A modular spectrum sensing node for Resources-Oriented Radio Monitoring

The existing radio monitoring practices suffer from two critical limitations: 1) current practices have been limited to isolated measuring and are unable to integrate data from different spectrum sensors; 2) these practices still require human operators to interpret signals, which prevent automated and intelligent processing. To address these limitations, we present a novel design of a spectrum sensing node based on the Representational State Transfer (REST) architecture. The spectrum sensing node exploits the REST architecture to integrate resources (data) into the web and to make it effortless to collect and share resources by exposing resources as service-oriented Web APIs. In addition, the spectrum sensing node introduces standardization and automation to radio monitoring. We outline the procedures of spectrum analysis and signal analysis for radio monitoring. Furthermore, to automate spectrum sensing node functionalities, we implement spectrum analysis and signal analysis using machine learning methods, which reliably extract and learn intrinsic features of complex data. The spectrum sensing node has been packaged and deployed in Honghe Hani and Yi Autonomous Prefecture of China for radio spectrum management. We describe and discuss our implemented prototype of the spectrum sensing node to demonstrate its practical advantages.