Abstract:
The continuous growth of demand experienced
by wireless networks creates a spectrum availability challenge. Cognitive radio (CR) is a promising solution capable
of overcoming spectrum scarcity. It is an intelligent radio
technology that may be programmed and dynamically configured to avoid interference and congestion in cognitive
radio networks (CRN). Spectrum sensing (SS) is a cognitive radio life cycle task aiming to detect spectrum holes.
A number of innovative approaches are devised to monitor
the spectrum and to determine when these holes are present.
The purpose of this survey is to investigate some of these
schemes which are constructed based on machine learning
concepts and principles. In addition, this review aims to
present a general classification of these machine learningbased schemes
