our example). The value will be a string of the items and their values, for example: “{pi_value2 = 34, pi_value1=77}”. We use a replace all text block to remove the “{” and “}” characters, then we pass the string to the ListView component. Note this same code will pass 2 or 200 tags in the pi tag section.

The when BT_PUT.click block will pass the text that we enter on the screen into the android/and_value1 item in our database.

After the screen layout and logic is complete, the menu item Build will compile the app. The app can be made available as an APK downloadable file or as a QR code link.

Below is picture of the final Android app synced with the Firebase data.

Python Data Monitoring Example

Our starting IoT Firebase realtime database was quite simple. Data monitoring or SCADA (Supervisory Control and Data Aquisition) projects usually require more information than a just a value. A more realistic sensor database would includes fields such as tag name, description, status, time and units.

By adding some indexing in the Firebase Security Rules it is possible to create some queries and sorts of the data. Some typical queries would be: “Points in alarm” or “Point values between 2:00 and 2:15”.

In the code listing below the syntax statement of .order_by_child(“status”).equal_to(“ALARM”).get() is used to show only records that have a status = ALARM. Some other filter options include: .start_at(time1).end_at(time2).get, .limit_to_first(n), and .order_by_value().

A good next step would be to pass important filtered information to Google Firebase’s messaging feature.

import pyrebase, random

# define the Firebase as per your settings
config = {
  "apiKey": "AIzaSyCq_AIzaSyCq_Wyt",
  "authDomain": "my-iot-7ded7.firebaseapp.com",
  "databaseURL": "https://my-iot-7ded7.firebaseio.com",
  "storageBucket": "my-iot-7ded7.appspot.com"
}

firebase = pyrebase.initialize_app(config)
db = firebase.database()

tag_sum = db.child("pi").order_by_child("status").equal_to("ALARM").get()

# print alarm points
for tag in tag_sum.each():
    info = tag.val()
    print (tag.key()," - ", info["description"],":", info["value"],info["units"], info["status"])

For more Python examples see the Pyrebase documentation.

Summary

Without a lot of configuration it is possible to configure a Google Firebase project and have Python, Node-Red and Android app read and write values. The level of programming complexity is on par with an MQTT implementation, however Google Firebase can offer a lot more future functionality that you wouldn’t have with MQTT, such as file storage, machine learning, messaging, and server side functions.

At the time of writing this article I found the Arduino Firebase library to be unreliable but Google offers some options like an MQTT bridge.