Cloud-based context-aware spectrum availability monitoring and prediction using crowd-sensing

Abstract

The scarcity of radio-frequency bands, due to its fixed allocation, is an emerging problem in wireless communications. Cognitive radio (CR) is a new paradigm which suggests reusing the frequency bands for unlicensed user at the time of licensed users’ inactivity. Therefore, unlicensed users must perform spectrum sensing to find the available spectrum opportunities. Cooperative spectrum sensing (CSS) is a method where unlicensed users individually sense and upload their sensing data to a fusion center. Moreover, crowd-sensing methods could be used by mobile users to provide more sensing data from various locations for the sake of improving the achieved decisions on spectrum status. Providing spectrum data from various sources makes the spectrum monitoring and management more complex. This chapter proposes a novel mechanism that (1) uses cloud computing technology as a well-suited platform for storing, processing such big data, and providing monitoring service in order to be used by, e.g., CR nodes; (2) considers the impact of contextual parameters such as location, time, building complexity around the user, etc. on the spectrum availability decision; and (3) takes the advantages of machine learning techniques to predict the future behavior of spectrum opportunities.