Air Quality Monitoring System [MONAQ]

⚠🌐Česká Verze ZDE!

---

A small backend system for collecting, storing, and displaying air quality data from sensors that use MQTT. This project provides a RESTful API to serve sensor data and is flexible enough to support any frontend, currently implemented using Streamlit.

This project was made as part of 2-Week "internship" at Datové centrum ústeckého kraje utilizing the data platform called Portabo

Note

Keep in mind that this was done in two weeks, so time was a bit pressing. Some parts or code could have been written/done better.

Table of Contents

• Screenshots
• Features
• Technologies Used
• Setup Instructions
  • Prerequisites
  • Installation
  • Running
• API Endpoints
• Frontend

Screenshots

Features

• MQTT Integration: Collects data from air quality sensors using the MQTT protocol.
• RESTful API: Serves the collected data via a FastAPI-based API.
• Data Storage: Utilizes MariaDB for storing historical air quality data.
• Data Analysis: Supports data analysis with Pandas and Plotly for data visualization.
• Flexible Frontend: Provides a flexible frontend interface with Streamlit, easily replaceable with any other technology.

Technologies Used

• Python: Core language for the backend.
• FastAPI: For creating RESTful APIs.
• MariaDB: For database storage.
• Docker: Containerization of the application.
• Paho-MQTT: For MQTT communication.
• Pandas: For data manipulation and analysis.
• Plotly: For creating interactive plots and graphs.
• Streamlit: (Current) Frontend framework.

Setup Instructions

Prerequisites

• Docker and Docker Compose (for containerized setup)
• MariaDB (optional)

Installation

1. Clone the repository:

   git clone https://github.com/DEPSTRCZ/MONAQ.git
   cd MONAQ

2. Configure Docker-Compose:

   Rename the example-docker-compose.yml to docker-compose.yml. Fill in all the enviroment variables.

   If wanted uncomment the MariaDB section for MariaDB server that can be used. (Follow step 2.1)

Important

Vendor does not need to be enabled! It just collects data and saves the to DB. So unless you want to use this seriously and give it data using MQTT you don't need to.

Note

MapBox api token is needed for the frontend to work.

2.1. Setting up sample data for database.:

• Connect to your database.
• Run/Import the sample-data.sql

That can be done using HeidiSQL, DBeaver (Or the included PhpMyAdmin that can be enabled in docker-compose.yml

Running

docker-compose up --build

Thats it! The project should be available:

• FrontEnd
• FastAPI Swagger UI

API Endpoints

• http://localhost:8002/getAllSensors Fetches all sensors from the database along with their latest data.

{
  "count": 2,
  "sensors": [
    {
      "temperature": "28.00",
      "updated_at": "2024-05-28T11:42:46",
      "sensor_id": 1,
      "co2": 852,
      "humidity": "43.00",
      "loc_lat": "0.000000",
      "loc_long": "0.000000"
    },
    {
      "temperature": "26.70",
      "updated_at": "2024-05-29T12:12:58",
      "sensor_id": 2,
      "co2": 724,
      "humidity": "40.00",
      "loc_lat": "0.000000",
      "loc_long": "0.000000"
    }
  ]
}

• http://localhost:8002/getSensor/{id} Fetches information about specified sensor.

{
  "sensor_id": 1,
  "times_posted": 1,
  "last_update": "2024-06-20T15:15:36",
  "records": [
    {
      "sensor_id": 1,
      "id": 25423,
      "co2": 1891,
      "humidity": "36.00",
      "loc_lat": "0.000000",
      "loc_long": "0.000000",
      "temperature": "28.50",
      "updated_at": "2024-06-20T15:15:36"
    }
  ]
}

• http://localhost:8002/getQualityInfo/{id} Fetches quality information from specified sensor. Including deltas.

{
  "humidity": "36.0",
  "temperature": "28.5",
  "co2": 1915,
  "delta_co2": 24,
  "delta_humidity": "0.0",
  "delta_temperature": "0.0"
}

Frontend

The current frontend is implemented using Streamlit. The BackEnd was made mostly flexible. So it should work with any backend.