O problema de lidar com a mobilidade dos usuários existe desde que os dispositivos móveis se tornaram capazes de lidar com conteúdo multimídia e ainda é um dos desafios mais relevantes na área de redes de computadores. A arquitetura de Internet convencional é inadequada em lidar com um número cada vez maior de dispositivos móveis que estão tanto consumindo quanto produzindo conteúdo. Named Data Networking (NDN) é uma arquitetura de rede que pode potencialmente superar este desafio de mobilidade. Ela suporta a mobilidade do consumidor nativamente, mas não oferece o mesmo nível de suporte para a mobilidade de conteúdo. A mobilidade de conteúdo exige garantir que os consumidores consigam encontrar e recuperar o conteúdo desejado mesmo quando o produtor correspondente (ou o hospedeiro principal) não estiver disponível. Nesta tese, propomos o PDRM (Proactive Data Replication Mechanism), um mecanismo de replicação de dados proativo e consciente de localização, que aumenta a disponibilidade de conteúdo através da redundância de dados no contexto da arquitetura NDN. Ele explora os recursos disponíveis dos usuários finais na vizinhança para melhorar a disponibilidade de conteúdo, mesmo no caso da mobilidade do produtor. Ao longo da tese, discutimos o projeto do PDRM, avaliamos o impacto do número de provedores disponíveis na vizinhança e a capacidade de cache na rede em sua operação e comparamos seu desempenho com NDN padrão e duas propostas do estado-da-arte. A avaliação indica que o PDRM melhora o suporte à mobilidade de conteúdo devido ao uso de informações de popularidade dos objetos e recursos extras na vizinhança para ajudar a replicação pró-ativa. Os resultados mostram que o PDRM pode reduzir os tempos de download até 53,55%, o carregamento do produtor até 71,6%, o tráfego entre domínios até 46,5% e a sobrecarga gerada até 25% em comparação com NDN padrão e os demais mecanismos avaliados. / The problem of handling user mobility has been around since mobile devices became capable of handling multimedia content and is still one of the most relevant challenges in networking. The conventional Internet architecture is inadequate in dealing with an ever-growing number of mobile devices that are both consuming and producing content. Named Data Networking (NDN) is a network architecture that can potentially overcome this mobility challenge. It supports consumer mobility by design but fails to offer the same level of support for content mobility. Content mobility requires guaranteeing that consumers manage to find and retrieve desired content even when the corresponding producer (or primary host) is not available. In this thesis, we propose PDRM, a Proactive and locality-aware Data Replication Mechanism that increases content availability through data redundancy in the context of the NDN architecture. It explores available resources from end-users in the vicinity to improve content availability even in the case of producer mobility. Throughout the thesis, we discuss the design of PDRM, evaluate the impact of the number of available providers in the vicinity and in-network cache capacity on its operation, and compare its performance to Vanilla NDN and two state-of-the-art proposals. The evaluation indicates that PDRM improves content mobility support due to using object popularity information and spare resources in the vicinity to help the proactive replication. Results show that PDRM can reduce the download times up to 53.55%, producer load up to 71.6%, inter-domain traffic up to 46.5%, and generated overhead up to 25% compared to Vanilla NDN and other evaluated mechanisms.

The Whereabouts Screens are a minimalistic publishing system offering location awareness at the workplace. The employees can share information with their colleagues, relevant mostly to their whereabouts, by publishing it on several situated displays located around the office space. The Whereabouts Screens have been installed with minimal changes since 2012 at the Mobile Life center and as a relatively new system, it has still not reached its full potential. Despite its “primitive” characteristics however, the employees show a continued interest in using them. Based on related research on situated displays the research questions to be answered are: What do the semi-public situated displays in the workplace have to offer to their users? How can they be augmented and better integrated in their environment using the existing technology? To investigate these questions, Human-Computer Interaction methods such as in-situ observations, interviews, a design workshop and prototyping were used. The result was a prototype of an augmented system, in the form of a notice board where the users can post notes of different types (text, images, etc.). The board is extended beyond the limits of the screen with LEDs. The intensity and colors of the LEDs are affected by the use of the system, and illuminate the area around the screen. The augmented Whereabouts Screens aim to promote social interaction and location awareness at the workplace. They were evaluated through user testing and the first results show a positive reception from the users. / Whereabouts Skärmar är ett minimalistiskt publicerings-system som bjuder närvarokännedom på arbetsplatsen. De anställda kan dela information med sina kollegor, mestadels kopplat till vad de befinner sig, genom att publicera det på flera  skärmar belägna runt om på arbetsplatsen. Whereabouts Skärmar har funnits installerade med minimala förändringar sedan 2012 vid Mobile Life centret men har  som ett relativt nytt system ännu inte nått sin fulla potential. Trots begränsningar i systemet visar anställda ett fortsatt intresse av att använda dem. Baserat på relaterad forskning kring situerade skärmar tar sig uppsatsen an två frågeställningar: Vad har semi-publika situerade skärmar på arbetsplatsen att erbjuda sina användare?  Hur kan deras funktionalitet utökas och integreras bättre i omgivningen med hjälp av befintlig teknik? För att undersöka dessa frågor, används människa-datorinteraktion metoderna in-situ observationer, intervjuer, designworkshops och prototypbyggande använts. Resultatet är en prototyp av ett system med utökad funktionalitet i form av en anslagstavla där användare kan lägga upp anteckningar av olika slag (text, bilder, etc.). Anslagstavlan sträcker sig över gränserna för skärmen med LEDljusen.  Intensitet och färger hos LEDljusen påverkas av användningen av systemet, och lyser upp området runt skärmen. De utökade Whereabouts skärmarna syftar till att understödja social interaktion och  närvarokännedom på arbetsplatsen. De utvärderades genom användartester och de första resultaten visar på ett positivt mottagande av användarna.

