Abstract:
The last decade has seen the rise of a set of information technologies, ranging from the expansion of the traditional Internet to interconnect smart objects, big data, cloud compu ting to the blockchain. As these technologies have evolved, they have converged to rise a new paradigm known as the Internet of Things. In its cloud-centric architecture, the Internet of Things links billions of objects and users worldwide. This has motivated researchers to focus on the problem of automating actions such as connecting new objects to the network, or self-organizing underground components such as wireless sensor networks. Our thesis focus on these automation mechanisms which aim to reduce human intervention to ensure functional safety and availability over time. At various levels we have identified crucial modules, such as the leader election at the perception layer, task scheduling and load balancing in the cloud environment at the core layer, or even more complex at the application layer, the large-scale deployment of a selfmanaged platiorm for calculating the ecological footprints of companies and individuals. We carried out a state-of-theart survey, a comparative and assessment study of the most recent solutions, before proposing our own approaches. Experimentation has shown very satisfactory results, making our works to be among the most relevant ones in the current literature.
