odern IoT environments rely on large-scale, heterogeneous networks of connected devices that continuously generate data requiring real-time processing, reliable transmission, and rapid adaptation to local context. While cloud-centric IoT architectures provide substantial storage and computational capabilities, they fail to meet the strict latency, adaptability, and Quality of Service (QoS) requirements of many emerging applications. These limitations have driven the development of the Edge-to-Cloud IoT Continuum, which distributes computing, communication, and storage resources from the cloud down to the network edge, closer to the devices.
My doctoral research is situated within this paradigm and aims to enhance the performance, reliability, and scalability of the Continuum. Current orchestration and infrastructure-management solutions remain limited in their ability to support large numbers of heterogeneous IoT and edge networks, and they often overlook local contextual data as well as dynamic QoS constraints. My objective is to thoroughly analyze these approaches, identify the key bottlenecks, and propose new architectures, mechanisms, and protocol-level improvements that enable more fine-grained, reactive, and context-aware orchestration decisions.
By rethinking communication flows, improving IoT protocol integration, and leveraging local intelligence at the edge, this research seeks to reduce latency, increase reliability, and support large-scale heterogeneous IoT deployments. Ultimately, this work aims to contribute to a new generation of Edge-to-Cloud Continuum solutions that reconcile the strengths of the cloud with the real-time operational requirements of modern IoT applications.