Today most cellular phones, personal digital assistants, PCs, etc. offer an electronic calendar. Electronic calendars are especially useful for people who have many different meetings each day and who need to know when the meetings start and who is involved in each meeting. With the aid of a program a calendar can be published on web, shared with other people to enable collaboration, or synchronized between different devices. Current calendaring software offers an almost unlimited set of features and services. However, today such software does not utilize context-awareness, for example exploiting knowledge of the user's location. When people collaborate they often need to meet in order to do a task jointly or discuss something. It can be difficult to plan a meeting because people have booked their available time differently in their calendars. Because of this there is a need to automatically schedule certain types of meetings. In this thesis, a program that schedules meetings automatically is designed, implemented and evaluated. This program facilitates collaboration by finding a commonly available time and/or meeting place for a meeting, thus making it easier for the meeting people to agree. When meetings are scheduled without requiring too much attention from a user and the number of human errors can be reduced while planing a meeting, users do not need to expends as much effort as it goes into scheduling meetings today. Because today a company planing a collaboration task collectively spends a lot of time and effort searching for a commonly available time with this effort increasing non-linearly with increased numbers of participants companies can obviously benefit from automated scheduling systems. Testing with the application reveals that incorporating of user's location information into scheduling is a great tool to facilitate collaboration. The survey also shows the need for extensions to the developed application; with the new features utilizing location information. The evaluation also shows that the developed scheduling program has managed to reduce the time and effort spent while scheduling meetings. / Idag finns det en elektronisk kalender i de flesta mobiltelefoner, datorer och PDA:n. Elektroniska kalendrar är användbara framför allt för människor som har flera möten varje dag och som behöver veta när mötena startar vilka som ska delta. Vissa elektroniska kalendrar kan publiceras på webben, delas med andra människor för att möjliggöra samarbete och synkroniseras mellan till exempel mobiltelefoner, datorer och PDA:n. Kalendermjukvara erbjuder idag ett nästan obegränsat antal funktioner och nyttiga tjänster. Denna typ av mjukvara är dock generellt sett inte medveten om information såsom användarens position, vilket i sammanhanget kalls context-awareness. När människor ska samarbeta krävs ofta att de träffas för att utföra uppgifter tillsammans eller diskutera viktiga ämnen. För att kunna ha möten krävs att möten först planeras, vilket kan vara svårt då de inbjudna är olika uppbokade i sina respektive kalendrar. Av den anledningen finns ett behov av att automatisera schemaläggningen för vissa typer av möten. I detta examensarbete skall ett program för automatisk schemaläggning designas, utvecklas och evalueras. Programmet skall underlätta samarbete mellan mötesdeltagare genom att ta över uppgiften att hitta en gemensam tid och/eller plats för ett möte. Programmet skall därmed också underlätta för mötesdeltagarna att komma överens. När möten kan schemaläggas utan att det kräver för mycket uppmärksamhet från användarna och antalet mänskliga fel kan reduceras när möten planeras, behöver man inte lägga lika mycket arbete, som idag, på att schemalägga möten. Eftersom det för tillfället krävs mycket tid och resurser för ett företag för att schemalägga ett möte, samtidigt som tiden för att planera ett möte inte ökar linjärt med antalet deltagare, kommer företag antagligen att dra nytta av ett automatiserat schemaläggningssystem. En undersökning genomförd av ett antal testpersoner som använt applikationen visade på att användarens position var en viktig parameter som kunde förbättra schemaläggningen av möten. Undersökningen visade också att applikationen hade ett stort behov av att vidareutvecklas genom nya potentiella funktioner som tar hänsyn till användarens position. Men viktigast av allt så visade undersökningen på att applikationen lyckats med att reducera tiden det tar för att planera möten.

No description available.

