Dissertations / Theses on the topic 'Monitoring Smart Environment'

This master thesis presents the design and implementation of a framework for monitoringdata related to activities of daily living (ADL) in a smart home environment, conducted for theHuman Health and Activity Laboratory (H2Al) at Luleå University of Technology. The generalaim of such environments is to increase the quality of life by enabling elderly to live longer athome while reducing the consumption of resources necessary. The complexity of collection,filtering and storing of data in smart home environments is however inherent due to oftenmany interworking sensor-systems, which allmay have different APIs and communicationpathways. This means that knowing whether ‘all systems are go’ when for example doing astudy is not easy, especially for persons not trained in data science.This work therefore aim to design and implement a framework for datamonitoring thattargets smart home environments in which activities of daily living are important for analysisof health-related conditions and for the personalised tailoring of interventions. The frameworkprimarily collects data from four selected systems, that for example track the position andmovements of a person. The data is stored in a database and visualised on a website toallow for monitoring of individual sensor data being collected. The framework was validatedtogether with a occupational therapist through a proof-of-concept trial in the Human Healthand Activity Laboratory, for which healthy subjects conducted a typical test (making a salad)used when assessing human performance.In conclusion, the developed framework works as expected, collecting data frommanysensor systems and storing the data in a common format, while the visualisation on a websiteis perceived as giving an easy overview of monitored data. Additional data can easily be addedto the framework and other processes beyond monitoring can be linked to the data, suchas further data refinement and algorithms for activity recognition (possibly using machinelearning techniques). Future work include to better distinguish data from multiple occupants,develop themanagement of synchronous and asynchronous data, and refine the web interfacefor additional simplicity

The study of the next-generation distributed systems for distributed monitoring and user localization in smart environment is treated in this thesis. In the last years, a growing amount of attention has been focused on the adoption of Wireless Sersor Networks (WSN) as a scalable and flexible backbone to implement innovative services in smart environments, like smart building and smart cities. In this framework, this thesis will describe heterogeneous solutions to improve the supervision, control, monitoring, and management of public and private spaces. All these systems exploit the wireless communication and sensing in combination with smart methodologies to provide advanced services to the end user in many application fields, from environmental monitoring to energy management in smart districts or private and public buildings, up to road security and indoor occupancy for management and security reason. The data acquired by the WSN technology are used as input of customized strategies and algorithms developed for the real-time processing, fast analysis and result visualization.

The study of the next-generation distributed systems for distributed monitoring and user localization in smart environment is treated in this thesis. In the last years, a growing amount of attention has been focused on the adoption of Wireless Sersor Networks (WSN) as a scalable and flexible backbone to implement innovative services in smart environments, like smart building and smart cities. In this framework, this thesis will describe heterogeneous solutions to improve the supervision, control, monitoring, and management of public and private spaces. All these systems exploit the wireless communication and sensing in combination with smart methodologies to provide advanced services to the end user in many application fields, from environmental monitoring to energy management in smart districts or private and public buildings, up to road security and indoor occupancy for management and security reason. The data acquired by the WSN technology are used as input of customized strategies and algorithms developed for the real-time processing, fast analysis and result visualization.

A home automation system usually contains a set of tools that users use to control devices in their homes, often remotely. These devices often include but are not limited to light switches, thermostats, thermometers, window blinds, and climate controls. The potential for these kinds of systems is huge because of the sheer number of devices that could be controlled and managed with minimal and inexpensive extra hardware. Many of the appliances in a normal home could benefit from being connected to a system that allows the owner to manage and control the devices in their home. Thus the number of potential devices is orders of magnitude larger than the number of homes connected to the system. There are several systems on the market that provide systems to monitor and control a home environment, however these systems only support specific in system devices. This uncovers a problem where a homeowner only has the opportunity to use specific products that fit into these systems. By introducing an open platform for the public that are not bound to any system we can allow more devices to be integrated in the home and contribute to further development of smarter homes. The goal with this project was to provide a scalable open platform with the possibility of asynchronous updating. This has been done by implementing multiple logical parts to both provide a web interface for the user and to allow us to handle communication and storage of data. All these parts are linked together to form a system of servers that handles all background operations. This thesis discusses and presents implementations of all of these servers, how they are implemented, communicate with each other, provide secure connections and how they can scale with increasing usage. In this process we also discuss and present techniques that were used, how to use them and their benefits, to help us reach our goal.
”Home automation” syftar till ett system som låter användaren kontrollera och styra olika apparater i hemmet, ofta sker detta utifrån. Dessa apparater inkluderar, men är inte begränsade till ljusbrytare, termostater, termometrar, persienner eller klimatanläggningar. Potentialen för ett sådant system är enormt då antalet apparater som skulle kunna övervakas med endast minimal och billig extra hårdvara är stort. Många av dessa apparater kan dra nytta av att vara ansluten till ett system som gör det möjligt för ägaren att hantera och styra enheter i deras hem. Antalet apparater är därför mångdubbelt fler än antalet hem som är kopplade till systemet. Det finns flera system på marknaden som ger användaren ett sätt att övervaka och styra en hemmiljö, men dessa system är ofta låsta och stödjer bara specifika enheter. Genom att införa en öppen plattform för allmänheten som inte är bunden till något system, kan vi tillåta att fler enheter kan integreras i hemmet och bidra till ytterligare utveckling av smartare hem. Målet med detta projekt var att skapa en skalbar öppen plattform med möjlighet till asynkron uppdatering. Detta har gjorts genom att implementera flera logiska delar för att förse användaren med ett webbgränssnitt och för att tillåta oss hantera kommunikation och lagring av data. Alla dessa delar är sammanlänkade för att bilda ett system av servrar som hanterar alla bakgrundsprocesser. Denna avhandling diskuterar och presenterar implementeringar av alla dessa servrar, hur de genomförs, kommunicera med varandra, ger säkra anslutningar och hur de kan skala med ökad användning. I denna process diskuterar och presenterar vi de tekniker som använts, hur man använder dem och deras fördelar.

Today semantic web technologies and Linked Data principles are providing formalism, standards, shared data semantics and data integration for unstructured data over the web. The result is a transformation from theWeb of Interaction to theWeb of Data and actionable information. On the crossroad lies our daily lives, containing plethora of unstructured data which is originating from low cost sensors and appliances to every computational element used in our modern lives, including computers, interactive watches, mobile phones, GPS devices etc. These facts accentuate an opportunity for system designers to combine these islands of data into a large actionable information space which can be utilized by automated and intelligent agents. As a result, this phenomenon is likely to institute a space that is smart enough to provide humans with comfort of living and to build an efficient society. Thus, in this context, the focus of my research has been to propose solutions to the problems in the domains of smart environment and energy management, under the umbrella of ambient intelligence. The potential role of semantic web technologies in these proposed solutions has been analysed and architectures for these solutions were designed, implemented and tested.

La dernière décennie a montré un intérêt croissant pour les bâtiments intelligents. Les bâtiments traditionnels sont les principaux consommateurs d’une partie importante des ressources énergétiques, d'où le besoin de bâtiments intelligents a alors émergé. Ces nouveaux bâtiments doivent être conçus selon des normes de construction durables pour consommer moins. Ces bâtiments intelligents sont devenus l’un des principaux domaines d’application des environnements pervasifs. En effet, une infrastructure basique de construction de bâtiment intelligent se compose notamment d’un ensemble de capteurs sans fil. Les capteurs basiques permettent l’acquisition, la transmission et la réception de données. La consommation d’énergie élevée de l’ensemble de ces appareils est un des problèmes les plus difficiles et fait donc l’objet d’études dans ce domaine de la recherche. Les capteurs sont autonomes en termes d’énergie. Etant donné que la consommation d’énergie a un fort impact sur la durée de vie du service, il existe plusieurs approches dans la littérature. Cependant, les approches existantes sont souvent adaptées à une seule application de surveillance et reposent sur des configurations statiques pour les capteurs. Dans cette thèse, nous contribuons à la définition d’une architecture de surveillance déclaratif durable par l’optimisation énergétique des interactions entre requêtes applicative orientées service et réseau de capteurs sans fil. Nous avons choisi le bâtiment intelligent comme cas d’application et nous étudions donc un système de surveillance d’un bâtiment intelligent. Du point de vue logiciel, un système de surveillance peut être défini comme un ensemble d’applications qui exploitent les mesures des capteurs en temps réel. Ces applications sont exprimées dans un langage déclaratif sous la forme de requêtes continues sur les flux de données des capteurs. Par conséquent, un système de multi-applications nécessite la gestion de plusieurs demandes de flux de données suivant différentes fréquences d’acq/tx de données pour le même capteur sans fil, avec des exigences dynamiques requises par les applications. Comme une configuration statique ne peut pas optimiser la consommation d’énergie du système, nous proposons une approche intitulée Smart-Service Stream-oriented Sensor Management (3SoSM) afin d’optimiser les interactions entre les exigences des applications et l’environnement des capteurs sans fil, en temps réel. 3SoSM offre une configuration dynamique des capteurs pour réduire la consommation d’énergie tout en satisfaisant les exigences des applications en temps réel. Nous avons conduit un ensemble d’expérimentations effectuées avec un simulateur de réseau de capteurs sans fil qui ont permis de valider notre approche quant à l’optimisation de la consommation d’énergie des capteurs, et donc l’augmentation de la durée de vie de ces capteurs, en réduisant notamment les communications non nécessaires
Recent researches and analysis reports declare that high energy consumption of buildings is major problem in developed countries. As a result, they show concretely that building energy management systems (BEMS) and deployed wireless sensor network environments are important for energy efficiency of building operations. In the literature, existing smart building management systems focus on energy consumption of the building, hardware deployed inside/outside of the building and network communication issues. They adopt static configurations for wireless sensor devices and proposed models are fitted to a single application. In this study, we propose a sustainable declarative monitoring architecture that focus on the energy optimisation of interactions between application service oriented queries and wireless sensor devices. We consider the monitoring system as a set of applications that exploit sensor measures in real time such as HVAC automation and control systems, real time supervision, security. These applications can be configured dynamically by the users or by the supervisor. In our approach, we take a data point of view: applications are declaratively expressed as a set of continuous queries on the sensor data stream. To achieve our objective of energy aware optimization of the monitoring architecture, we formalize sensor device configuration and fit data acquisition and data transmission to actual applications requirements. We present a complete monitoring architecture and an algorithm that handles dynamic sensor configuration. We introduce a platform that covers physical and also simulated wireless sensor devices

