Aspectos de segurança, persistência e nomeação em agentes de recursos para Internet das coisas

• Autor(a) principal: Potter, Henrique Alberto Brittes
• Data de Publicação: 2015
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/7708
• topic: Internet of things; Resource agents; Middleware; Security; Data models; Context aware applications; Ontology; Internet das coisas; Agentes de recurso; Middleware; Segurança; Modelos de dados; Aplicações cientes de contexto; Ontologia; Internet; Internet - Medidas de segurança; Modelagem de dados; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO

The Internet of Things (IoT) needs the integration of services and devices with heterogeneous communication protocols. To this end a common infrastructure with standard data models and protocols have to be established in order to enable different devices to talk to each other. Context aware solutions have solved some of the IoT challenges, providing mechanisms that bind sensor data to a context descriptive data model. Nevertheless, the IoT needs to go beyond these solutions since it also have to address security, privacy and scalability issues. The integration of all these technologies will compose the necessary infrastructure that will enable the IoT. However, standardized protocols and data models will not be enough. A middleware support will be necessary to process data, control resources access among other needed features. Nonetheless, it must be considered that each device may need a specific API, a centralized solution with an API for each device will pose unnecessary maintenance challenges and may not scale; therefore, the integration of new devices needs to be modular. This can be delegated software agents, which are responsible for wrapping each specific device interface and communication protocol. This work presents the ContQuest, a modular framework in which these software agents are named Resource Agents. This framework aims to ease the integration of new resources and the development of IoT and context aware applications. It defines a service architecture, data models and communication protocols. The ContQuest also includes consistent solutions for security, data persistence and access control to its services and Resource Agents. The ContQuest uses OWL to describe resources, which also enables the usage of the data semantics and reasoning. A prototype of this framework was developed and validated with the integration of three physical devices: an Arduino board with sensors, a RFID card reader and the IRIS sensor network. The impact of the security system was analyzed with a performance evaluation on the prototype. The results of this validation and performance evaluation were satisfactory. It is also discussed the origins of the Internet of Things, the relationship to context aware solutions, IoT security challenges and learned lessons from previous works