Arquitectura Software para el Enriquecimiento del Contexto y Detección de Eventos Significativos en Dominios Smart Everything del Internet de las Cosas Colaborativo

The exponential growth of the Internet of Things (IoT) and the proliferation of Smart Everything applications have driven the development of intelligent environments capable of monitoring, processing and reacting to large volumes of data in real time. However, most current solutions continue to operate in isolation within a single application domain, which limits information contextualisation and misses a significant opportunity to improve it by discarding data from domains without an apparent relationship. Furthermore, the absence of a common criterion for defining and structuring data generated across different domains hinders their integration and correlation. This doctoral thesis addresses these limitations by proposing a software architecture aimed at improving contextualisation through the detection of situations of interest and their management via contextualised actions in collaborative IoT and Smart Everything environments. First, a taxonomy is defined that enables the homogeneous definition of data from multiple IoT domains, providing a set of extensible common attributes that can adapt to specific requirements without losing semantic coherence. Based on this taxonomy, a JSON-based data structure is designed to facilitate the structured exchange of information among different IoT domains. Building on this taxonomy and data structure, a software architecture is proposed that integrates messaging brokers, real-time processing through Complex Event Processing (CEP), and storage and notification systems. This architecture enables the integration and correlation of simple events from different domains to detect situations of interest in real time, thereby improving contextualisation in multi-domain environments. However, it is not possible to define contextualised actions exclusively at the CEP engine level. Moreover, CEP engines can become complex and difficult to maintain as the number of CEP patterns and situations of interest to be detected increases. To overcome this limitation, the thesis incorporates Business Process Management (BPM) as a complementary mechanism within the proposed software architecture for the definition and management of contextualised actions, as well as for the definition of the CEP pattern logic and their real-time updates. This integration improves the maintainability, adaptability and usability of the system, facilitating its evolution in response to changes in context or domain requirements. The feasibility and usefulness of the proposal are validated through the implementation of the software architecture and the development of representative case studies, together with performance and usability evaluations. The results obtained demonstrate that the solution enables the integration of data from heterogeneous domains, improves information contextualisation, and manages intelligent actions in a more scalable and maintainable way when handling an increasing number of situations of interest compared with traditional solutions based on isolated domains. In summary, this thesis presents a proposal that combines a multi-domain taxonomy, a CEP-based software architecture, and BPM integration for the management of contextualised actions in collaborative IoT and Smart Everything environments.

Author dc:creator

• Bazán Muñoz, Adrián

Advisors dc:contributor.advisor

• Ortiz Bellot, Guadalupe
• García de Prado Fontela, Alfonso

• Context Awareness
• Consciencia del contexto
• Internet of Things
• Internet de las Cosas
• Business Process Model
• Modelo de Procesos de Negocio
• Complex Event Processing
• Procesamiento de Eventos Complejos
• Collaborative Internet of Things
• Internet de las Cosas Colaborativo
• Taxonomy
• Taxonomía

Statement dc:rights

• Attribution-NonCommercial-NoDerivatives 4.0 Internacional

Licence dc:rights.uri

• http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/10498/39860

OAI identifier oai:identifier

oai:rodin.uca.es:10498/39860