TUFAN: Building an AI-Powered Urban Intelligence Platform with AWS for Cleaner and Safer Cities
TUFAN is an IoT-based urban monitoring platform combining indoor and outdoor sensing modules with a centralized government dashboard to collect, visualize, and analyse environmental data for smarter urban management.
TUFAN: Building an Indoor and Outdoor Environmental Monitoring System
Introduction
The idea behind TUFAN started with a simple problem. We can collect environmental data using sensors, but how do we make that data useful to the people who need it?
Air quality can change from one location to another. Conditions inside a room may be different from those outside the building. If readings are collected at different places without a common platform, comparing them becomes difficult.
Our team at the University of Engineering and Management, Kolkata, is working on TUFAN, an IoT based urban monitoring project that brings indoor monitoring, outdoor monitoring and a government dashboard into one system.
We want to make it easier to collect environmental readings and view them in one place.
The two monitoring modules
We are developing two types of monitoring modules for TUFAN. One is intended for indoor use and the other is designed for outdoor environments.
The indoor module
The indoor module is designed to monitor environmental conditions inside buildings and enclosed spaces. It uses a microcontroller based approach so that sensors can collect readings and pass them to the software system.
The exact measurements depend on the sensors installed in the module. The aim is to give users a way to observe conditions over time instead of relying on occasional manual readings.
For example, a building administrator could use the system to review available readings from different rooms and identify areas that may need further investigation.
Our presentation estimates the indoor module cost at Rs. 4,500. This is the current project estimate and may change as the hardware and enclosure are refined.
The outdoor module
The outdoor module is intended for collecting environmental data in open spaces. The proposed hardware includes an ESP32 microcontroller, a PMS7003 particulate matter sensor and a LoRa wireless communication module.
The PMS7003 measures particulate matter, including PM2.5 and PM10. The ESP32 handles the sensor integration, while LoRa can support communication between compatible devices over a wireless link.
This setup could be useful when monitoring locations are spread across an area and connecting every device directly to a WiFi network is not practical.
There are still some things to work through before outdoor deployment. The enclosure needs to protect the electronics from weather, the power supply must be reliable and the sensors need to be checked for measurement accuracy. We also need to consider how readings will reach the central server.
The estimated cost of our outdoor module is Rs. 1,550, based on our current project presentation. This figure should be treated as a prototype estimate rather than the final cost of a complete outdoor installation.
Why we need a common dashboard
Collecting readings is only part of the work. We also need a convenient way to view them.
The TUFAN government dashboard is intended to bring data from connected monitoring modules into a single web interface. An authorized user should be able to see available readings, check the location of a monitoring device and review previous observations.
Over time, the dashboard could help users compare environmental conditions between locations and look for changes in the collected data. It could also support reports for municipal teams and other organizations involved in environmental monitoring.
We want the dashboard to make the information understandable to someone who does not need to work directly with the hardware or read raw sensor output.
How the system is expected to work
The process begins with the monitoring module collecting a reading from its connected sensor. The device then prepares the data for transmission.
Depending on the final hardware setup, the reading may travel through WiFi or through a LoRa connection to a suitable gateway. The gateway or connected backend can then forward the information to the server.
The backend checks the incoming data and stores the readings with details such as the device identifier and timestamp. The dashboard can retrieve these records and display them to the user.
This is the workflow we want to establish between the physical modules and the software platform. Each stage needs to be tested so that missing readings or connection problems do not lead to incorrect information on the dashboard.
Using AWS for the cloud backend
As we work on connecting the hardware to the dashboard, AWS is one option we are considering for the cloud infrastructure.
AWS IoT Core could be used to receive messages from compatible connected devices or gateways. AWS Lambda could handle tasks such as validating incoming readings and processing data before storage.
Amazon DynamoDB could store structured sensor records, while Amazon S3 could hold exported datasets and reports. Amazon API Gateway could provide endpoints for the dashboard to communicate with the backend.
Amazon CloudWatch could help us inspect logs and identify processing errors when something goes wrong.
These services would give us a possible route towards a cloud connected version of TUFAN. The final architecture will depend on how the hardware communicates, how frequently readings arrive and how much data the system needs to retain.
We have to account for one important detail. A LoRa module does not automatically connect to AWS. The outdoor setup would need a suitable gateway or another network connection to forward its readings to the cloud.
The challenges we are working through
One of the main challenges is making sure that the readings are reliable. Sensors need proper placement and calibration, and measurements collected indoors may need to be handled differently from those collected outdoors.
Connectivity is another concern. If a device loses its connection, the system needs a sensible way to handle delayed or missing data. The dashboard should not present an old reading as if it were current.
We also need to consider device security and access permissions. This becomes particularly important if the platform is used by an institution or a government department.
These details may not be the most visible part of a demonstration, but they matter when turning a working prototype into something that people can depend on.
What comes next
Our next steps are to improve the connection between the monitoring modules and the dashboard, test the data collection process and build a dependable backend for storing and retrieving readings.
We also want to improve how the dashboard presents historical data so that users can compare observations from different monitoring locations.
Once the system has been tested under realistic conditions, we can evaluate its performance and identify what needs to change before a wider deployment.
Conclusion
TUFAN is our attempt to connect environmental sensing hardware with a dashboard that makes the collected information easier to use.
The indoor module is intended for monitoring enclosed spaces, while the outdoor module is designed to collect particulate matter readings in open environments. The government dashboard brings the software side of the project together and provides a place to review the available data.
We still have engineering work to complete, particularly around data reliability, communication and cloud integration. Working through these details will help us understand what it takes to move from a sensor prototype to a useful monitoring system.
For our team, the next milestone is simple. We want to get the modules collecting dependable readings and make sure those readings can be viewed correctly through the dashboard.
That is the foundation we want to build with TUFAN.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article