A cooperative spectrum sensing algorithm based on principal component analysis and K-medoids clustering
Chenhao Sun, Yonghua Wang, Pin Wan, Yiqi Du · 2018
In order to decrease the effects of noise on the signal feature extraction and improve the spectrum sensing performance of cognitive radio system. This paper proposes an improved cooperative spectrum sensing (CSS) method based on the principal component analysis (PCA) and K-medoids clustering. Firstly, principal component analysis extracts signal principal components by applying multiple antenna systems, then the corresponding signal features are extracted through the signal principal component matrix. Finally, these features are classified using the K-medoids clustering algorithm. The experimental simulations are performed with different features and K-medoids clustering algorithms. The simulation results show that the proposed method has better detection performance than traditional spectrum sensing methods and can effectively improve the efficiency of spectrum sensing.