Two Novel Spectrum Sensing Algorithms Based On Eigenvalue under Different Noise

In the field of spectrum sensing for cognitive radio networks, there exists a category of eigenvalue based algorithms by analyzing the limiting eigenvalue distribution of the sample covariance matrix. In this paper, an novel algorithm based on the ratio of the maximum eigenvalue to signal energy (MEE) is proposed under Gaussian white noisy environment. The corresponding probabilities of detection and the detector robustness in low signal to noise ratio (SNR) are derived. Moreover, an improved difference between the maximum eigenvalue and the minimum eigenvalue (DMM) algorithm in alpha-stable distribution noise is proposed. The relation between the probability of false alarm and the decision threshold is derived as well. Both the theoretical analysis and numerical simulations demonstrate that the proposed MEE algorithm can effectively overcome the noise uncertainty without the prior knowledge of the signal transmitted and has better detection performance. The improved DMM algorithm outperforms the original counterpart under alpha-stable distribution noise in terms of suppression impulse noise.