Abstract: This paper presents a new fractional lower order moment-based 2-D direction finding method-robust covariation DOA matrix method in the presence of impulsive noise which can be modeled as a symmetric alpha-stable(SaS) process. The method extends signal models and application situation of original DOA matrix algorithm by exploiting the features of impulsive noise and SaS processes. The approach is robust against additive impulsive noise. It enhances universality and robustness of the algorithm, and remedies the shortcoming of the traditional subspace techniques employing both second-order and higher order moments, which cannot be applied in impulsive noise environments. The simulation results show the feasibility and effectiveness of the new algorithm.