Learning based Cooperative Spectrum Sensing for Primary User Detection in Cognitive Radio Networks

In Cognitive Radio Networks (CRN) spectrum sensing plays an important role in achieving spectrum utilization fast and accurately. Due to interference, power levels and hidden terminal problem, it becomes challenging to detect the presence of primary users accurately with better spectrum efficiency. Thus detection of primary users has become an important research problem in cognitive radio network. In this paper proposed a learning methods to detect the presence of primary user with high accuracy. The proposed classifiers has been trained using the extracted features to detect PU’s signal in low SNR condition. The Support vector machine (SVM), wavelet transforms, K-Nearest-Neighbour (KNN) and Reinforcement Learning-based classification techniques are implemented for Cooperative Spectrum Sensing (CSS). This approach is based on training learning models with energy vectors in presence and absence of primary users. The results provides the analysis of the learning techniques in accordance with Receiver Operating Characteristics (ROC) and shown the finest learning model for accurate primary user detection.