Creación de una infraestructura auto adaptativa de monitoreo mediante un middleware de configuración de componentes sensibles al contexto en ambientes de vida asistida

Abstract

Today technology is part of our daily life, since it helps to solve different needs and provides solutions that facilitate the development of activities, be they business, educational, health, environmental, etc. In addition to providing self-sufficiency to people facing daily tasks, not only common users, but also people with special abilities and skills, such as the elderly or people with disabilities; this to improve their quality of life and grant them greater independence, whether in care centers or family residential settings. Several investigations about assistive technologies have been carried out, based on a new paradigm of information technology called "ambient intelligence"; This paradigm helps to enhance the capacities of the elderly, people with functional diversity or people with chronic diseases, through digital environments that are sensitive, adaptable and respond to human needs. From the sense, support and support to the elderly, the concept of Assisted Living Environments (AAL - Ambient Assisted Living) was born, which can be described as intelligent assistance systems for a better and safer life. Therefore, an intelligent system must be composed of different “objects” (household appliances, sensors, actuators, etc.) from everyday life interconnected to each other to create services based on the detection of a physical event without the need for human intervention. direct and to facilitate tasks at home. Nowadays, the provision of advanced services through the interconnection of objects is what is called the internet of things (IoT - Internet of things). Bringing together the terms of AAL and IoT results in modern execution environments, which have become increasingly decentralized, heterogeneous, uncertain, and changeable. This has led the software engineering community to propose innovative ways to build, run and manage systems and services that allow the software and its models to reconfigure itself without the need to be rebuilt. In addition, it is necessary to address new theories, mechanisms, and appropriate software modeling methods to build self-adaptive environments endowed with the IoT, which is why runtime models are a tool that allows addressing the complexity of dynamic adaptations, to maintain an abstract model of a system at run time. Therefore, it is necessary to conceptualize the run-time models as simple representations that are built for specific purposes, in such a way that through a causal relationship, if a system changes its associated model also changes or vice versa. Consequently, the challenge in our degree work is to adapt and self-configure digital entities (cloud platforms, gateway, IoT devices) and physical entities (system users, environments), automatically, in an architecture for IoT deployed without need to stop the running system. In this aspect, to overcome the challenge, our vision will be focused through the models at runtime (M@rt - Models at runtime) to monitor and collect data, as well as to reduce the time it would take to restart the system and lose context sensitive information. In the present degree work, the design of an infrastructure focused on monitoring is proposed, which allows generating data for the subsequent support of the self-adaptation of the different components and / or IoT devices through a monitoring middleware at runtime. In addition, as an implementation of the proposed solution and as an additional deliverable, a DSL (Domain Specific Language - Lenguaje de dominio especifico) called Monitor-IoT has been developed, which provides the designer and developer of solutions applied to a case study with a tool for the development of architectures for the internet of things and allows analyzing them from the point of view of user perceptions.