Edge-Cloud Intelligence in Self-Diagnostic of Land Mobile Radio Systems

IIoT sensors are usually deployed on a massive scale with stringent scalability, modularity, and interoperability requirements. It is indisputable that they produce a large amount of high-speed and heterogeneous data streams that pose many challenges to perform management, processing, and analytical tasks. This paper proposes an integrated edge-cloud continuum platform that can harvest IIoT data streams from a variety of sensors deployed at a remote RF site; and can harmonize different machine learning models for diagnosing problems that enhance infrastructure monitoring and long-term structural resilience. A real-world experiment was carried out to evaluate the proposed platform for supporting a self-diagnostic process for intelligent maintenance of Land Mobile Radio (LMR) infrastructures.