This work has two objectives: a) to predict people's future locations, and b) to predict when they will be at given locations. Current location-based applications react to the user's current location. The progression from location-awareness to location-prediction can enable the next generation of proactive, context-predicting applications. Existing location-prediction algorithms predict someone's next location. In contrast, this dissertation predicts someone's future locations. Existing algorithms use a sequence of locations and predict the next location in the sequence. This dissertation incorporates temporal information as timestamps in order to predict someone's location at any time in the future. Sequence predictors based on Markov models have been shown to be effective predictors of someone's next location. This dissertation applies a Markov model to two-dimensional, timestamped location information to predict future locations. This dissertation also predicts when someone will be at a given location. These predictions can support presence or understanding co-workers’ routines. Predicting the times that someone is going to be at a given location is a very different and more difficult problem than predicting where someone will be at a given time. A location-prediction application may predict one or two key locations for a given time, while there could be hundreds of correct predictions for times of the day that someone will be in a given location. The approach used in this dissertation, a heuristic model loosely based on Market Basket Analysis, is the first to predict when someone will arrive at any given location. The models are applied to sparse, WiFi mobility data collected on PDAs given to 275 college freshmen. The location-prediction model predicts future locations with 78-91% accuracy. The temporal-prediction model achieves 33-39% accuracy. If a tolerance of plus/minus twenty minutes is allowed, the prediction rates rise to 77%-91%. This dissertation shows the characteristics of the timestamped, location data which lead to the highest number of correct predictions. The best data cover large portions of the day, with less than three locations for any given timestamp. / Ph. D.

Master of Philosophy (Architecture) / Context-aware computing is a mobile computing paradigm in which applications can discover, use, and take advantage of contextual information, such as the location, tasks and preferences of the user, in order to adapt their behaviour in response to changing operating environments and user requirements. A problem that arises is the inability to respond to contextual information that cannot be classified into any known context. Many context-aware applications require all discovered contextual information to exactly match a type of context, otherwise the application will not react responsively. The ability to learn and recall contexts based on the contextual information discovered has not been very well addressed by previous context-aware applications and research. The aim of this thesis is to develop a component middleware technology for mobile computing devices for the discovery and capture of contextual information, using the situated reasoning concept of constructive memory. The research contribution of this thesis lies in developing a modified architecture for context-aware systems, using a constructive memory model as a way to learn and recall contexts from previous experiences and application interactions. Using a constructive memory model, previous experiences can be induced to construct potential contexts, given a small amount of learning and interaction. The learning process is able to map the many variations of contextual information currently discovered by the user with a predicted type of context based on what the application has stored and seen previously. It only requires a small amount of contextual information to predict a context, something common context-aware systems lack, as they require all information before a type of context is assigned. Additionally, some mechanism to reason about the contextual information being discovered from past application interactions will be beneficial to induce contexts for future experiences.

This thesis describes research on methods for Ubiquitous/Pervasive Computing to better suit users in an Intelligent Environment. The approach is to create and equip a computing environment, such as our Active Office, with technologies that can identify user needs and meet these need in a timely, efficient and unobtrusive manner.¶ The critical issues in the Intelligent Environment are how to enable transparent, distributed computing to allow continued operation across changing circumstances and how to exploit the changing environment so that it is aware of the context of user location, the collection of nearby people and objects, accessible devices and changes to those objects over time.¶ Since the Intelligent Environment is an environment with rapid and rich computing processing, the distributed context processing architecture (DiCPA) was developed to manage and respond to rapidly changing aggregation of sensor data. This architecture is a scalable distributed context processing architecture that provides: 1. continued operation across changing circumstances for users, 2. the collection of nearby people and objects, 3. accessible devices and 4. the changes to those objects over time in the environment. The DiCPA approach focuses on how the Intelligent Environment provides context information for user location, user mobility and the user activity model. Users are assumed mobile within the Intelligent Environment and can rapidly change their access to relevant information and the availability of communications and computational resources.¶ Context-Aware Computing is a new approach in software engineering for Intelligent Environment. It is an approach in the design and construction of a context-aware application that exploits rapid changes in access to relevant information and the availability of communication and computing resources in the mobile computing environment. The goal of Context-Aware Computing is to make user interaction with the computer easier in the smart environment where technology is spread throughout (pervasive), computers are everywhere at the same time (ubiquitous) and technology is embedded (ambient) in the environment. Context-aware applications need not be difficult, tedious or require the acquisition of new skills on the part of the user. They should be safe, easy, simple to use and should enable new functionality without the need to learn new technology. They should provide relevant information and a simple way for a user to manage.¶ The Intelligent Environment requires a context-aware application to improve its efficiency and to increase productivity and enjoyment for the user. The context awareness mechanism has four fundamental cores i.e. identity (who), activity (what), location (where) and timestamp (when). Based on DiCPA architecture, the model of user location (where), user mobility (where), user activity (what) and Intelligent Environment response (what) were developed. Prototypes were also developed to proof the Context-Aware Computing concept in the Intelligent Environment.¶ An Intelligent Environment uses the multi-disciplinary area of Context-Aware Computing, which combines technology, computer systems, models and reasoning, social aspects, and user support. A “good quality” project for Context-Aware Computing requires core content and provides iterative evaluation processes, which has two types of iteration: design and product iteration of the evaluation. The aim of the development of an evaluation program in Context-Aware Computing is to determine what to test, how to test and the appropriate metrics to use. This work presents the metrics for a good quality project in the Context-Aware Computing area, which is followed by the evaluation of the prototypes of this work.

