Advances in smart technologies, wireless networking, and increased interest in contextual services have led to the emergence of ubiquitous and pervasive computing as one of the most promising areas of computing in recent years. Smart Spaces, in particular, have gained significant interest within the research community. Currently, most Smart Spaces rely on physical components, such as sensors, to acquire information about the real-world environment. Although current sensor networks can acquire some useful contextual information from the physical environment, their information resources are often limited, and the data acquired is often unreliable. We argue that by introducing social network information into such systems, smarter and more adaptive spaces can be created. Social networks have recently become extremely popular, and are now an integral part of millions of people’s daily lives. Through social networks, users create profiles, build relationships, and join groups, forming intermingled sets and communities. Social Networks contain a wealth of information, which, if exploited properly, can lead to a whole new level of smart contextual services. A mechanism is therefore needed to extract data from heterogeneous social networks, to link profiles across different networks, and to aggregate the data obtained. We therefore propose the design and implementation of a Smart Spaces framework that utilizes the social context. In order to manage services and sessions, we integrate our system with the IP Multimedia Subsystem. Our system, which we call SocioSpace, includes full design and implementation of all components, including the central server, the location management system, the social network interfacing system, the service delivery platform, and user agents. We have built a prototype for proof of concept and carried out exhaustive performance analysis; the results show that SocioSpace is scalable, extensible, and fault-tolerant. It is capable of creating Smart Spaces that can truly deliver adaptive services that enhance the users’ overall experience, increase their satisfaction, and make the surroundings more beneficial and interesting to them. / Thesis (Ph.D, Computing) -- Queen's University, 2012-04-30 11:14:29.054

A key enabler of pervasive computing is the ability to drive service delivery through the analysis of situations: Semantically meaningful classifications of system state, identified through analysing the readings from sensors attached to the everyday objects that people interact with. Situation recognition is a mature area of research, with techniques primarily falling into two categories. Knowledge-based techniques use inference rules crafted by experts; however often they compensate poorly for sensing peculiarities. Learning-based approaches excel at extracting patterns from noisy training data, however their lack of transparency can make it difficult to diagnose errors. In this thesis we propose a novel hybrid approach to situation recognition that combines both techniques. This offers improvements over each used individually, through not sacrificing the intelligibility of the decision processes that the use of machine learning alone often implies, and through providing better recognition accuracy through robustness to noise typically unattainable when developers use knowledge-based techniques in isolation. We present an ontology model and reasoning framework that supports the uniform modelling of pervasive environments, and infers additional knowledge from that which is specified, in a principled way. We use this as a basis from which to learn situation recognition models that exhibit comparable performance with more complex machine learning techniques, while retaining intelligibility. Finally, we extend the approach to construct ensemble classifiers with either improved recognition accuracy, intelligibility or both. To validate our approach, we apply the techniques to real-world data sets collected in smart-office and smart-home environments. We analyse the situation recognition performance and intelligibility of the decision processes, and compare the results to standard machine learning techniques and results published in the literature.

