Uma Abordagem de Middleware com Suporte à Qualidade de Contexto voltada para Aplicações de Internet das Coisas.

Abstract

Quality of Context (QoC) in general, that is, both Quality of Information (QoI) and Quality of Service (QoS), has a direct and significant impact on IoT applications users’ experiences, which can be positive or negative, depending on the type and level of QoC available for application. This doctoral thesis approaches as a research problem the fact that it is very complex to develop IoT applications, especially those having context quality requirements. Although several middleware approaches have been proposed in the literature, few of them offer support for applications that have QoC requirements. In general, related work still has limitations on some aspects, such as heterogeneous sensors management, distributed communication, reliability in mobile environments, provisioning, evaluation and monitoring of QoC, mainly in relation to quality of service, which suffers from the lack of parameters. In order to overcome some of these limitations and facilitate the development of IoT applications that have QoC requirements, this thesis proposed a new approach to context middleware, called M-hub/CDDL. The proposed solution presents mechanisms to support management and context quality that are broader and more efficient than the related approaches in several aspects, mainly in relation to management sensors, provisioning, evaluation, and monitoring of QoC. The quantitative evaluation of middleware has proven the efficiency of the M-Hub/CDDL in relation to data delivery time, reliability in unstable networks, monitoring time and resource consumption (memory and battery). The proof of concept has shown that the M-Hub/CDDL mechanisms apply to several application domains, while the results of the case study with users showed a high degree of satisfaction with the programming abstractions, services and QoC support provided by middleware. Therefore, the results show the efficiency of the MHub/CDDL as a facilitating approach to the IoT applications development with QoC requirements, showing that the objectives were achieved.