Performance evaluation of cyclic polyspectrum estimators

Results of analysis and computer simulations are used to compare two methods of estimating the cyclic polyspectrum. The computational requirements of the two methods are determined and compared. A cycle leakage phenomenon is explained and it is shown that, for a given amount of data, only one of the two methods can be modified to reduce this leakage, but at the cost of increased computations. Numerical examples of measurements of fourth-order cyclic polyspectra are provided to illustrate the leakage problem.<
>