Optimized Extreme Learning Machine for Intelligent Spectrum Sensing in 5G systems

A two-level learned distributed networking (LDN) structure that uses existing machine learning (ML) algorithms and the novel Optimized Extreme Learning Machine (OELM) algorithm to perform intelligent spectrum sensing for 5G systems has been proposed and implemented. This novel technique uses input vectors like received signal strength indicator, the distance between Cognitive Radio users and gateways, and energy vectors to train the model. Extreme Learning Machine optimized by BAT algorithm outperforms the existing Machine Learning techniques in terms of detection accuracy, false alarm, detection probability and cross validation curves at different SNR scenarios.