一种基于ARMA模型的目标极点提取新方法

针对传统极点提取算法需要确定目标后时响应起点的不足,本文提出了一种提取目标极点的新方法,该方法使用ARMA模型逼近整个目标响应,并利用全局最小二乘技术求解模型参数获取目标极点。基于实测细导线和有限长柱体散射场数据的处理结果表明,本文新方法较之KT法和矩阵束法具有更高的极点提取精度。

A New ARMA-Model Based Method for Target Pole Extraction

In traditional pole extraction methods, there exists the shortcoming that the beginning of the late-time response must be determined before the processing can go forward. In this paper, a new pole extraction method is presented. Without the determination of the late-time beginning of the target transient response, the proposed method assumes an ARMA representation to approximate the complete target transient response, a total least square method is applied to the calculation of ARMA model parameters. Afterward,target' s poles are extracted from the model parameters. The proposed method is validated using a set of measured scattering data of a thin wire and a finite cylinder. The processing results show that the proposed method is more precise than the KT method and matrix pencil method in pole extraction.