Adaptive cooperative spectrum sensing method of enhanced covariance residual matrix

The invention relates to a self-adaptive cooperative spectrum sensing method of an enhanced covariance residual matrix. The self-adaptive cooperative spectrum sensing method is mainly completed through the following steps: step 1, a stochastic resonance system parameter optimization method: solving an optimal stochastic resonance system parameter by using an artificial fish swarm algorithm introduced in a background technology and a new optimization function; 2, a two-channel feature extraction step of an enhanced signal: enhancing the received signal by using a stochastic resonance system with set parameters, and extracting I and Q paths of signals by using orthogonal demodulation introduced in the background technology, and 3, a multi-user data fusion method: calculating a covariance matrix of the I and Q paths of signals of each SU, according to the method, the covariance residual matrix is constructed through stochastic resonance signal preprocessing and matrix cancellation, noise interference is reduced