Research Project focused on developing innovative devices using the low-cost sensors to obtain the concentrations of greenhouse gases (GHGs) such as carbon dioxide (CO2) and methane (CH4) as well as obtain a good water quality as a 2nd treatment in the Wastewater Treatment Plant. In addition to sensor calibration, the multi-parameter monitor prototype were tested in several contexts: a) Laboratory scale with natural soil columns, to figure out the sensor response under controlled conditions, calibration and validation; b) Field scale in many geological contexts, for Air-Soil quality (methane and carbon dioxide measurements): Natural Gas Storage Site in Minerbio, Italy; Drilling and Hydraulic Fracturing activities in Greeley, CO, USA; for Water Quality: Wastewater Treatment Plant in Algarve, Portugal. The monitoring system provided a huge set of data for which can be used statistical analysis, management and processing (Big DATA). The source identification of greenhouse gas emissions is identified in several IPCC reports that climate change is the major emergency for the socio / economic / environmental equilibrium of Earth planet. No outliers were identified as methane gas concentrations at Minerbio gas storage site, Italy and at Hydraulic activities in Greeley, Colorado. The soil column experiments for infiltration basins in the Wastewater treatment plant in Algarve, Portugal, gave us good results, the water quality was improved after the 2nd treatment. The low-cost sensors (gas – water) gave as a good calibration and validation with r2 coefficient of correlation of 0.70 – 0.96.
Il progetto di ricerca si è concentrato sullo sviluppo di dispositivi innovativi utilizzando i sensori a basso costo per ottenere le concentrazioni di gas (GHG) quali anidride carbonica (CO2) e metano (CH4) e ottenere una buona qualità dell'acqua come secondo trattamento nelle acque reflue nell’impianto di trattamento. Oltre alla calibrazione del sensore, il prototipo di monitoraggio multiparametro è stato testato in diversi contesti: a) Nel laboratorio con colonne di terreno naturali, suoli, per determinare la risposta del sensore in condizioni controllate, calibrazione e validazione; b) Scala di campo in molti contesti geologici, per la qualità Aria-suolo (misure di metano e anidride carbonica, radon) nel sito di stoccaggio di gas naturale a Minerbio, Italia; Attività di perforazione e fratturazione idraulica in Greeley, Colorado, USA; per la qualità dell'acqua: impianto di trattamento delle acque reflue in Algarve, Portogallo. Il sistema di monitoraggio ha fornito un enorme set di dati per i quali è stato possibile utilizzare analisi statistiche, gestione ed elaborazione (Big DATA). L'identificazione della fonte delle emissioni di gas è stata identificata in diversi rapporti dell'IPCC secondo cui i cambiamenti climatici rappresentano l'emergenza principale per l'equilibrio socio / economico / ambientale del pianeta Terra. Non sono stati identificati valori anomali come concentrazioni di gas metano nel sito di stoccaggio di Minerbio (Italia) e nelle attività di perforazione in Greeley, Colorado, USA. Gli esperimenti con la colonna di terreno per i bacini di infiltrazione nell'impianto di trattamento delle acque reflue in Algarve, in Portogallo, ci hanno dato buoni risultati, la qualità dell'acqua è stata migliorata dopo il 2 ° trattamento. I sensori a basso costo (gas - acqua) per la qualita’ dell’aria e del suolo, hanno fornito una buona calibrazione e validazione con coefficiente di correlazione r2 di 0,70 - 0,96.

Internet of Things (IoT) and Machine Type Communication (MTC) have got more momentum in the last few years but still, need to be penetrated with their full swing in our daily life. This can be possible with general framework that provides global network coverage. Non-terrestrial networks comprised of satellites and aerial platforms are expected to provide next-generation communication services in underserved and un-served areas by ensuring the quality of service that cannot be covered by existing terrestrial networks owing to economical and geographical limitations. The aim of this thesis is to formulate a set of massive and critical MTC use cases such as global environment monitoring, tracking of shipping containers and smart agriculture, and assess their comprehensive requirements like data size, sensor node density and uplink capacity and discuss possible network architectures and deployments focusing on satellite or aerial networks. A rigorous discussion on different network architectures to address the requirements have been presented, that involve (1) Low Earth Orbit (LEO) satellite based network, (2) High Altitude Platform (HAP) based network, and (3) HAP and UAV based network. The proposed network architectures have been simulated and analyzed using MATLAB tools for respective use cases in terms of required number of satellites or aerial platforms. The criteria for selection of network architectures for the use cases are based on the minimum number of satellites or aerial platforms. The results show that LEO constellation consisting of 260 satellites are feasible concerning deployment and management for global environment monitoring network. Similarly, 1440 LEO satellites provide global coverage for tracking of shipping containers. Smart agriculture use case requires high throughput, and hence HAP and UAV integrated network architecture is more realistic for a fully autonomous system as compared to other network architectures. Cooperative control and management of set of agricultural machines can be performed at the UAV. Simulation results show that single UAV can be capable of commanding and controlling the agricultural smart machines in one square kilometer crop field and can send the summary of events to the central station via a HAP.
Internet of Things (IoT) och maskintypkommunikation (MTC) har fått mer fart under de senaste åren men måste fortfarande penetreras med sin fulla sväng i vårt dagliga liv. Detta kan vara möjligt med allmän ramverk som ger global nätverkstäckning. Icke- markbundna nät bestående av satelliter och flygplattformar förväntas tillhandahålla nästa generations kommunikationstjänster i undervärdiga och obetjänade områden genom att säkerställa kvaliteten påtjänster som inte kan täckas av befintliga marknät pågrund av ekonomiska och geografiska begränsningar. Syftet med den här avhandlingen är att formulera en uppsättning massiva och kritiska MTC-användningsfall som global miljöövervakning, spårning av fraktcontainrar och smart jordbruk, och utvärdera deras omfattande krav som datastorlek, sensornoddensitet och upplänkkapacitet och diskutera möjliga nätverk arkitekturer och distributioner med fokus påsatellit- eller flygnät. En rigorös diskussion om olika nätverksarkitekturer för att möta kraven har presenterats, som involverar (1) Low Earth Orbit (LEO) satellitbaserat nätverk, (2) High Altitude Platform (HAP) baserat nätverk, och (3) HAP och UAV baserat nätverk. De föreslagna nätverksarkitekturerna har simulerats och analyserats med MATLAB-verktyg för respek- tive användningsfall i termer av det nödvändiga antalet satelliter eller flygplattformar. Kriterierna för val av nätverksarkitekturer för användningsfallen är baserade pådet minsta antalet satelliter eller flygplattformar. Resultaten visar att LEO-konstellationen bestående av 260 satelliter är möjlig när det gäller distribution och hantering för globalt miljöövervakningsnätverk. Påliknande sätt ger 1440 LEO-satelliter global täckning för spårning av fraktcontainrar. Småjordbruksanvändningsfall kräver hög kapacitet, och följaktligen är HAP och UAV integrerad nätverksarkitektur mer realistisk för ett helt autonomt system jämfört med andra nätverksarkitekturer. Kooperativ kontroll och hantering av jordbruksmaskiner kan utföras vid UAV. Simuleringsresultat visar att en enda UAV kan vara kapabel att kommandera och kontrollera jordbrukssmarta maskiner i ett kvadratkilometer grödningsfält och kan skicka sammanfattningen av händelser till centralstationen via HAP.

