A novel refinement technique for 2-D PPS parameter estimation

This paper considers refinement in parameter estimation of 2-D polynomial phase signals (PPSs). For this purpose, a method based on phase unwrapping, filtering and polynomial fitting is developed. In the method, 2-D parameter estimation is reduced to 1-D refinement procedure, thus significantly gaining in calculation efficiency. As initial estimates, the proposed method takes the estimates provided by any 2-D parameter estimator. In this paper, we use initial estimates provided by the standard Francos–Friedlander algorithm. Numerical simulations performed on third-order 2-D PPSs show that mean squared error of the proposed estimator reaches the Cramér–Rao lower bound above the signal-to-noise ratio threshold.