Indoor localization is a very important theme in the industry 4.0 era. This thesis presents some of the main problems and solutions that regards this topic, by also analyzing hybrid approaches combining different techniques. Besides the generic overview of the topic and the presentation of the most commons techniques and the current literature review, in this thesis there also results and methodologies of a practical work with the goal of analyzing the utilization of Ultra-wideband (UWB) and WiFi wireless protocols, together with localization techniques Received signal strength (RSS) and two-way ranging (TWR), which is based on Time of flight (TOF). The practical work is divided in two main parts: the first one consists in setting up a system that is able to collect data by exploiting both technologies and using it in some experiments that reproduce both Line-of-sight (LOS) and Non-line-of-sight (NLOS) conditions; the second one is about the analysis of the data that had been collected, in which Received signal strength indication (RSSI) WiFi is predicted through the data relative to UWB in order to study the relation between the two protocols, RSSI values obtained with both UWB and WiFi are compared in a visual way and localization is performed by using information of both protocols. The devices that were chosen in order to analyze the performances of UWB and WiFi protocols are respectively the DW1000 Decawave chip and some simple Raspberry Pi 3 model B. The results obtained at the end of this work show that by combining the data that were collected during the measurements process with the coordinates estimated by the DecaWave system with Machine Learning techniques it is possible to improve the precision of the Decawave system itself, up to double it in some scenarios.
La localizzazione in ambienti interni è una tematica molto importante nell'era dell'industria 4.0. Questa tesi presenta alcuni dei principali problemi e soluzioni che caratterizzano questo argomento, analizzandone anche approcci ibridi che combinano più tecniche tra di loro. La panoramica generale di questa tematica, la presentazione delle tecniche di uso più comune e l'analisi dello stato attuale della letteratura sono accompagnate da un lavoro pratico che ha l'obiettivo di analizzare l'utilizzo dei protocolli wireless Ultra-wideband (UWB) e WiFi, affiancati alle tecniche di localizzazione Received signal strength (RSS) e two-way ranging (TWR), la quale è basata su Time of flight (TOF). Il lavoro pratico è diviso in due parti principali: la prima consiste nella messa in piedi di un sistema che possa raccogliere dati sfruttando entrambe le tecnologie e nell'impiego di quest'ultimo in esperimenti che riproducono sia condizioni di linea di vista che non; la seconda consiste nell'analisi dei dati raccolti, in cui viene predetto il Received signal strength indication (RSSI) del WiFi mediante i dati relativi ad UWB in modo da studiare la relazione tra i due protocolli, vengono confrontati i valori di RSSI ottenuti sia con UWB che con WiFi in maniera visiva e viene effettuata la localizzazione usando le informazioni di entrambi i protocolli. I dispositivi scelti per analizzare le prestazioni dei protocolli UWB e WiFi sono rispettivamente il chip DW1000 della DecaWave e dei semplici Raspberry Pi 3 model B. I risultati ottenuti alla fine di questo lavoro mostrano che combinando i dati raccolti durante il processo di misurazioni con le coordinate stimate dal sistema DecaWave mediante techiche di Machine Learning è possibile migliorare la precisione del sistema DecaWave stesso, fino anche a raddoppiarla in alcuni scenari.
Realization and performance evaluation of a hybrid UWB/WiFi indoor localization system
IENI, MARCO
2017/2018
Abstract
Indoor localization is a very important theme in the industry 4.0 era. This thesis presents some of the main problems and solutions that regards this topic, by also analyzing hybrid approaches combining different techniques. Besides the generic overview of the topic and the presentation of the most commons techniques and the current literature review, in this thesis there also results and methodologies of a practical work with the goal of analyzing the utilization of Ultra-wideband (UWB) and WiFi wireless protocols, together with localization techniques Received signal strength (RSS) and two-way ranging (TWR), which is based on Time of flight (TOF). The practical work is divided in two main parts: the first one consists in setting up a system that is able to collect data by exploiting both technologies and using it in some experiments that reproduce both Line-of-sight (LOS) and Non-line-of-sight (NLOS) conditions; the second one is about the analysis of the data that had been collected, in which Received signal strength indication (RSSI) WiFi is predicted through the data relative to UWB in order to study the relation between the two protocols, RSSI values obtained with both UWB and WiFi are compared in a visual way and localization is performed by using information of both protocols. The devices that were chosen in order to analyze the performances of UWB and WiFi protocols are respectively the DW1000 Decawave chip and some simple Raspberry Pi 3 model B. The results obtained at the end of this work show that by combining the data that were collected during the measurements process with the coordinates estimated by the DecaWave system with Machine Learning techniques it is possible to improve the precision of the Decawave system itself, up to double it in some scenarios.| File | Dimensione | Formato | |
|---|---|---|---|
|
thesis.pdf
accessibile in internet per tutti
Descrizione: Thesis1
Dimensione
1.28 MB
Formato
Adobe PDF
|
1.28 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/142318