This thesis contributes to the field of acoustic environmental monitoring by developing novel semiautomated methods of processing long audio recordings to conduct species richness surveys efficiently. These methods allow a machine to select rich subset of the recordings though estimations of acoustic variety, which can then be presented to the human listener for species identifications. This work represents a step towards more effective biodiversity monitoring of vocal species that can be performed at a larger scale than is possible with traditional methods.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 49).
Recent architectural trends have included exploring open space and the extensive use of glass as building material. While the details of these large, light-exposed, open-air environments can be modeled as thermal fluid systems in CFD simulations, the use of dense sensor networks can provide real-time monitoring of a building's airflow and thermal management systems without the need for computationally-intensive theoretical models, and can use this data to inform and advance these models. Sensor networks can provide an accurate picture of the actual conditions of a building and how those conditions can change over time, due to deterioration or external influences. The information gathered from such networks will be critical in determining the energy efficiency of a building. To do this, a sensor network made of two types of sensors, temperature-humidity and airflow, was deployed in the large, glass-enclosed atrium of the recently-completed MIT Media Lab Extension (E14) in late March 2010. Their performance was calibrated, monitored, and the preliminary results analyzed in conjunction with the external weather conditions in the Boston metropolitan area. The results show that while the use of the sensors in monitoring temperature and humidity is successful, the airflow sensors currently require a different solution to solve both the need for low-power consumption and resolution, range, and stability in its measurements.
by Ashley K. Turza.
S.B.

If the UK is to address its energy reduction targets, it is vital to understand energy use behaviours and to devise technology that positively encourages domestic occupants to use less energy. This study is cross-over research that spans energy research, social science and socio-technology. The work presented in this dissertation reveals the domestic energy saving potential of the use of In-hone Displays (IHDs) by quantifying changes in actual energy consumption and then evaluating these changes using social science research techniques to document the psychological nature of the human interaction with a digital user interface (UI). Many studies have investigated how IHDs for domestic electricity use change behaviour; the findings of this unique 37 month pre-normative study, the first of its kind in the UK, show that the coloured dual-fuel IHD had a positive effect on consumption behaviour and energy reduction. However, the exact difference in energy consumption between experimental groups is dependent on the type of normalisation condition applied to the recorded energy consumption. After the first six months of monitoring, those with a coloured IHD reduced their gas consumption by an average of 20% compared to a control group; this was tested to be statistically significant (p < .05). This difference in consumption was similar for those living in flats and those living in houses. The quantitative figures are reinforced by the findings from questionnaire and the semi-structured interviews, which show that those with an IHD were significantly more likely to reduce their gas consumption and reported increased use of the controls and settings like thermostats for heat-related appliances. Thirty-one months later, this change in gas use behaviour persisted. Over the total 37 month monitoring period, the majority of participants continued to engage with the IHD on a daily basis and consumed 27% less gas than the control group. This difference reached statistical significance (p=.05). The questionnaires conducted 31 months after the initial findings found that those in the intervention group had statistically higher gas reducing behaviour change scores (p < .05). The first six months of energy data show that the sample group with the IHD used 7% less electricity than the control group. The difference in group means was found to not be statistically significant (p > .05). The difference in electricity consumption was considerably higher in the sample living in houses than in the sample living in flats. Qualitative feedback from the participants suggests that the use of the IHD had a slight positive effect on users' consciousness of reducing electricity consumption. However, a larger portion of the occupants with no IHD were similarly confident in ingrained methods of regulating and reducing their electricity consumption. Thirty-one months later, the difference in electricity consumption was substantially higher than was measured for the first six months. Over the total 37 month monitoring period, the intervention group consumed 21% less electricity than the control group. This was not statistically significant (p > .05), the interviews found that those with an IHD did not directly attribute their reduced use of electricity to the IHD. Rather, they maintained low levels of electricity use because it was an ingrained habit long before they used the IHD and for fire and safety reasons. Between the 6 month report and 31 month report, both experimental groups reduced the amount of electricity and gas they consumed. This was attributed to changes in weather patterns and occupants growing more accustomed to their new home. The properties with highest gas consumption reduced their consumption closer to that predicted by the Standard Assessment Procedure (SAP). The research found contrasting differences in how the two utilities where perceived and used. This was evident when the energy data was divided into groups based on occupancy. Larger savings in gas consumption was seen in the intervention group with lower occupancy: the intervention group consumed considerable more electricity than the control group in the lower occupancy dwellings, and consumed considerably less in the larger occupancy dwellings. Electricity was described as a luxury, used to maintain a certain quality of life. Those with younger dependents felt it necessary to provide them with as much electronic luxury as they could. Electricity was relatively freely accessed and used by all residents with little resistance if a justified reason was given for its use. However, space heating was perceived as a sacrificial commodity. Heat was described as being relatively easy to regulate with the use of blankets and extra clothing. Heating controls were perceived to be out of reach for many but one or two in the household. This tended to be in control of the person responsible for the majority of household tasks. The users of Ewgeco IHD commented more on the device's ability to promote new gas saving behaviour in order to reduce gas consumption. In contrast, the visual representation of real-time electricity consumption was used more as a safety feature, and appears to fail to produce significant electricity reduction. The participants used the electricity consumption information to reinforce their existing levels of electricity use awareness and it highlighted electrical appliances that had been left on to them. This was reported to be specifically useful at times when the occupants were retiring from the living spaces in the home. These findings demonstrate that a simple ‘push-information' style IHD may need to evolve further with greater smart home control functionality, internet capability and user interaction for this technology to be part of the low-carbon solution. However, it has also been demonstrated that, for particular household groups, IHDs can lead to longer term changes in energy consumption behaviour, specifically for heat.

Nowadays words like Smart City, Internet of Things, Environmental Awareness surround us with the growing interest of Computer Science and Engineering communities. Services supporting these paradigms are definitely based on large amounts of sensed data, which, once obtained and gathered, need to be analyzed in order to build maps, infer patterns, extract useful information. Everything is done in order to achieve a better quality of life. Traditional sensing techniques, like Wired or Wireless Sensor Network, need an intensive usage of distributed sensors to acquire real-world conditions. We propose SenSquare, a Crowdsensing approach based on smartphones and a central coordination server for time-and-space homogeneous data collecting. SenSquare relies on technologies such as CoAP lightweight protocol, Geofencing and the Military Grid Reference System.

The Internet of Things (IoT) is the paradigm that allows us to interact with the real world by means of networking-enabled devices and convert physical phenomena into valuable digital knowledge. Such a rapidly evolving field leveraged the explosion of a number of technologies, standards and platforms. Consequently, different IoT ecosystems behave as closed islands and do not interoperate with each other, thus the potential of the number of connected objects in the world is far from being totally unleashed. Typically, research efforts in tackling such challenge tend to propose a new IoT platforms or standards, however, such solutions find obstacles in keeping up the pace at which the field is evolving. Our work is different, in that it originates from the following observation: in use cases that depend on common phenomena such as Smart Cities or environmental monitoring a lot of useful data for applications is already in place somewhere or devices capable of collecting such data are already deployed. For such scenarios, we propose and study the use of Collective Awareness Paradigms (CAP), which offload data collection to a crowd of participants. We bring three main contributions: we study the feasibility of using Open Data coming from heterogeneous sources, focusing particularly on crowdsourced and user-contributed data that has the drawback of being incomplete and we then propose a State-of-the-Art algorith that automatically classifies raw crowdsourced sensor data; we design a data collection framework that uses Mobile Crowdsensing (MCS) and puts the participants and the stakeholders in a coordinated interaction together with a distributed data collection algorithm that prevents the users from collecting too much or too less data; (3) we design a Service Oriented Architecture that constitutes a unique interface to the raw data collected through CAPs through their aggregation into ad-hoc services, moreover, we provide a prototype implementation.

This research reviews and extends the impedance-based structural health monitoring technique in order to detect and identify structural damage on various complex structures. The basic principle behind this technique is to apply high frequency structural excitations (typically higher than 30 kHz) through the surface-bonded piezoelectric transducers, and measure the impedance of structures by monitoring the current and voltage applied to the transducers. Changes in impedance indicate changes in the structure, which in turn can indicate that damage has occurred. Several case studies, including a pipeline structure, a composite reinforced aluminum plate, a precision part (gear), a quarter-scale bridge section, and a steel pipe header, demonstrate how this technique can be used to detect damage in real-time. A method to process impedance measurements to prevent significant temperature and boundary condition changes registering as damage has been developed and implemented. Furthermore, the feasibility of using the technique for high temperature structures and for condition monitoring of critical facilities subjected to a severe natural disaster has been investigated. While the impedance-based structural health monitoring technique indicates qualitatively that damage has occurred, more information on the nature of damage is necessary for remote structures. In this research, two different damage identification schemes have been combined with the impedance method in order to quantitatively assess the state of structures. One is based on a wave propagation modeling, and the other is the use of artificial neural networks. A newly developed wave propagation model has been developed and combined with the impedance method in order to estimate the severity of damage. Numerical and experimental investigations on 1-dimensional structures were presented to illustrate the effectiveness of the combined approach. Furthermore, to avoid the complexity introduced by conventional computational methods in high frequency ranges, multiple sets of artificial neural networks were integrated with the impedance-based health monitoring technique. By incorporating neural network features, the technique is able to detect damage in its early stage and to determine the severity of damage without prior knowledge of the model of structures. The dissertation concludes with experimental examples, investigations on a quarter-scale steel bridge section and a space truss structure, in order to verify the performance of the proposed methodology.
Ph. D.

