论文部分内容阅读
传统的后多普勒自适应处理方法,如因子法(FA)和扩展因子法(EFA)虽然能大大降低自适应处理时的运算量和独立同分布样本的需求量,但由于实际中均匀训练样本数目的限制,当天线阵元数进一步增大时,FA和EFA抑制杂波和检测动目标的能力会显著恶化。针对这一问题,提出了一种空域数据重排的后多普勒自适应处理方法。该方法将多普勒滤波后的空域数据重排为一行列数相近的矩阵,空域滤波器权系数也表示成可分离的形式,从而得到一双二次代价函数,利用循环迭代的思想求解权系数。实验表明该方法具有快速收敛,所需训练样本少的优点,尤其在大阵列、小样本条件下该方法抑制杂波的性能明显优于FA和EFA。
Traditional post-Doppler adaptive processing methods, such as factor method (FA) and expansion factor method (EFA), can greatly reduce the computational complexity and the demand of independent and identically distributed samples in adaptive processing. However, When the number of antenna elements is further increased, the ability of the FA and EFA to suppress clutter and detect moving targets will be significantly deteriorated. To solve this problem, a post-Doppler adaptive processing method for spatial data rearrangement is proposed. This method rearranges the Doppler-filtered spatial data into a matrix with a similar number of rows and columns, and the weight coefficients of the spatial filter are also expressed in a separable form, so as to obtain a biquadratic cost function. The iterative method is used to solve the weighting coefficient . Experiments show that the proposed method has the advantages of fast convergence and less training samples. Especially in the case of large array and small sample, the proposed method is superior to FA and EFA in suppressing clutter.