FIR system identification based on higher-order cyclic cumulants

This paper focuses on nonminimum phase FIR system identification using higher-order cyclic cumulants alone, and proposes two linear normal equation algorithms for parameter estimation and SVD-based order determination methods. The system unidentifiable from second-order cyclostationary statistics are shown to be identifiable using the proposed higher-order cyclic cumulants based algorithms. Simulations are given to verify the high performance of the new methods.