Kako se efekat globalnog zagrevanja odigrava širom planete, tako se svetska populacija suočava sa verovatno jednim od najvažnijih socijalnih i naučnih fenomena-promenom parametara životne sredine usled zagađenja. Preduzimanje bilo kakve akcije zahteva precizna i tačna merenja parametara životne sredine u više desetina hiljada tačaka, postavljenih širom sveta. Pošto je finansijski skupo, a i praktično nemoguće napraviti tako veliki broj mernih stanica koje bi premrežile celu planetu, očigledno je da se moraju pronaći neka alternativna rešenja. Napravljen je merni sistem i realizovane su merne metode za udaljeno merenje parametara životne sredine. Ovaj sistem može biti realizovan kao stacionarna ili kao pokretna merna stanica. Radna hipoteza se zasniva na korišćenju statističke analize izmernih podataka, gde se dolazi do pretpostavke i dokaza o mogućnosti smanjenja broja senzora na mernoj stanici, jer se praćenjem jedne veličine (koncentracija ugljen-monoksida) može doći do pretpostavljene vrednosti druge veličine (koncentracija azot-dioksida) u slučaju da potiču iz istog izvora. A korišćenjem predikcije, pomoću regresionog modela – interpolacije i ekstrapolacije pokazala se mogućnost smanjenja broja mernih stanica. Naime, korišćenjem interpolacionih krivih moguće je na teritoriji jednog grada prikazati estimacije koncentracija gasova na osnovu podataka sa pokretne merne stanice. Takođe, na osnovu matematičkog ARMA modela pokazana je estimacija koncentracije gasova na osnovu prethodnih merenja.
As the effects of global warming are spreading globally, the world population encounters one of the most important social and scientific phenomena- changing the parameters of the environment due to pollution. Any conducted action requires precise and accurate measuring of the environmental parameters at several dozens of thousands points deployed around the world. Since financially, as well as practically, it is impossible to create such a large number of measuring stations which would network all over the planet, it is obvious that some alternative solutions must be found. A new measuring system is developed and measuring methods for remote measurement of environmental parameters are implemented. This system can be implemented as a stationary or mobile measuring station. The working hypothesis is based on the use of statistical analysis of measurement data. It leads to the possibility of reducing the number of sensors at measure station, as based on the monitoring of one value-gas concentration (the concentration of carbon monoxide) can be estimated values of other gas (the concentration of nitrogen - dioxide) in the case that they originate from the same source. Using prediction and regression models - interpolation and extrapolation have shown the possibility to reduce the number of measuring stations. Specifically, in the territory of one observedcity, by using interpolation curves, the estimation of concentrations of gases based on data from the measuring system can be shown. Also, based on a mathematical model (ARMA) estimation of concentrations of gases based on previous measurements is shown.

Human body temperature is an important indicator of physical performance and condition in terms of comfort, heat or cold stress. The aim of this research was to develop Temperature Sensing Fabric (TSF) for continuous temperature measurement in healthcare applications. The study covers the development and manufacture of TSF by embedding fine metallic wire into the structure of textile material using a commercial computerised knitting machine. The operational principle of TSF is based on the inherent propensity of a metal wire to respond to changes in temperature with variation in its electrical resistance. Over 60 TSF samples were developed with combinations of different sensing elements, two inlay densities and highly textured polyester yarn as the base material. TSF samples were created using either bare or insulated wires with a range of diameters from 50 to 150 μm and metal wires of nickel, copper, tungsten, and nickel coated copper. In order to investigate the Temperature-Resistance (T-R) relationship of TSF samples for calibration purposes, a customised test rig was developed and monitoring software was created in the LabVIEW environment, to record the temperature and resistance signals simultaneously. TSF samples were tested in various thermal environments, under laboratory conditions and in practical wear trials, to analyse the relationship between the temperature and resistance of the sensing fabric and to develop base line specifications such as sensitivity, resistance ratio, precision, nominal resistance, and response time; the influence of external parameters such as humidity and strain were also monitored. The regression uncertainty was found to be less than in ±0.1°C; the repeatability uncertainty was found to be less than ±0.5°C; the manufacturing uncertainty in terms of nominal resistance was found to be ± 2% from its mean. The experimental T-R relationship of TSF was validated by modelling in the thermo-electrical domain in both steady and transient states. A maximum error of 0.2°C was found between the experimental and modelled T-R relationships. TSF samples made with bare wire sensing elements showed slight variations in their resistance during strain tests, however, samples made with insulated sensing elements did not demonstrate any detectable strain-dependent-resistance error. The overall thermal response of TSF was found to be affected by basal fabric thickness and mass; the effect of RH was not found to be significant. TSF samples with higher-resistance sensing elements performed better than lower-resistance types. Furthermore, TSF samples made using insulated wire were more straightforward to manufacture because of their increased tensile strength and exhibited better sensing performance than samples made with bare wire. In all the human body wear trials, under steady-state and dynamic conditions both sensors followed the same trends and exhibited similar movement artifacts. When layers of clothing were worn over the sensors, the difference between the response of the TSF and a high-precision reference temperature were reduced by the improved isothermal conditions near the measurement site.

Non-residential users contribute to a significant portion of the total water delivered by water supplying agencies. However, a very limited number of studies have attempted to investigate the water use behavior of non-residential users. With the emergence of newer “smart” meters, water use now can be measured and recorded at a very high temporal frequency. Smart meters can help determine total water use, timing, and component end uses to better understand water use practices by non-residential users. Water end use disaggregation is the process of separating the water used by each fixture or process within a facility. This is useful because having a breakdown of the consumption of all end uses may encourage users to consume less water and gives them indications on how to do so. This project involved collecting and working with three different datasets with three different temporal scales (monthly billing data, 5-minute water use data, and 5-second water use data). We analyzed monthly billing data to solicit potential participating facilities for the study. For each participating facility, new smart devices were installed on their existing water meters, including an advanced water meter register and a pulse counting data logger. The newer registers logged and transmitted data to a web-accessible data portal at 5-minute intervals, while the pulse counters recorded water use at 5- second intervals. These devices enabled us to measure the timing and volume of different water uses (e.g., indoor versus outdoor versus industrial processes uses). In this project, we identified different water use events, average water used by each end use (from plumbing fixtures to industrial machinery), variability in end uses (faucets/toilets versus showers), variability in use by the type of user (manufacturing facilities versus assisted living homes), and the impact of the business type on the water use.

This thesis concerns the development of optimisation algorithms to determine the relative co-location, (localisation), of a number of freely-flying 'Smart Dust mote' sensor platform elements using a non-deterministic data-set derived from the duplex wireless transmissions between elements. Smart dust motes are miniaturised, microprocessor based, electronic sensor platforms, frequently used for a wide range of remote environmental monitoring applications; including specific climate synoptic observation research and more general meteorology. For the application proposed in this thesis a cluster of the notional smart dust motes are configured to imitate discrete 'Radio Drop Sonde' elements of the wireless enabled monitoring system in use by meteorological research organisations worldwide. This cluster is modelled in software in order to establish the relative positions during the 'flight' ; the normal mode of deployment for the Drop Sonde is by ejection from an aeroplane into an upper-air zone of interest, such as a storm cloud. Therefore the underlying research question is, how to track a number of these independent, duplex wireless linked, free-flying monitoring devices in 3-dimensions and time (to give the monitored data complete spatio-temporal validity). This represents a significant practical challenge, the solution applied in this thesis was to generate 3-dimensional geometries using the only 'real-time' data available; the Radio Signal Strength Indicator (RSSI) data is generated through the 'normal' duplex wireless communications between motes. Individual RSSI values can be considered as a 'representation of the distance magnitude' between wireless devices; when collated into a spatio-temporal data-set it 'encodes' the relative, co-locational, 3-dimensional geometry of all devices in the cluster. The reconstruction, (or decoding), of the 3-dimensional geometries encoded in the spatio-temporal data-set is a complex problem that is addressed through the application of various algorithms. These include, Random Search, and optimisation algorithms, such as the Stochastic Hill-climber, and various forms of Evolutionary Algorithm. It was found that the performance of the geometric reconstruction could be improved through identification of salient aspects of the modelled environment, the result was heuristic operators. In general these led to a decrease in the time taken to reach a convergent solution or a reduction in the number of candidate search space solutions that must be considered. The software model written for this thesis has been implemented to generalise the fundamental characteristics of an optimisation algorithm and to incorporate them into a generic software framework; this then provides the common code to all model algorithms used.

Two of the most important environmental challenges in the 21st century are to protect the quality of fresh water resources and to utilize renewable energy sources to lower greenhouse gas emissions. This study contributes to the solution of the first challenge by providing methodologies for optimal design of real-time water quality monitoring systems and interpretation of data supplied by the monitoring system to identify potential pollution sources in river networks. In this study, the optimal river water quality monitoring network design aspect of the overall monitoring program is addressed by a novel methodology for the analysis of this problem. In this analysis, the locations of sampling sites are determined such that the contaminant detection time is minimized for the river network while achieving maximum reliability for the monitoring system performance. The data collected from these monitoring stations can be used to identify contamination source locations. This study suggests a methodology that utilizes a classification routine which associates the observations on a contaminant spill with one or more of the candidate spill locations in the river network. This approach consists of a training step followed by a sequential elimination of the candidate spill locations which lead to the identification of potential spill locations. In order to contribute the solution of the second environmental challenge, this study suggests utilizing available excess energy in water distribution systems by providing a methodology for optimal design of energy recovery systems. The energy recovery in water distribution systems is possible by using micro hydroelectric turbines to harvest available excess energy inevitably produced to satisfy consumer demands and to maintain adequate pressures. In this study, an optimization approach for the design of energy recovery systems in water distribution networks is proposed. This methodology is based on finding the best locations for micro hydroelectric plants in the network to recover the excess energy. Due to the unsteady nature of flow in water distribution networks, the proposed methodology also determines optimum operation schedules for the micro turbines.

