Spatio-temporal spectrum sensing in cognitive radio networks using Beamformer-Aided SVM algorithms

This paper addresses the problem of spectrum sensing in multi-antenna cognitive radio system using support vector machine (SVM) algorithms. First, we formulated the spectrum sensing problem under multiple primary users scenarios as a multiple state signal detection problem. Next, we propose a novel, beamformer aided feature realization strategy for enhancing the capability of the SVM for signal classification under both single and multiple primary users conditions. Then, we investigate the error correcting output codes (ECOC) based multi-class SVM algorithms and provide a multiple independent model (MIM) alternative for solving the multiple state spectrum sensing problem. The performance of the proposed detectors is quantified in terms of probability of detection, probability of false alarm, receiver operating characteristics (ROC), area under ROC curves (AuC) and overall classification accuracy. Simulation results show that the proposed detectors are robust to both temporal and joint spatio-temporal detection of spectrum holes in cognitive radio networks