Spectrum Sensing Using SVM Based Energy Detection in Cognitive Radio Systems

Cognitive Radio (CR) is an intelligent wireless communication system, capable of improving the spectral environment utilization. Spectrum sensing is one of the most important phases in the cognitive radio cycle. This operation consist of detecting the presence of signals in certain frequency bands. This paper discusses about energy spectrum sensing in cognitive radio systems, especially about primary unit (PU) signal energy spectrum sensing in cognitive radio systems using SVM learning. The SVM model is used to classify the presence of signals based on energy and standard deviation dataset. The results show that the proposed SVM spectrum sensing method is able to well recognize the presence of signals from primary units at several SNR levels and differentiate them from noise. The classification accuracy level (testing) reaches more than 60% at an SNR of 0 dB and more than 80% or maximum at an SNR level of 10 dB.