In the last 50 years, flooding has figured as the most frequent and widespread natural disaster globally. Extreme precipitation events stemming from climate change could alter the hydro-geological regime resulting in increased flood risk. Near real-time precipitation monitoring at local scale is essential for flood risk mitigation in urban and suburban areas, due to their high vulnerability. Presently, most of the rainfall data is obtained from ground‐based measurements or remote sensing that provide limited information in terms of temporal or spatial resolution. Other problems may be due to the high costs. Furthermore, rain gauges are unevenly spread and usually placed away from urban centers. In this context, a big potential is represented by the use of innovative techniques to develop low-cost monitoring systems. Despite the diversity of purposes, methods and epistemological fields, the literature on the visual effects of the rain supports the idea of camera-based rain sensors but tends to be device-specific. The present thesis aims to investigate the use of easily available photographing devices as rain detectors-gauges to develop a dense network of low-cost rainfall sensors to support the traditional methods with an expeditious solution embeddable into smart devices. As opposed to existing works, the study focuses on maximizing the number of image sources (like smartphones, general-purpose surveillance cameras, dashboard cameras, webcams, digital cameras, etc.). This encompasses cases where it is not possible to adjust the camera parameters or obtain shots in timelines or videos. Using a Deep Learning approach, the rainfall characterization can be achieved through the analysis of the perceptual aspects that determine whether and how a photograph represents a rainy condition. The first scenario of interest for the supervised learning was a binary classification; the binary output (presence or absence of rain) allows the detection of the presence of precipitation: the cameras act as rain detectors. Similarly, the second scenario of interest was a multi-class classification; the multi-class output described a range of quasi-instantaneous rainfall intensity: the cameras act as rain estimators. Using Transfer Learning with Convolutional Neural Networks, the developed models were compiled, trained, validated, and tested. The preparation of the classifiers included the preparation of a suitable dataset encompassing unconstrained verisimilar settings: open data, several data owned by National Research Institute for Earth Science and Disaster Prevention - NIED (dashboard cameras in Japan coupled with high precision multi-parameter radar data), and experimental activities conducted in the NIED Large Scale Rainfall Simulator. The outcomes were applied to a real-world scenario, with the experimentation through a pre-existent surveillance camera using 5G connectivity provided by Telecom Italia S.p.A. in the city of Matera (Italy). Analysis unfolded on several levels providing an overview of generic issues relating to the urban flood risk paradigm and specific territorial questions inherent with the case study. These include the context aspects, the important role of rainfall from driving the millennial urban evolution to determining present criticality, and components of a Web prototype for flood risk communication at local scale. The results and the model deployment raise the possibility that low‐cost technologies and local capacities can help to retrieve rainfall information for flood early warning systems based on the identification of a significant meteorological state. The binary model reached accuracy and F1 score values of 85.28% and 0.86 for the test, and 83.35% and 0.82 for the deployment. The multi-class model reached test average accuracy and macro-averaged F1 score values of 77.71% and 0.73 for the 6-way classifier, and 78.05% and 0.81 for the 5-class. The best performances were obtained in heavy rainfall and no-rain conditions, whereas the mispredictions are related to less severe precipitation. The proposed method has limited operational requirements, can be easily and quickly implemented in real use cases, exploiting pre-existent devices with a parsimonious use of economic and computational resources. The classification can be performed on single photographs taken in disparate conditions by commonly used acquisition devices, i.e. by static or moving cameras without adjusted parameters. This approach is especially useful in urban areas where measurement methods such as rain gauges encounter installation difficulties or operational limitations or in contexts where there is no availability of remote sensing data. The system does not suit scenes that are also misleading for human visual perception. The approximations inherent in the output are acknowledged. Additional data may be gathered to address gaps that are apparent and improve the accuracy of the precipitation intensity prediction. Future research might explore the integration with further experiments and crowdsourced data, to promote communication, participation, and dialogue among stakeholders and to increase public awareness, emergency response, and civic engagement through the smart community idea.
Negli ultimi 50 anni, le alluvioni si sono confermate come il disastro naturale più frequente e diffuso a livello globale. Tra gli impatti degli eventi meteorologici estremi, conseguenti ai cambiamenti climatici, rientrano le alterazioni del regime idrogeologico con conseguente incremento del rischio alluvionale. Il monitoraggio delle precipitazioni in tempo quasi reale su scala locale è essenziale per la mitigazione del rischio di alluvione in ambito urbano e periurbano, aree connotate da un'elevata vulnerabilità. Attualmente, la maggior parte dei dati sulle precipitazioni è ottenuta da misurazioni a terra o telerilevamento che forniscono informazioni limitate in termini di risoluzione temporale o spaziale. Ulteriori problemi possono derivare dagli elevati costi. Inoltre i pluviometri sono distribuiti in modo non uniforme e spesso posizionati piuttosto lontano dai centri urbani, comportando criticità e discontinuità nel monitoraggio. In questo contesto, un grande potenziale è rappresentato dall'utilizzo di tecniche innovative per sviluppare sistemi inediti di monitoraggio a basso costo. Nonostante la diversità di scopi, metodi e campi epistemologici, la letteratura sugli effetti visivi della pioggia supporta l'idea di sensori di pioggia basati su telecamera, ma tende ad essere specifica per dispositivo scelto. La presente tesi punta a indagare l'uso di dispositivi fotografici facilmente reperibili come rilevatori-misuratori di pioggia, per sviluppare una fitta rete di sensori a basso costo a supporto dei metodi tradizionali con una soluzione rapida incorporabile in dispositivi intelligenti. A differenza dei lavori esistenti, lo studio si concentra sulla massimizzazione del numero di fonti di immagini (smartphone, telecamere di sorveglianza generiche, telecamere da cruscotto, webcam, telecamere digitali, ecc.). Ciò comprende casi in cui non sia possibile regolare i parametri fotografici o ottenere scatti in timeline o video. Utilizzando un approccio di Deep Learning, la caratterizzazione delle precipitazioni può essere ottenuta attraverso l'analisi degli aspetti percettivi che determinano se e come una fotografia rappresenti una condizione di pioggia. Il primo scenario di interesse per l'apprendimento supervisionato è una classificazione binaria; l'output binario (presenza o assenza di pioggia) consente la rilevazione della presenza di precipitazione: gli apparecchi fotografici fungono da rivelatori di pioggia. Analogamente, il secondo scenario di interesse è una classificazione multi-classe; l'output multi-classe descrive un intervallo di intensità delle precipitazioni quasi istantanee: le fotocamere fungono da misuratori di pioggia. Utilizzando tecniche di Transfer Learning con reti neurali convoluzionali, i modelli sviluppati sono stati compilati, addestrati, convalidati e testati. La preparazione dei classificatori ha incluso la preparazione di un set di dati adeguato con impostazioni verosimili e non vincolate: dati aperti, diversi dati di proprietà del National Research Institute for Earth Science and Disaster Prevention - NIED (telecamere dashboard in Giappone accoppiate con dati radar multiparametrici ad alta precisione) e attività sperimentali condotte nel simulatore di pioggia su larga scala del NIED. I risultati sono stati applicati a uno scenario reale, con la sperimentazione attraverso una telecamera di sorveglianza preesistente che utilizza la connettività 5G fornita da Telecom Italia S.p.A. nella città di Matera (Italia). L'analisi si è svolta su più livelli, fornendo una panoramica sulle questioni relative al paradigma del rischio di alluvione in ambito urbano e questioni territoriali specifiche inerenti al caso di studio. Queste ultime includono diversi aspetti del contesto, l'importante ruolo delle piogge dal guidare l'evoluzione millenaria della morfologia urbana alla determinazione delle criticità attuali, oltre ad alcune componenti di un prototipo Web per la comunicazione del rischio alluvionale su scala locale. I risultati ottenuti e l'implementazione del modello corroborano la possibilità che le tecnologie a basso costo e le capacità locali possano aiutare a caratterizzare la forzante pluviometrica a supporto dei sistemi di allerta precoce basati sull'identificazione di uno stato meteorologico significativo. Il modello binario ha raggiunto un'accuratezza e un F1-score di 85,28% e 0,86 per il set di test e di 83,35% e 0,82 per l'implementazione nel caso di studio. Il modello multi-classe ha raggiunto un'accuratezza media e F1-score medio (macro-average) di 77,71% e 0,73 per il classificatore a 6 vie e 78,05% e 0,81 per quello a 5 classi. Le prestazioni migliori sono state ottenute nelle classi relative a forti precipitazioni e assenza di pioggia, mentre le previsioni errate sono legate a precipitazioni meno estreme. Il metodo proposto richiede requisiti operativi limitati, può essere implementato facilmente e rapidamente in casi d'uso reali, sfruttando dispositivi preesistenti con un uso parsimonioso di risorse economiche e computazionali. La classificazione può essere eseguita su singole fotografie scattate in condizioni disparate da dispositivi di acquisizione di uso comune, ovvero da telecamere statiche o in movimento senza regolazione dei parametri. Questo approccio potrebbe essere particolarmente utile nelle aree urbane in cui i metodi di misurazione come i pluviometri incontrano difficoltà di installazione o limitazioni operative o in contesti in cui non sono disponibili dati di telerilevamento o radar. Il sistema non si adatta a scene che sono fuorvianti anche per la percezione visiva umana. I limiti attuali risiedono nelle approssimazioni intrinseche negli output. Per colmare le lacune evidenti e migliorare l'accuratezza della previsione dell'intensità di precipitazione, sarebbe possibile un'ulteriore raccolta di dati. Sviluppi futuri potrebbero riguardare l'integrazione con ulteriori esperimenti in campo e dati da crowdsourcing, per promuovere comunicazione, partecipazione e dialogo aumentando la resilienza attraverso consapevolezza pubblica e impegno civico in una concezione di comunità smart.

