IoT/WI - Finding Dory - Final Project [A.Y. 2021/2022]

This is the final project regarding the courses Internet of Things and Wireless Internet, A.Y. 2021/2022.

💡 Introduction

The main purpose of the project is to find the coordinates $(X, Y)$ of the position of Dory by using a fingerprint dataset composed of Received Signal Strength Indicator (RSSI) values from six sniffer-anchors and the RSSIs emitted by Dory’s smartphone in its corresponding coordinates.

The fingerprint dataset has been divided into small fragments of data hidden inside MQTT publishes/subscriptions and CoAP request/response packets and it was required to obtain all these fragments, clean them and remove the outliers.

After reconstructing the entire fingerprint dataset, it was necessary to find a model that was able to determine the position of the device with the most similar fingerprint, comparing the RSSI measurements from Dory’s device with the database’s entries.

This model, implemented within a Python script, exploits the Euclidean Distance to find the final coordinates $(X,Y)$ of Dory.

📚 Contents

This repository contains the following files:

• input.txt (created manually after cleaning fragments), containing all fragments obtained via CoAP and MQTT and given as input to the Python script that computes odd positions and Dory’s estimate position
• output.txt (obtained with Python), containing the whole dataset (even and odd positions), which is the matrix used by the Python script to compute Dory’s position
• parser.py, the Python script file that we created to compute Dory's position
• Report file, explaining project development steps, the algorithm, and additional assumptions

👨🏼‍💻 Contributors

• Andrea Prisciantelli (@priscia99)
• Riccardo Reggiani (@riccardoreggiani)