The rapid urbanization process in the last century has deeply changed the way we live and interact with each other. As most people now live in urban areas, cities are experiencing growing demands for more efficient and sustainable public services that may improve the perceived quality of life, specially with the anticipated impacts of climatic changes. In this already complex scenario with increasingly overcrowded urban areas, different types of emergency situations may happen anywhere and anytime, with unpredictable costs in human lives and economic losses. In order to cope with unexpected and potentially dangerous emergencies, smart cities initiatives have been developed in different cities, addressing multiple aspects of emergencies detection, alerting, and mitigation. In this thesis, three main issues relating to environmental and health risk management in smart cities were addressed. The first topic focuses on the possibility of using drone-femtocell systems to locate devices under rubble in post-earthquake scenarios by devising location algorithms with very low error and very high drone energy efficiency. The second topic addresses the detection and classification of rainfall levels using different types of signals (audio, video, radio) and deep learning techniques. Classifying rainfall intensity precisely and in real-time would mean providing smart cities with a system that predicts and manages hydrogeological risk conditions (landslides, floods, and inundations) in cities. And thus, it provides information in terms of road safety and much more, such as aspects related to the management of mobile radio connections in order to maintain good radio signal quality during heavy rainfall situations. The third topic concerns healthcare management within smart cities, in particular, the possibility of defining innovative algorithms for the detection and classification (in very short times and with high precision) of heart disease using ECG and PCG signals in conjunction with deep learning techniques.
Il rapido processo di urbanizzazione dell'ultimo secolo ha cambiato profondamente il modo in cui viviamo e interagiamo gli uni con gli altri. Poiché oggi la maggior parte delle persone vive in aree urbane, le città stanno sperimentando una crescente richiesta di servizi pubblici più efficienti e sostenibili che possano migliorare la qualità della vita percepita, specialmente con gli impatti previsti dei cambiamenti climatici. In questo scenario già complesso, con aree urbane sempre più sovraffollate, diversi tipi di situazioni di emergenza possono verificarsi ovunque e in qualsiasi momento, con costi imprevedibili in termini di vite umane e perdite economiche. Per far fronte a emergenze inaspettate e potenzialmente pericolose, sono state sviluppate iniziative di smart city in diverse città, che affrontano molteplici aspetti di rilevamento, allerta e mitigazione delle emergenze. In questa tesi sono stati affrontati tre temi principali relativi alla gestione del rischio ambientale e sanitario nelle città intelligenti. Il primo tema si concentra sulla possibilità di utilizzare sistemi basati sull'uso congiunto di droni e femtocelle per localizzare dispositivi sotto le macerie in scenari post-terremoto, ideando algoritmi di localizzazione con un errore molto basso e un'efficienza energetica del drone molto elevata. Il secondo argomento riguarda il rilevamento e la classificazione dei livelli di pioggia utilizzando diversi tipi di segnali (audio, video, radio) e tecniche di deep learning. Classificare con precisione e in tempo reale l'intensità delle precipitazioni significherebbe dotare le smart city di un sistema in grado di prevedere e gestire le condizioni di rischio idrogeologico (frane, alluvioni e inondazioni) nelle città. Quindi, fornire informazioni in termini di sicurezza stradale e molto altro ancora, come, ad esempio, gli aspetti legati alla gestione delle connessioni radio mobili per mantenere una buona qualità del segnale radio in situazioni di forti precipitazioni. Il terzo argomento riguarda la gestione dell'assistenza sanitaria all'interno delle smart city, in particolare la possibilità di definire algoritmi innovativi per il rilevamento e la classificazione (in tempi brevissimi e con elevata precisione) di patologie cardiache utilizzando segnali ECG e PCG in combinazione con tecniche di deep learning.
Tecniche innovative di telemedicina e gestione del rischio ambientale nelle Smart Cities attraverso sensori IoT basati su DL
AVANZATO, ROBERTA
2022
Abstract
The rapid urbanization process in the last century has deeply changed the way we live and interact with each other. As most people now live in urban areas, cities are experiencing growing demands for more efficient and sustainable public services that may improve the perceived quality of life, specially with the anticipated impacts of climatic changes. In this already complex scenario with increasingly overcrowded urban areas, different types of emergency situations may happen anywhere and anytime, with unpredictable costs in human lives and economic losses. In order to cope with unexpected and potentially dangerous emergencies, smart cities initiatives have been developed in different cities, addressing multiple aspects of emergencies detection, alerting, and mitigation. In this thesis, three main issues relating to environmental and health risk management in smart cities were addressed. The first topic focuses on the possibility of using drone-femtocell systems to locate devices under rubble in post-earthquake scenarios by devising location algorithms with very low error and very high drone energy efficiency. The second topic addresses the detection and classification of rainfall levels using different types of signals (audio, video, radio) and deep learning techniques. Classifying rainfall intensity precisely and in real-time would mean providing smart cities with a system that predicts and manages hydrogeological risk conditions (landslides, floods, and inundations) in cities. And thus, it provides information in terms of road safety and much more, such as aspects related to the management of mobile radio connections in order to maintain good radio signal quality during heavy rainfall situations. The third topic concerns healthcare management within smart cities, in particular, the possibility of defining innovative algorithms for the detection and classification (in very short times and with high precision) of heart disease using ECG and PCG signals in conjunction with deep learning techniques.File | Dimensione | Formato | |
---|---|---|---|
PhD_thesis_Avanzato_Roberta.pdf
accesso aperto
Dimensione
15.06 MB
Formato
Adobe PDF
|
15.06 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/72825
URN:NBN:IT:UNICT-72825