Frequency estimation by correlation based two-dimensional Prony-type method

In this paper, an extended two-dimensional (2-D) Prony type method is presented to extract frequencies from observed data contaminated by the 2-D additive noise. The additive noise is generated by 2-D moving average (MA) model driven by non-Gaussian white noise. This method involves two parts, each utilizing a 1-D singular value decomposition-based technique, and is capable of locating frequencies anywhere in the 2-D frequency plane. This procedure also has the advantage not requiring a pairing of x and y-poles. Simulations show that the algorithm can restraint the additive noise of 2-D MA model effectively. The performance of the proposed algorithm is also illustrated by Monte-Carlo simulations and compared with other algorithms.