In this work, IoT* (AmbiSense), we present our developed IoT system as a solution for Building and Energy Management using visualization tools to identify heuristics and create automatic savings. Our developed prototypes communicate using LoRa, one of the latest IoT technologies, and are composed of a set of battery-operated sensors tied to a System on Chip. These sensors acquire environmental data such as temperature, humidity, luminosity, air quality, and also motion. For small to medium-size buildings where system management is possible, a multiplatform dashboard provides visualization templates with real-time data, allowing to identify patterns and extract heuristics that lead to savings using a set of pre-defined actions or manual intervention. LoBEMS (LoRa Building and Energy Management System), was validated in a kindergarten school during a three-year period. As an outcome, the evaluation of the proposed platform resulted in a 20% energy saving and a major improvement of the environment quality and comfort inside the school. For larger buildings where system management is not possible, we created a 3D visualization tool, that presents the system collected data and warnings in an interactive model of the building. This scenario was validated at ISCTE-IUL University Campus, where it was necessary to introduce the community interaction to achieve savings. As a requested application case, our system was also validated at the University Data Center, where the system templates were used to detect anomalies and suggest changes. Our flexible system approach can easily be deployed to any building facility without requiring large investments or complex system deployments.
Nesta dissertação de mestrado, IoT * (AmbiSense), é apresentado um sistema IoT desenvolvido como uma solução para Gestão de Edifícios e Energia recorrendo a ferramentas de visualização para identificar heurísticas e criar poupanças automáticas. Os protótipos desenvolvidos comunicam utilizando LoRa, e são compostos por um conjunto de sensores ligados a um microcontrolador alimentado por bateria. Os sensores adquirem dados como temperatura, humidade, luminosidade, qualidade do ar e movimento. Para edifícios de pequena e média dimensão onde a gestão do sistema é possível, um dashboard fornece templates de visualização com dados em tempo real, permitindo extrair heurísticas, que introduzem poupanças através de um conjunto de ações predefinidas ou intervenção manual. O sistema LoBEMS (LoRa Building and Energy Management System), foi validado numa escola local durante um período de três anos. A avaliação do sistema resultou numa poupança de energia de 20% e uma melhoria significativa da qualidade do ambiente e conforto no interior da escola. Para edifícios de maior dimensão onde a gestão do sistema não é possível, criámos uma ferramenta de visualização 3D, que apresenta os dados e alertas do sistema, num modelo interativo do edifício. Este cenário foi validado no campus do ISCTE-IUL, onde foi necessária a interação da Comunidade para obter poupanças. Foi nos também solicitada uma validação do sistema no centro de dados da Universidade, onde os templates do sistema foram utilizados para detetar anomalias e sugerir alterações. A flexibilidade do sistema permite a sua implementação em qualquer edifício, sem exigir um grande investimento ou implementações complexas.

In recent years, the improvements of small-scale Unmanned Aerial Vehicles (UAVs) in terms of flight time, automatic control, and remote transmission are promoting the development of a wide range of practical applications. In aerial video surveillance, the monitoring of broad areas still has many challenges due to the achievement of different tasks in real-time, including mosaicking, change detection, and object detection. In this thesis work, a small-scale UAV based vision system to maintain regular surveillance over target areas is proposed. The system works in two modes. The first mode allows to monitor an area of interest by performing several flights. During the first flight, it creates an incremental geo-referenced mosaic of an area of interest and classifies all the known elements (e.g., persons) found on the ground by an improved Faster R-CNN architecture previously trained. In subsequent reconnaissance flights, the system searches for any changes (e.g., disappearance of persons) that may occur in the mosaic by a histogram equalization and RGB-Local Binary Pattern (RGB-LBP) based algorithm. If present, the mosaic is updated. The second mode, allows to perform a real-time classification by using, again, our improved Faster R-CNN model, useful for time-critical operations. Thanks to different design features, the system works in real-time and performs mosaicking and change detection tasks at low-altitude, thus allowing the classification even of small objects. The proposed system was tested by using the whole set of challenging video sequences contained in the UAV Mosaicking and Change Detection (UMCD) dataset and other public datasets. The evaluation of the system by well-known performance metrics has shown remarkable results in terms of mosaic creation and updating, as well as in terms of change detection and object detection.

碩士
朝陽科技大學
資訊與通訊系
105
The techniques of IoT(Internet of Thing) and the technology changed into more advanced in recent year. More and more products combine with the IoT techniques. Smart home connects with variety of daily necessities. By using the IoT, products can communicate with each other and making our daily life become safer and more convenient. The point of smart home is sensor, control and immediate inform. Our system is using sensor model and control model to sensing environment’s data and control home appliances. We can monitored by using the platform on web browser and inform alarm. The monitoring platform also provide history which is recorded data’s variation in recent three days. The user can observe environment data from the chart. Moreover, we also design the system in mobile application. The user can monitor data on the platform in anytime. Mobile application also provide inform function, if sensor get the value is overflow, mobile application will inform on user’s cellphone and mobile application will tell user which sensor is overflow. Our system also provide multi-model monitor in the same time. When the multi-model sent the data to same server, this server’s monitoring platform can show multi-model’s sensing data. We use temperature and humidity sensor, smoking sensor, ultrasonic sensor and ambient light sensor in this system. The user only need to put the sensor on the Arduino board which need to follow by our system’s specifications. You can get the correspond data on the monitor.

Project Work presented as the partial requirement for obtaining a Master's degree in Information Management, specialization in Knowledge Management and Business Intelligence
The need to answer ever-increasing urban challenges and the integration of new technologies in every aspect of our daily lives prompt the creation of the Smart City concept. These new technologies gather an enormous quantity of data that reveals interesting patterns about cities, enabling opportunities to enhance public decision-making and problem-solving. One critical aspect of urban life is its relationship with the environment. There’s a real need to work on a unified global approach on how to tackle issues such as pollution and waste management, but there is not a single framework or guideline to handle these new data gathered. This work focuses on structuring a framework of analysis for waste collection and transportation purposes. The objective was to develop monitoring dashboards for Departamento de Higienização Urbana (DMHU), using proven metrics of efficiency and effectiveness, result of a comprehensive literature review. A BI framework was developed using Power BI to perform efficiency-based analysis of waste collection circuits – the main process of DMHU. The receptivity of DMHU towards the solution presented seems to indicate that a BI solution is indeed valuable for complex monitoring problems such as solid waste management.

碩士
長庚大學
醫療機電工程研究所
103
Obstructive Sleep Apnea Syndrome (OSAS) is caused by partial obstruction or complete blockage of the upper air way during sleep. Overnight polysomnography (PSG) is golden standard device for the diagnosis of OSAS. It monitors and records several biophysiological signals throughout the night by attaching several sensors to patients. This procedure is complex and requires a specialized sleep center. The purpose of this work is to develop a sleep activity behavior monitoring platform and provide self-monitoring of abnormal events in home environment. Besides, there are no clear and consistent criteria to follow in calculating the baseline of airflow or oxygen saturation for apnea and hypopnea events. Therefore, we proposed a method to calculate the baseline using tidal volume reduction ratio and oxygen saturation reduction. Experimental results demonstrated the proposed method can correctly determine the apnea and hypopnea events. In addition, previous literatures show occurrence of that the upper air way blockage may be dependent on movement of the head position during sleep. In this research, a smart pillow integrated with physiological measurements, head posture biofeedback and close-loop pneumatic control mechanism was developed. The pillow has four airbags with two degrees of freedom. It can provide the adjustment of positions with tilt angle range (0° to 21°) and rotational angle (0° to 24°) with less than the 1°. This work also evaluated the identification of apnea and hypopnea events using proposed baseline calculation method for clinical applicability.

