Spectrum Hole Detection for Cognitive Radio through Energy Detection using Random Forest

Abstract

The growth of wireless data is the major driving force for an exponential increase in wireless communication. Cognitive Radio is one of the emerging wireless technologies that can be used for smart utility networks. Optimum utilization of the wireless spectrum is the objective of Cognitive Radio. Finding a spectrum hole through intelligent means is essential for the success of Cognitive Radio. Dynamic spectrum allocation is also an efficient technique for spectrum allocation. It will lead to a better spectrum utilization. In this paper, some of the machine learning techniques are used to find a frequency range for dynamic spectrum allocation. Different machine learning techniques such as Logistic Regression, Support Vector Machine, Adaboost Classifier, and Random Forests were used to find spectrum holes in skewed data. Random Forest outperforms all the other models with an accuracy of 91% for determining the spectrum bandwidth (i.e. hole) for Cognitive Radio applications.