Utilization of digital polyspectral analysis to estimate transfer functions of cubically nonlinear systems with nonGaussian inputs

Spectral estimation of the nonlinear transfer functions (linear, quadratic, and cubic) of a nonlinear system with stationary nonGaussian inputs, which can be modeled by single-input/single-output third-order Volterra functionals is presented. This method is general in the sense that it does not depend on the input statistics. The practicality and the feasibility of the digital implementation of the method are tested by analyzing simulated input-output data of known cubically nonlinear systems. In addition, the notion of coherency is extended to cubically nonlinear systems and generalized for nonGaussian inputs