Les défis que l'Internet des objets posent sont à la mesure des transformations que cette technologie est susceptible d'entraîner dans notre rapport quotidien à notre environnement. Nos propres objets, et des milliards d'autres, disposeront de capacités de traitement des données et de connexion au réseau, certes limitées mais effectives. Alors, ces objets se doteront d'une dimension numérique, et deviendront accessibles d'un façon tout à fait nouvelle. Ce n'est pas seulement la promesse d'un accès original à l'objet, mais bel et bien l'avènement d'une nouvelle perception et interaction avec ce qui nous entoure. Les applications de l'Informatique ubiquitaire utiliseront majoritairement les interactions entre objets, et la somme de leurs actions/réactions offrira une véritable valeur ajoutée. Mais l'hétérogénéité des composants matériels et des réseaux empruntés freine considérablement l'essor de l'Internet des objets. L'objectif de cette thèse est de proposer une solution effective et le cadre nécessaire à la construction de telles applications. Après avoir montré la pertinence des solutions chorégraphiées et quantifié le gain acquis sur des structures de communication arborescentes, nous présenterons D-LITe, notre framework, qui appréhende chaque objet comme étant fournisseur de services. Grâce à son approche REST assurant l'interopérabilité dans l'assortiment des composants et réseaux de l'Internet des objets, le framework D-LITe, hébergé par chaque objet (et adapté à ses contraintes), fournit un contrôle distant, aussi bien pour sa reprogrammation dynamique que les échanges avec ses partenaires. Nous poursuivrons en présentant SALT, le langage de programmation compris par D-LITe, basé sur les transducteurs à états fini. Outre son expressivité étendue aux particularités du domaine, SALT accorde un accès aux fonctionnalités de l'objet au travers d'une couche d'abstraction matérielle. Enfin, profitant de la standardisation offerte par D-LITe pour la programmation de chaque composant en particulier, une solution de composition, BeC3, va offrir un moyen efficace pour construire une application complète par assemblage des comportement distribués, tout en respectant la cohérence des interactions entre objets, par l'intermédiaire d'une abstraction des échanges et de leur modélisation. Aussi sommes-nous, par la résolution des problématiques rencontrées à chacun des différents niveaux, capables de présenter une solution simple, cohérente et fonctionnelle à même de bâtir réellement et efficacement des applications robustes pour l'Internet des objets / The challenges which the Internet of objects put are for the measure of the transformations which this technology may pull in our daily report to our environment. Our own objects, and billions of others, will have processing capacities of the data and connection to the network, certainly limited but effective. Then, these objects will be equipped with a digital dimension, and will become accessible in a completely new way. It is not only the promise of an original access to the object, but well and truly the succession of a new perception and an interaction with what what surrounds us. The applications of the ubiquitaire Computing will use mainly the interactions between objects, and the sum of their actions / reactions will offer a real added value

Submitted by Cássia Santos (cassia.bcufg@gmail.com) on 2014-09-19T12:27:42Z No. of bitstreams: 2 Dissertacao Marcos Paulino Roriz Junior.pdf: 6606924 bytes, checksum: 51f41fd9bffd47d74d5b5433034ffe62 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2014-09-19T12:58:45Z (GMT) No. of bitstreams: 2 Dissertacao Marcos Paulino Roriz Junior.pdf: 6606924 bytes, checksum: 51f41fd9bffd47d74d5b5433034ffe62 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2014-09-19T12:58:45Z (GMT). No. of bitstreams: 2 Dissertacao Marcos Paulino Roriz Junior.pdf: 6606924 bytes, checksum: 51f41fd9bffd47d74d5b5433034ffe62 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2013-05-17 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / According to Mark Weiser, ubiquitous computing focuses on seamlessly integrating computing tasks into people’s daily lives. Because of current technology limitations, the realization of ubiquitous computing observes a limited set of aspects of ubiquitous computing, such as, mobility and context, which are based on services that integrate the users with the resources that are present on a delimited ubiquitous environment (such as in smart spaces). Instead, we explored a different approach, in which services are used not to integrate an individual user with the environment, but to integrate the users present in the environment with one another. One way to realize this aspect is by using content sharing, first-class application dat, that serve as integration medium between users. However, due to the environment complexity and lack of middleware platforms, applications that follow this approach are repeatedly built from scratch using raw techniques. Aiming to provide an infrastructure for the development of this kind of applications, we propose Content Sharing for Smart Spaces (C3S), a middleware that offers a high-level programming model using primitives that are based on a set of content sharing semantics and ubiquitous application concepts. The primitives express a small set of behaviors, such as move, clone, and mirror, which serve as building blocks for developers to implement sharing and content