- Overview
- Files
- Libraries
- PostgreSQL Database
- DbVisualiser
- Particle Electron Code
- The Python Code
- Further References
Currently, our project uses two Arduinos, a Particle Electron and a computer to get GPS coordinates from the Arduino to our database stored on a computer. To do that. The Arduino communicates with another Arduino via the LoRa Network adn transmits the data to it. The Arduino then uses serial connections (physical wires) to send the data to the ParticleElectron. The Particle Electron will publish an event, containing the time at which the event was published, the data (in this case, the GPS coordinates from the first Arduino) and which device it came from.
When the Particle Electron publishes an event, it is saved on the Particle Cloud and can be pulled from anywhere at any time. The final part of the system uses a computer to grab the data from the Particle Cloud and save it in a database. This database can then be used later, for e.g -
Presenting data in apps
Or presenting data on a website
This repository focuses on publishing and retreving events from the Particle Cloud
The files which you will be using are -
subscribePublishEvent.py
Publish_Event.ino
The subscibePublishEvent.py
file is used on the computer to store the data in a database. The Publish_Event.ino
file is flashed on the Particle Electron which allows to Particle Electron to publish events.
You can download the files by simply clicking the Clone or Download
button and selecting the .zip version. Once it has downloaded, you can simply go into the folder where the file has been downloaded and you should the folder Retreiving-Events-master
which contains all the files in it.
If you having some trouble with the Python file, especially some module not found errors, then run the following command in the terminal -
pip install -r requirements.txt.
You can find the requirements.txt file here - https://github.com/rudrathegreat/Retreiving-Events/blob/master/requirements.txt
You also need to install PostgreSQL. PostgreSQL allows you to make databases and search for data in databases. To install PostgreSQL, you simply go to the following link - https://www.postgresql.org
Download the .dmg (macOS) file and follow the installation instructions. After that, you should receive a username and password. REMEMBER THAT!!!
Next, open up PSQL (SQL Shell). You can do this by searching it in Spotlight Search (Command+Space
) -
or you can go to the PostgreSQL folder and run it from there (The PostgreSQL folder is usually in the Applications Folder). -
Inside it, it will first confirm your username and password. Just press enter if the parameter in the square brackets matches what you were going to give to psql. My username is postgres, so I am just going to hit enter to tell PSQL that I want to use the parameter inside the square brackets -
Once all of that is configured, you can run the following command to create a database -
create database mydb;
Once we have made our database, we need to copy our username, password and database name to the Python file and insert it this line -
connection = psycopg2.connect(host="localhost", database="{{database name}}", user="{{your username}}", password="{{your password}}")
The Python file is ready to go! Now we need a way to view the data in the database and for that, we will need an application called DbVisualiser.
DbVisualier allows you to view data inside database and search for data using queries. We can go to this link and download/install DbVisualiser - https://www.dbvis.com
Once you are done installing (remember to follow the instructions), open it up and create a new database connection. This database connection simple allows you to connect to that database.
Enter in your username in the UserId parameter, the password in the password parameter, the name of your database in the database parameter and the name of your database connection in the database connection parameter (see image above for reference). Change the port if necessary.
Once you are connected, you are good to go!
So basically the Particle Electron code is not to hard to understand. Basically, it is creating a bunch of variables and then sending their values to the Particle Cloud. It is also publishing an event which will then be recorded in the database. The main line is this -
Particle.publish("motion-detected", "Kangroo is moved!", PRIVATE);
This is the line that publishes the event.
The Python Code then obtains the event using get requests (basically asking the Cloud for the data) in the following line -
messages = SSEClient('https://api.particle.io/v1/devices/events/motion-detected?access_token=1c476acb47bd0b944a031e2859ef7160e4b72a66')
And then splitting it up into its constituent parts -
for msg in messages:
eventName = str(msg.event)#.encode('utf-8')
data = str(msg.data)#.encode('utf-8')
Then it will check if there is anything in data. If there is, then it will save in the database -
dataJson = json.loads(data)
print(dataJson)
eventData = dataJson["data"]
coreid = dataJson["coreid"]
published_at = dataJson["published_at"]
ttl = str(dataJson["ttl"])
save_data(eventName, eventData, coreid, published_at, ttl)
This is how the program works.
