Iterative Prony Spectral Estimation Algorithm for Low SNR Signals

Abstract

In this paper we propose a high resolution spectral estimation algorithm for sinusoidal signals corrupted by white noise. This algorithm could estimate the frequencies of measured signals more accurately than IFA⁹⁾ can do, under the low SN ratio.
For Hildebrand's Prony method, the AR spectral estimation procedure is extended. The proposed algorithm employs the extended Prony method iteratively and then designs a prefilter which has parallel structure by the first order filter with complex coefficients. As a result, the SN ratio of measured signals is increased and the estimation accuracy could be improved.