基于空频分解信号子空间向量的时间反演成像

Time Reversal Imaging Algorithm Based on Signal-subspace Vectors from the Spatial-frequency Decomposition

论文提出基于空频分解信号子空间向量的时间反演成像新方法。利用天线阵列采集的散射场数据建立空频矩阵,奇异值分解该矩阵得到信号子空间向量,以此实现对目标的选择性成像。基于完全散射场数据的成像伪谱包含多个子向量贡献,相当于多次成像叠加,具有统计特性。新方法既避免了传统的空空分解时间反演算法产生的随机相位的影响,又具有较好的抗干扰性能,即使叠加信噪比10 d B的高斯白噪声,也能实现对多个目标的准确成像。

Basing on the signal-subspace vectors from the spatial-frequency decomposition, a novel time-reversal imaging algorithm is proposed. Using the backscattered data recorded by the antenna array, a spatial-frequency multistatic matrix is set up. Singular value decomposition is applied to the matrix to obtain the signal-subspace vectors, which are employed to focus the targets imaging selectively. The imaging pseudo-spectrum based on the full backscattered data includes the contributions of multiple sub-vectors and can be viewed as the superposition of multiple images. The algorithm is statistically stable. The random phases, generated by the conventional time-reversal imaging method based on the spatial-spatial decomposition, do not arise in the algorithm. It has excellent capability to resist the noise interference and can accurately focus the multi-targets even when noise with 10 dB SNR is added to the measured data.