A fine search method for the cubic-phase function-based estimator

In this paper, we propose a refinement procedure for the estimation of polynomial-phase signal (PPS) parameters. The PPS parameters are estimated using a method based on the cubic-phase function (CPF), known to outperform the standard high-order ambiguity function (HAF) in terms of both the estimation accuracy and performance threshold. The method is referred to as the hybrid CPF-HAF (HCH). Since in the calculation of the HCH we cannot use fast Fourier transform algorithms, it is very important to decrease the number of HCH samples required for precise parameter estimation. To that end, we propose an approach based on the dichotomous search of the HCH peak. With the calculation of only several additional HCH samples, we are able to estimate parameters accurately without using time consuming oversampling or maximum-likelihood approach.