This dissertation presents a system for monitoring heart rate variability (HRV) by electrocardiogram (ECG) and photoplethysmography (PPG) for the detection of attention state in a learning environment. The ECG and PPG signs were acquired and processed through the development of a multi-microcontroller embedded system. When the waves of these signals are extracted, the heart rate is calculated through the time intervals between the R peaks for the electrocardiogram and between the beats for photopletismography. Finally, in order to indicate the level of attention on the part of the user, an input was added in which the volunteer marks the periods he or she considers to be most attentive. This data is associated with a millisecond resolution time stamp and sent in real time via Internet WiFi to a database in the cloud. After this data is stored, the HRV is analysed based on algorithms developed with the Matlab tool. These algorithms allow the study of cardiac variability according to time and frequency domains and how non-linear HRV measurements were also considered. Finally, a module for measuring skin conductivity was added, relating it to the analysis of the level of stress during the learning process. In order to prove the reliability of the system, several volunteers were tested in real environments according to a stipulated protocol. These records were analysed as a starting point to classify situations of greater attention of the volunteer in an educational scenario.
Esta dissertação apresenta um sistema de monitorização da variabilidade da frequência cardíaca (VFC) através da Eletrocardiograma (ECG) e Fotoplestimografria (PPG) para a deteção do estado de atenção num ambiente de aprendizagem. Os signais de ECG e PPG foram adquiridos e processados através do desenvolvimento de um sistema embebido multi-microcontrolador. Ao extrair-se as ondas destes sinais, calcula-se a frequência cardíaca através dos intervalos de tempo entre os picos R para o eletrocardiograma e entre os batimentos para a fotopletismografia. Por fim, para poder indicaro nível da atenção da parte do utente adicionou-se um input em que o voluntário marca os períodos que considera estar mais atento. Estes dados são associados a uma marca temporal com resolução ma base de dados na nuvém. Após o armazenamento destes dados, analisa-se a VFC com base em algoritmos desenvolvidos com a ferramenta Matlab. Estes algoritmos permitem estudar a variabilidade cardíaca segundo os domínios do tempo, da frequência e como tambem foram consideradas medidas VFC não lineares. Por fim, adicionou-se um módulo para a medição da condutividade da pele, relacionando-a com a avaliação do nível de stress durante o processo de aprendizagem. Para comprovar a fiabilidade do sistema realizaram-se testes a diversos voluntários em ambientes reais de acordo com um protocolo estipulado. Estes registos foram analisados como ponto de partida para classificar situações de maior atenção do voluntário perante um cenário educativo.

In this research work, we describe the development and subsequent validation of EnerMon a flexible, efficient, edge-computing based IoT LoRa System to monitor power consumption. This system provides real-time information and a descriptive analytics process to provide a ‘big picture’ about energy consumption over time and identify energetic waste. The solution is based on Arduinos, Current Transformer Sensors, Raspberry PI as an application server and LoRa communication alongside a description and information on what is to be expected of it. It describes the development process from the Design phase to the Validation phase with all the steps in between. Due to LoRa low debit communication, an edge computing approach was implemented to create a real-time monitoring process based on this technology. This solution, with the help of descriptive analysis, allows the creation of an energetic local footprint, using a low-cost developed solution for less than 80€ per three-phases monitoring device. It also allows for an easy installation without communication range and obstacles limitations making it easy to be used in a different set of situations from big complex building to smaller consumers, such as electric boilers, or simply to measure the energetic footprint of tourists in a small local tourist apartment.
O presente estudo, realizado no âmbito da tese de mestrado, descreve o desenvolvimento e subsequente validação de EnerMon, um sistema IoT(Internet of Things / Internet das Coisas) LoRa flexível, eficiente e baseado em edge-computing capaz de monitorizar consumo de energia em tempo real. Através de processos de análise descritiva é também apresentada uma visão geral sobre o consumo energético ao longo do tempo com a identificação de desperdícios de energia. Neste trabalho o leitor irá tomar conhecimento do processo de desenvolvimento completo do sistema, desde a fase de design à fase de testes, criado com o uso de Arduinos, Sensores de transformadores de corrente e Raspberry PIs como servidores aplicacionais, bem como informação relacionada com a comunicação LoRa e o que lhe é expectável. Esta solução, com a ajuda de analíticas descritivas, permitem a identificação de pegadas energéticas locais, usando materiais de baixo custo por menos de 80€ por sensor de monitorização de três fases. Permite também a fácil instalação sem limitações de alcance e obstáculos na comunicação, simplificando a sua utilização em diferentes ambientes desde edifícios complexos a consumidores energéticos mais pequenos, como caldeiras elétricas ou simplesmente medir a pegada energética de turistas em alojamento local.

Quella del cambiamento climatico è una delle più grandi sfide che l'umanità dovrà affrontare nei prossimi decenni. Per questo, negli ultimi anni, gli sforzi della ricerca si sono focalizzati sullo sviluppo ed implementazione di sistemi distribuiti di monitoraggio ambientale. Questi sistemi sono in grado di produrre grandi quantità di dati, che possono essere usati per descrivere i cambiamenti climatici e, sperabilmente, indirizzare le future decisioni politiche allo scopo di mitigarne gli effetti. Ad ogni modo, per rendere questi sistemi effettivamente intelligenti, è necessario tenere in considerazione diversi aspetti, inclusi i due su cui si è focalizzato questo lavoro di tesi. Il primo, spesso sottovalutato, riguarda la progettazione dell'esperimento di acquisizione dei dati: infatti, un setting sperimentale poco consono porta a dati in qualche modo affetti da un bias, e, di conseguenza, a risultati non significativi. Il secondo aspetto invece riguarda l'algoritmo usato per modellare i dati, che dovrebbe essere scelto per riflettere la natura degli stessi. Questo lavoro prova quindi a dare un (primo) contributo ad entrambi questi aspetti, descrivendo i risultati di due specifici scenari di utilizzo, e mostrando come gli esperimenti possano beneficiare da alcune semplici linee guida. L'obiettivo finale a cui tende questo lavoro è quindi quello di definire una pipeline di elaborazione dei dati ambientali, che possa, a lungo andare, diventare abbastanza flessibile da essere adattata a scenari eterogenei e relativi ad una varietà di fenomeni ambientali.
Climate change is one of the biggest challenges that humanity will face in the upcoming decades. Hence, over the last few years, the environmental engineering research community has focused its effort on the development and deployment of (often distributed) smart sensor systems, specifically designed for environmental monitoring. These sensors produce large amounts of data, which can be used to describe climate changes and, hopefully, suggest future actions to prevent further damages to the environment. However, to enable the ’smart’ capabilities in such systems, researchers must pay attention to several aspects, including two on which this thesis work is focused. The first one, which is often underestimated, is the design of the data acquisition phase: a poor experimental setting will lead to biased data, and therefore ineffective results. The second one concerns the algorithm used to model data, which should be chosen to reflect their intrinsic nature. This work tries to give a first contribution to both these aspect, describing the results of two specific use case scenarios, and highlighting how experiments can greatly benefit from some simple, yet effective, design guidelines. The final goal is to define an initial working pipeline for environmental data processing, which can be both flexible to be adapted to different scenarios, and accurate enough to give an effective description of the observed phenomena.

碩士
正修科技大學
電機工程研究所
106
This manuscript provides a smart monitoring system design of aquaculture, used for the control of aquaculture water quality to be the optimal state, by an IoT structure. This design integrates multiple sensors, such as temperature sensor, PH sensor, TDS sensor, EC sensor, ORP sensor as well as smart functions. The smart functions include communication, time scheduling, power monitoring, and safety protections. All these datas can upload into cloud server via Wifi communication. Not only it will alert when water is under the bad quality, also it can offer informations of water quality to users. Moreover, the system include solar energy to provide indipendent power supply.By comparing with the conventional controller, this design not only significantly reduces the required area of wiring, but also improves the setup efficiency.

碩士
國立成功大學
電機工程學系碩博士班
100
This thesis presents an integration of load management hardware module and thermal humidity sensor, where Zigbee wireless network is included to establish an energy management and environmental sensing system bases on the Internet-of-Things (IoT) structure. In this developed system platform, it is equipped with the functions of real-time monitoring, load control and data storage, in which the electricity management strategies consisting of electricity usage scheduling, environment regulation and demand response are all concerned. It is found that through this scheme design, the developed system comes with high flexibility along with the improved power consumption efficiency, achieving the goal energy conservation. In order to confirm the feasibility and practicality of the proposed system design, it has been simulated and tested under different scenarios. From the test results, they have demonstrated the effectiveness of this system on environment supervision and home appliance management applications, which is meanwhile served as the beneficial reference for electricity management operating model and future IoT development.

