A conceptual model and prototype of Cognitive Radio Cloud Networks in TV White Spaces

Abstract

A Cognitive Radio Cloud Network (CRCN) model is proposed for wireless communications in TV White Spaces (TVWS). Making use of the flexible and vast computing capacity of the Cloud, a database and a sparse Bayesian learning (SBL) algorithm are developed for cooperative spectrum sensing (CSS) and implemented on Microsoft's Windows Azure Cloud platform. A medium access control (MAC) scheme is also prototyped for this CRCN model to collect sensing reports and access channels with Rice University's wireless access research platform (WARP). Through this CRCN prototype, important network parameters such as the mean squared errors in CSS, the time to detect the presence and/or the absence of primary users, and the channel vacating delay are measured and analyzed for the design and deployment of the future CRCN.