An efficient and fast algorithm for estimating the frequencies of 2-D superimposed exponential signals in presence of zero-mean multiplicative and additive noise

作者：Bian Jiawen; Li Hongwei^*; Peng Huiming

来源：Journal of Statistical Planning and Inference, 2011, 141(3): 1277-1289.

DOI：10.1016/j.jspi.2010.09.028

• 2-D superimposed exponential signals
• Three step iterative algorithm
• Least squares estimators
• Consistent estimators
• Texture

In this paper, a computationally efficient algorithm is proposed for estimating the parameters of two-dimensional (2-D) superimposed exponential signals in presence of independently and identically distributed (i.i.d.) zero-mean multiplicative and additive noise. It is observed that the estimator is consistent and works quite well in terms of biases and mean squared errors. Moreover, the algorithm is efficient when multiple 2-D frequencies pairs share a same 1-D frequency component and the estimators attain the same convergence rate with the least squares estimator (LSE) in presence of additive noise. Finally, it is observed that the algorithm can be used to estimate the frequencies of the evanescent component of texture accurately.

• 出版日期2011-3
• 单位中国地质大学（北京）