Robust two dimensional spectral estimation based on AR model excited by a t-distribution process

A new robust two dimensional (2-D) spectral estimation method based on an AR model is proposed. The optimal coefficient is selected by assuming that the excitation signal is a t-distribution t(α) with α degrees of freedom. When α=∞, we get the conventional least square (L₂) method. Thus, the proposed method can be regarded as a generalization of the L₂ method. Simulation results show that the obtained estimates using the proposed method with small α are more efficient, the standard deviation (SD) of the estimation results are smaller, and more accurate than that with large α. The proposed estimator with small α is more efficient and more accurate than the recursive method based on Huber´s (1981) M-estimate