A self-managing IoT-Edge-Cloud architecture for improved robustness in environmental monitoring

The new distributed paradigm Edge-Cloud Continuum (ECC) is fundamental to provide a continuum of computing from Edge to Cloud and create more dependable Cloud-powered IoT applications. In line with this computing paradigm, this paper presents a self-managing IoT-Edge-Cloud computing architecture for enhancing robustness in environmental monitoring. In addition to the sensor data transmission to a cloud server node for late processing, a semi-decentralized self-adaptation architecture with feedback control components running on edge nodes (gateways) is adopted to monitor the end devices and make decisions in the edge. The proposed control architecture was implemented as part of SEIROB, an IoT-Edge-Cloud exemplar for air quality monitoring, using the Python programming language and the open-source ChirpStack LoRaWAN network server stack. In this paper, we share the architecture design of SEIROB, report our experiments about its effectiveness in achieving some robustness scenarios, and discuss faced challenges and lessons learned in using a ready-to-use IoT infrastructure (like the ChirpStack platform) for ECC.