博士
國立交通大學
電機資訊國際學程
107
Advances of micro-electromechanical system (MEMS), sensing technology, and wireless communication have significantly encouraged the development of wireless sensor networks (WSNs) in the past decade. A WSN is widely used for habitat and environmental monitoring, medical application (with the purpose of improving quality of health care), agricultural assistance, and as solutions to military problems. For monitoring applications, sufficient sensing coverage is essential. In this dissertation, we propose a Coverage-Aware Sensor Automation (CASA, which means "home" in Spanish) protocol suite, including: a global sensor deployment scheme and sensing coverage recovery in the presence of sensor failures. An Enhanced Virtual Forces Algorithm with Boundary Forces (EVFA-B) is proposed for deployment scheme of omnidirectional mobile sensors to ensure sufficient sensing coverage. In the face of sensing node failures, a Sensor Self-Organizing Algorithm (SSOA) is devised to efficiently recover the sensing void and restore the required sensing coverage. Different from omnidirectional sensors, the coverage region of a directional sensor is determined by not only the sensing radius (distance), but also its sensing orientation and spread angle. Heterogeneous sensing distance and spread angles are likely to exist among directional sensors. In this dissertation, we deal with heterogeneous directional sensors equipped with locomotion and rotation facilities to enable sensors self-deployment. Two Enhanced Deployment Algorithms (EDA-I and EDA-II) are proposed to achieve high sensing coverage ratio in the monitored field. EDA-I leverages the concept of virtual forces (for sensors movements) and virtual boundary torques (for sensors rotations), whereas EDA-II combines Voronoi diagram directed movement and boundary torques guided rotations. Performance results demonstrate that our enhanced deployment mechanisms are capable of providing desirable surveillance level, while consuming moderate moving and rotating energy under reasonable computation time. Once the deployment algorithm produces moving destinations (goals) for all sensors, however, the problem of how to schedule moving paths to reach the goals without collisions remains largely neglected in the sensor networking field. Only with proper path planning and scheduling, all robots will be able to reach their intended destination successfully. We propose a Collision-Free Path Planning (CFPP) mechanism, based on geometric formulations and batched movements, to address the sensors deployment problem. Our proposed CFPP method incurs little computation latency, moderate energy consumption, and ensures 100$\%$ goal reachability. A proof-of-concept prototype is implemented to corroborate our protocol feasibility in a real-life environment.

碩士
國立宜蘭大學
環境工程學系碩士班
104
Disclosure of critical effects of various indoor air pollutants (IAPs) for human health risk is attributable to issue of the Sick Building Syndrome (SBS) widely discussed all over the world nowadays. Hereafter, in 2010 the World Health Organization (WHO) tried to propose a set of worldwide guidelines, criteria, information, and standards on some IPAs associated with human health, benzene and formaldehyde for example, to lead any countries in the world to scientifically carry out their certain management solutions or means in terms of the indoor air quality (IAQ). Along with emissions of various IAPs from building materials, excessive interior decorations as well as redundant aged buildings, the climatic characteristics of subtropical island Taiwan, higher temperature and relative humidity for instance, have been proven to be sort of unfavorable conditions for deteriorating the IAQ. Whereas above mentioned causes speak for themselves- the IAQ will for sure play an influential role on vital health of anyone who inevitably stays approximately 90% of his/her time in any generalized indoor circumstance in Taiwan. Consequently, Taiwan was the second country in the world, following the South Korea, to promulgate an enforcement domestic law ‘The Indoor Air Quality Management Act’ for regulating IAQ of Taiwan in 2011; it has been officially implemented in 2012. A variety of principal measures were definitely declared in this Act, including the TEPA (Taiwan Environmental Protection Administration) should announce a list of locations required to follow the act beforehand plus this list will be further expanded in the future, several certain locations should employ qualified and dedicated IAQ control specialists, some specific locations should build up their continuous emission monitoring system based on real-time requirements, to name a few. Meanwhile, a number of auxiliary regulations were enacted by the TEPA to implement this Act by a comprehensively approach. Owing to drive by the accessibility of Internet, innovative development of Information and Communication Technology (ICT) as well as global adoption of the ‘Open’ thinking in regard to software, hardware and data over the years, our contemporary society has undoubtedly been evolving into a new and ubiquitous pattern of cyber world centered by the conception of Internet of Things (IOT). In brief, the idea of IOT attempts to envision real-time application of automation intelligence and smart decision making by further developing a cyber-physical system (CPS) and an added-value data Cloud analytic model, Big Data analysis for example, on the basis of a large amount of data widely collected from nature or anthropogenic surrounding. Therefore, the IOT has been globally regarded as one kind of most well-known topic, outlook or action plan with trendy, innovative, interdisciplinary attributes and opportunities for upgrading any industries as well. Take the conception of what is called Industry 4.0 ( the fourth industrial revolutio or smart factory) for example, it in a manner aims to realize and generalize the core philosophy of the IOT through the current trend of automation intelligence, smart decision making thinking and data exchange in industrial technologies to facilitate transformation opportunities or turning points for any industrial sectors such as conventional industries, manufacturing industries, and so on. Compared the priority to enact an enforcement act for regulating IAQ with other countries in the world, this study proposes that the facts of health effect and risk correlated to IAQ should be disclosed and highlighted on the basis of intention and necessity of ‘The Indoor Air Quality Management Act’ promulgated in Taiwan in 2011. Obviously, it now is in urgent opportunity and need of making integration of multidisciplinary studies, and developing a comprehensive, smart and automation framework for systematically disclosing the environmental risk information to make a lot of sense to the public via merging additional, innovative IOT-oriented and automation thinking into our current IAQ regulating mechanism and managing system. Hence, this study not only takes above mentioned urgent necessities into account, but some required elements are also weighed within this study before making the definitions of our research requirements, limitations and assumptions; such as low energy consumption, technique and cost- effectiveness feasibility, system deployment facilitation, and so on. Accordingly, this study tries to take advantage of novel open hardware, Arduino for example, and be in compliance with the methodology of software engineering to practically build up a systematic research system for coming to a preliminary IOT-oriented IAQ management framework conclusion including comprehensive and smart automation managing mechanisms, environmental risk information disclosures, and CFD Simulation. In pragmatic, the results of this study could be concluded as a workable framework suggestion for facilitating management operation as well as communicating the real-time condition and health risk information of IAQ. Furthermore, the results of this study also could expectably call attention to the issues in terms of IAQ and some inputs of other researches as well.

One way to reduce the lifecycle cost and environmental impact of a product in a circular economy is to extend its lifespan by either creating longer-lasting products or managing the product properly during its use stage. Life extension of a product is envisioned to help better utilize raw materials efficiently and slow the rate of resource depletion. In the case of manufacturing equipment (e.g., an electric motor on a machine tool), securing reliable service life as well as the life extension are important for consistent production and operational excellence in a factory. However, manufacturing equipment is often utilized without a planned maintenance approach. Such a strategy frequently results in unplanned downtime, owing to unexpected failures. Scheduled maintenance replaces components frequently to avoid unexpected equipment stoppages, but increases the time associated with machine non-operation and maintenance cost.

Recently, the emergence of Industry 4.0 and smart systems is leading to increasing attention to predictive maintenance (PdM) strategies that can decrease the cost of downtime and increase the availability (utilization rate) of manufacturing equipment. PdM also has the potential to foster sustainable practices in manufacturing by maximizing the useful lives of components. In addition, advances in sensor technology (e.g., lower fabrication cost) enable greater use of sensors in a factory, which in turn is producing greater and more diverse sets of data. Widespread use of wireless sensor networks (WSNs) and plug-and-play interfaces for the data collection on product/equipment states are allowing predictive maintenance on a much greater scale. Through advances in computing, big data analysis is faster/improved and has allowed maintenance to transition from run-to-failure to statistical inference-based or machine learning prediction methods.

Moreover, maintenance practice in a factory is evolving from equipment “health management” to equipment “wellness” by establishing an integrated and collaborative manufacturing system that responds in real-time to changing conditions in a factory. The equipment wellness is an active process of becoming aware of the health condition and of making choices that achieve the full potential of the equipment. In order to enable this, a large amount of machine condition data obtained from sensors needs to be analyzed to diagnose the current health condition and predict future behavior (e.g., remaining useful life). If a fault is detected during this diagnosis, a root cause of a fault must be identified to extend equipment life and prevent problem reoccurrence.

However, it is challenging to build a model capturing a relationship between multi-sensor signals and mechanical failures, considering the dynamic manufacturing environment and the complex mechanical system in equipment. Another key challenge is to obtain usable machine condition data to validate a method.

A goal of the proposed work is to develop a systematic tool for maintenance in manufacturing plants using emerging technologies (e.g., AI, Smart Sensor, and IoT). The proposed method will facilitate decision-making that supports equipment maintenance by rapidly detecting a worn component and estimating remaining useful life. In order to diagnose and prognose a health condition of equipment, several data-driven models that describe the relationships between proxy measures (i.e., sensor signals) and machine health conditions are developed and validated through the experiment for several different manufacturing-oriented cases (e.g., cutting tool, gear, and bearing). To enhance the robustness and the prediction capability of the data-driven models, signal processing is conducted to preprocess the raw signals using domain knowledge. Through this process, useful features from the large dataset are extracted and selected, thus increasing computational efficiency in model training. To make a decision using the processed signals, a customized deep learning architecture for each case is designed to effectively and efficiently learn the relationship between the processed signals and the model’s outputs (e.g., health indicators). Ultimately, the method developed through this research helps to avoid catastrophic mechanical failures, products with unacceptable quality, defective products in the manufacturing process as well as to extend equipment service life.

To summarize, in this dissertation, the assessment of technical, environmental and economic performance of the AI-driven method for the wellness of mechanical systems is conducted. The proposed methods are applied to (1) quantify the level of tool wear in a machining process, (2) detect different faults from a power transmission mini-motor testbed (CNN), (3) detect a fault in a motor operated under various rotation speeds, and (4) to predict the time to failure of rotating machinery. Also, the effectiveness of maintenance in the use stage is examined from an environmental and economic perspective using a power efficiency loss as a metric for decision making between repair and replacement.