ubiquity features, while the ubiquitous concepts supported by the middleware allow the manipulation of users, groups and ubiquitous applications. We validated our proposal using two different case studies that allowed us to explore these features. Our results show that our middleware provides an easier way to develop sharing-based applications compared to related work found in the literature. / De acordo com Mark Weiser, a computação ubíqua se concentra na integração de maneira despercebida e sem rupturas (seamlessy) de tarefas da computação no cotidiano das pessoas. Por causa das atuais limitações tecnológicas, a realização dessa integração segue um ou mais aspectos da computação ubíqua, por exemplo, de mobilidade ou de contexto, que são baseados em serviços que integram o usuário em um ambiente ubíquo delimitado (como espaços inteligentes). Neste trabalho exploramos uma abordagem diferente, em que os serviços não são utilizados para integrar um usuário individual ao ambiente, mas são utilizados para integrar os usuários presentes no ambiente uns com os outros. Uma maneira de realizar esse aspecto é usando o compartilhamento de conteúdo, dados de primeira classe da aplicação que servem como meio de integrar os usuários. No entanto, devido à complexidade do ambiente de computação ubíqua e à falta de plataformas de middleware, aplicações que seguem esta abordagem são repetidamente construídas a partir “do zero”, usando técnicas não convencionais. Com o objetivo de fornecer uma infraestrutura para o desenvolvimento deste tipo de aplicação, propomos o Content Sharing for Smart Spaces (C3S), um middleware que oferece um modelo de programação de alto nível, usando primitivas baseadas em um conjunto de semânticas de compartilhamento de conteúdo e em conceitos de aplicações ubíquas. As primitivas expressam um conjunto de comportamentos, tais como mover, clonar, e espelhar, que servem como blocos de construção para os desenvolvedores implementarem funcionalidades de compartilhamento, enquanto que os conceitos de ubiquidade permitem a manipulação de usuários, grupos e aplicações ubíquas. A proposta foi validada por meio de dois estudos de caso que exploram esses recursos. Os resultados permitiram concluir que o middleware fornece uma maneira mais fácil de desenvolver aplicativos baseados em compartilhamento em comparação com trabalhos semelhantes encontrados na literatura.

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2017-05-19T17:39:41Z No. of bitstreams: 2 Tese - Leandro Alexandre Freitas - 2017.pdf: 10640805 bytes, checksum: 61b69f91cb32e17075d698eecf19b8c4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-05-22T10:42:45Z (GMT) No. of bitstreams: 2 Tese - Leandro Alexandre Freitas - 2017.pdf: 10640805 bytes, checksum: 61b69f91cb32e17075d698eecf19b8c4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-05-22T10:42:45Z (GMT). No. of bitstreams: 2 Tese - Leandro Alexandre Freitas - 2017.pdf: 10640805 bytes, checksum: 61b69f91cb32e17075d698eecf19b8c4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-04-04 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / The growth and popularization of wireless connectivity and of mobile devices has allowed the development of smart spaces that were previously only envisaged in the approach proposed by Mark Weiser. These smart spaces are composed of many computational resources, such as devices, services and applications, along with users, who must be able to associate with these features. However, programming these environments is a challenging task, since smart spaces have a dynamic nature, resources are heterogeneous, and it is necessary that interactions between users and devices are coordinated with one another. In this work, we present a new approach for smart spaces programming using Models@RunTime. In this regard, we propose a high level modeling language, called Smart Spaces Modeling Language (2SML), in which the user is able to model the smart space with all elements that can be part of it. Such models are developed by the users, interpreted and effected in the physical space by a model execution engine, called Smart Space Virtual Machine (2SVM), whose development is part of this work. / O crescimento e a popularização cada vez maior da conectividade sem fio e dos dispositivos móveis, tem permitido a construção de espaços inteligentes que antes eram vislumbrados apenas na proposta de computação ubíqua do cientista da Xerox PARK, Mark Weiser. Esses espaços inteligentes são compostos por diversos recursos computacionais, como dispositivos, serviços e aplicações, além de usuários, que devem ser capazes de se associar a esses recursos. Entretanto, a programação destes ambientes é uma tarefa desa- fiadora, uma vez que os espaços inteligentes possuem uma natureza dinâmica, os recursos se apresentam de forma heterogênea e é necessário que as interações entre usuários e dispositivos sejam coordenadas. Neste trabalho desenvolvemos uma nova abordagem para programação de espaços inteligentes, por meio de modelos em tempo de execução. Para isso, propomos uma linguagem de modelagem de alto nível, denominada Smart Space Modeling Language (2SML), em que o usuário é capaz de modelar o espaço inteligente com todos os elementos que dele podem fazer parte. Esse modelo desenvolvido pelo usuá- rio é interpretado e realizado no espaço físico por uma máquina de execução de modelos, denominada Smart Space Virtual Machine (2SVM), cujo desenvolvimento é parte deste trabalho.