Currently, our project uses two Arduinos, a Particle Electron and a computer to get GPS coordinates from the Arduino to our database stored on a computer. To do that. The Arduino communicates with another Arduino via the LoRa Network adn transmits the data to it. The Arduino then uses serial connections (physical wires) to send the data to the ParticleElectron. The Particle Electron will publish an event, containing the time at which the event was published, the data (in this case, the GPS coordinates from the first Arduino) and which device it came from.
When the Particle Electron publishes an event, it is saved on the Particle Cloud and can be pulled from anywhere at any time. The final part of the system uses a computer to grab the data from the Particle Cloud and save it in a database. This database can then be used later, for e.g -
Presenting data in apps
Or presenting data on a website
This repository focuses on publishing and retreving events from the Particle Cloud
The files which you will be using are -
LoRa_Send.py
LoRa_Receive.py
The file LoRa_Send.py
is used to allow a LoRa module to send data over the LoRa Network. The LoRa_Receive.py
file however is used to receive data from the LoRa Network.
There are several modules which we are using, most of which is coded in an implementation of Python called MicroPython
. MicroPython is a language that is similar to Python that can be used for small microcontrollers.
We will also be using ftp
, a deprecated command if you are using the latest MacOS system. To install it, you need homebrew
, also known as brew
. To install brew, you can follow the documentation from here - https://brew.sh. Once you have done that, yuo need to install a package of utilities called inetutils
, which contains deprecated commands and programs, including ftp
. To do that, you simply type the following command in the terminal -
brew install inetutils
With this, you can access the PyCom Module, a module which can be used to connect, transmit and receive data from the LoRa Network, and view all the files inside. We need this to place all the files we want to run in there.
The first piece of the code is the LoRa_Send.py
file. This program allows the PyCom Module to connect and send messages on the LoRa Network. The program first creates a stable connection on the LoRa Network. After that, it sends a message over and over and over and over again... forever. This can be summed up in these three lines -
while True:
s.send('Ping')
time.sleep(1)
The receiving code is not too dissimilar. The program first creates a stable connection as with the previous program. It then waits until a message Ping
has been received. If it has, then it shall send Pong
back.
Ftp
is a program that allows you to see, view and edit files inside somebody else's device. There are some basic commands.
To start a ftp connection, just type -
ftp xxx.xxx.x.x
Next, you need to type the username and password for that device. After that, you can list all the directories by doing -
ftp> ls
You can check the current directory by doing -
ftp> pwd
You can place files by doing the following -
ftp> put filename.filetype
And then you can exit it by simply -
ftp> quit
To upload the code, we need to place our code from our computer/laptop to the PyCom module. To do that, we first need to rename our program to main.py
. This is because the program which the PyCom module executes is main.py
. The PyCom module cannot run any other program besides main.py
, unless we tell is to from boot.py
.
Next, we need to connect to the PyCom Router, which is built in the PyCom module, similar to how you connect to your WiFi Router. Next, go to the folder in which your main.py
file is in via the terminal. Then connect to the device via ftp in the terminal -
ftp xxx.xxx.x.x
After that, we need to go inside the flash
folder. Inside this folder, there are the files boot.py
and main.py
. We need to replace the main.py
file inside the flash
folder with your main.py
in your computer. To do that, simply use the put command -
put main.py
Now that the main.py
file in the PyCom module has been updated, we need to reboot the the pyCom module. Once that is done, the LED on the PyCom module should be changing colour and signals should be being sent on the LoRa Network.
And that's how you do it.
https://www.youtube.com/watch?v=fD7x8hd9yE4
print('Thanks for reading!')