2-D HARMONIC RETRIEVAL USING HIGHER-ORDER STATISTICS

We consider the problem of estimating the harmonics of a noisy 2-D sig nal. The observed data is modeled as a 2-D sinusoidal signal, with eit her random or deterministic phases, plus additive Gaussian noise of un known covariance. Our method utilizes recently defined higher-order st atistics, referred to as ''mixed-cumulants'', which permit a formulati on that is applicable to both the random and deterministic case. In pa rticular, we first estimate the frequencies in each dimension using an overdetermined Yule-Walker ''type'' approach. Then, the 1-D frequenci es are paired using a matching criterion. To support our theory, we ex amine the performance of the proposed method via simulations.