<p>For the past recent years mobile developers and content providers have been aware of consumers’ engagement in online social networking. New opportunities are presented when using a mobile device as a medium for social interaction, also referred to as mobile social software. An ever growing effort aimed at various mobile design processes and applications have taken place. To achieve a widespread adoption of mobile social software amongst the main stream mobile consumer, several problem areas have to be taken into consideration, such as the technological possibilities and limitations of mobile phones, security issues regarding user privacy and the different aspects of consumers appeal for social media. This research looks into the overall situation of the current field of research and focuses on the interests in mobile social software amongst users. The study was geographically limited to Malmö, Sweden and focused on a population at the ages of 13-18.</p>

El uso creciente de dispositivos móviles, infraestructuras inalámbricas e Internet está cambiando nuestra vida diaria, no solo la manera en que nos comunicamos o compartimos información, sino también cómo nos relacionamos con el entorno. A través de nuestras interacciones con estas tecnologías, accedemos y generamos una membrana de información que se cierne sobre los espacios donde vivimos y que visitamos. Sin embargo, esta capa de información solo modela de manera imperfecta la realidad debido a una digitalización tosca y a limitaciones tecnológicas, que hacen peligrar la interacción humana. Por otro lado, la presencia de esta información geográfica ubicua generada por los usuarios abre nuevas perspectivas para la comprensión de las actividades humanas en el espacio y el tiempo. Esta tesis acepta el reto de investigar estos aspectos de las interacciones humanas con la información geográfica ubicua. Con un enfoque cualitativo y cuantitativo, discernimos las implicaciones de la integración de información geográfica ubicua y las resultantes estrategias de los usuarios para hacer frente a la incertidumbre espacial. Entonces, explotamos esta contribución para analizar enfoques novedosos con el objetivo de inferir actividades espacio-temporales de individuos y grupos respetando su privacidad. Demostramos la aplicabilidad de nuestras soluciones en los ámbitos de la investigación de mercados y el urbanismo. / The increasing use of mobile devices, wireless infrastructures, and the Internet is changing our daily lives, not only in the way we communicate with each other or share information but also how we relate to the environment. Through our interactions with these technologies we access and generate an informational membrane, hovering over the spaces we live in and visit. However, this information layer only imperfectly models the reality due to coarse digitization and technological limitations, challenging the human interaction. On the other hand, the presence of this user-generated ubiquitous geographic information opens novel perspectives in understanding human activities over space and time. This thesis takes on the challenge of exploring these aspects of human interactions with ubiquitous geographic information. Through qualitative lenses, we discern the implications of the integration of ubiquitous geographic information and the resulting users strategies to cope with spatial uncertainty. Then, we exploit this contribution to explore novel approaches to infer individuals and groups time-space activities with respect to their privacy. We demonstrate the applicability of our solutions in the domains of market research and urbanism.

Master of Philosophy (Architecture) / Context-aware computing is a mobile computing paradigm in which applications can discover, use, and take advantage of contextual information, such as the location, tasks and preferences of the user, in order to adapt their behaviour in response to changing operating environments and user requirements. A problem that arises is the inability to respond to contextual information that cannot be classified into any known context. Many context-aware applications require all discovered contextual information to exactly match a type of context, otherwise the application will not react responsively. The ability to learn and recall contexts based on the contextual information discovered has not been very well addressed by previous context-aware applications and research. The aim of this thesis is to develop a component middleware technology for mobile computing devices for the discovery and capture of contextual information, using the situated reasoning concept of constructive memory. The research contribution of this thesis lies in developing a modified architecture for context-aware systems, using a constructive memory model as a way to learn and recall contexts from previous experiences and application interactions. Using a constructive memory model, previous experiences can be induced to construct potential contexts, given a small amount of learning and interaction. The learning process is able to map the many variations of contextual information currently discovered by the user with a predicted type of context based on what the application has stored and seen previously. It only requires a small amount of contextual information to predict a context, something common context-aware systems lack, as they require all information before a type of context is assigned. Additionally, some mechanism to reason about the contextual information being discovered from past application interactions will be beneficial to induce contexts for future experiences.