Implementation of an Internet of Things Device Management Interface

摘要

The Internet is growing from connecting only computers and mobile devices to connecting also objects in the real life to the Internet, to create an Internet of Things (IoT). Interconnected Internet of Things unveils the environmental data from these objects to the use of complex applications and systems in the Internet and the cloud, which is supposed to make the boundaries between the real world and the digital world transparent. However, these devices will mainly be resource constrained, simple, sleepy devices such as sensors and the number of devices connected to the net- work is expected to be very high. Hence, these two factors introduce the problem of device management in IoT networks.The thesis proposes the design and implementation of a fundamental device management interface for IoT networks and devices, combining IoT-specific features and protocols such as CoAP, LWM2M and Publish/Subscribe with the existing Web frameworks and protocols such HTTP and WebSocket. Real-time management and monitoring of large-scale devices is one of the IoT-specific core features of the inter- face along with other management features. The interface analyzes the integration of additional features such as anomaly detection in IoT device data and error reporting mechanisms. Moreover, the management interface is designed as a standalone application over the existing Capillary Networks architecture, which targets at providing connectivity for resource constrained devices and optimizing IoT devices with cloud instances. Hence, the management interface extensively uses the features and entities provided by the Capillary Networks via large set of REST APIs.The design of the interface focuses on the IoT-specific problems of device management, which structures the implementation accordingly. The implementation of the interface is evaluated at the end of the thesis with stress tests and comparison with initial requirements. The evaluation is then followed by possible future work to enhance the interface performance and extendibility to future IoT networks.