Machine Learning for Improved Resource Block Detection in 4G LTE Cognitive Radio Networks

Accurate resource block detection through spectrum sensing is vital for cognitive radios in order to identify the free spaces in the licensed spectrum. The emerging field of machine learning is frequently being used to enhance the performance of spectrum sensing. Therefore, in this paper, a detailed study has been presented regarding the performance improvement of spectrum sensing through machine learning algorithms targeting the 4G LTE resource block detection. A novel spectrum sensing scheme has been proposed which achieves optimal performance at high as well as low Signal-to-Noise Ratio (SNR) conditions. The proposed scheme combines energy detection for high SNR conditions and matched filter detection with Naive Bayes algorithm for low SNR conditions.