A computationally efficient and interference tolerant nonparametric algorithm for LTI system identification based on higher order cyclostationarity (Q2734319)

A new nonparametric algorithm for the identification of linear time-invariant systems is proposed. It exploits the higher order cyclostationarity properties of the input and output signals and hence, under the mild assumption that the useful signal and the disturbance term do not share the same cycle frequency, is inherently unaffected by noise and interference. Specifically the proposed method regenerates the hidden periodicities of the input signal by exploiting the cyclic correlations of the input and output signals with a nonlinear transform. The asymptotic performance analysis has shown that the proposed algorithm provides asymptotically unbiased and consistent estimates of the system transfer function. Simulation results are presented to illustrate the effectiveness of the new algorithm and its superiority in terms of computational complexity and performance.