Aiming at the problem of soft fault diagnosis of nonlinear analog circuits, a novel approach of fault signature extraction is proposed based on the 2nd-order distribution—Wigner Ville distribution(WVD) of the Volterra kernel. Firstly, the Volterra kernel of the circuit under test(CUT) is calculated. Then the Volterra kernel of the CUT is transformed based on the WVD and the WVD functions are obtained. In the WVD functions, the fault signatures are extracted and the soft fault diagnosis of the nonlinear analog circuit is achieved. The simulation results show that the proposed method can solve the fault aliasing problem effectively and improve the capability of detecting and locating fault components.