A new Levinson-Durbin based 2-D AR model parameter estimation method

作者：Zeinali Mahdi; Shafiee Masoud^*

来源：Multidimensional Systems and Signal Processing, 2016, 27(2): 341-357.

DOI：10.1007/s11045-014-0305-4

• Quarter-plane autoregressive model
• Parameter estimation
• Two dimensional series
• Levinson-Durbin algorithm

This paper presents a new method for estimating the parameters of quarterplane two dimensional (2-D) autoregressive model based on the Levinson-Durbin algorithm. To achieve this aim, one-dimensional formulations related to Levinson-Durbin algorithm are extended to 2-D case. Online parameter estimation, capability of parameters variation detection, estimation improvement by using new data and less computational requirement are the significant advantages of the proposed method. Because of not involving complex and time consuming matrix computations, the presented method is computationally efficient. Numerical simulations are presented to show the efficiency of the proposed approach.

• 出版日期2016-4