IAA Spectral Estimation in the Selective Range

Recently, the iterative adaptive approach (IAA) has been proposed to be a high-resolution spectrum estimator. Its main idea is reformulating the nonlinear frequency estimation problem as a linear one, with parameters being updated iteratively according to weighted least squares. Since the derivation is based on grid searching in a fixed frequency range \([0,2\pi )\), the accurate of the IAA is limited by the number of grid. In this paper, we proposing two generalized versions of IAA, which can work well in a flexible frequency range, so that the performance can be improved in the same grid points. Simulation results are included to demonstrate the superior of our proposed methods.