最小方差无失真响应（MVDR）自适应波束形成方法拥有高空间分辨率、低旁瓣级等优势。基于构造时域解析信号的时域MVDR方法TMVDR（Time-domain MVDR）有效提高了MVDR最优权向量估计的稳定性。但是对于大孔径主动声呐，由于累积快拍数过少或者阵型失配引起的最优权向量估计误差，会导致TMVDR方法输出功率失真。本文在TMVDR基础上提出一种双约束条件的时域MVDR自适应波束形成方法DCTMVDR（Double constraint Time-domain MVDR）。该方法能够有效提高最优权向量估计的稳定性，改善输出功率失真问题，并且实现在单次快拍内进行最优权向量的估计，规避MVDR需要大量快拍支持的缺陷，适用于主动声呐。数值仿真和实验数据分析验证了该方法的有效性。
MVDR adaptive beamforming algorithm has the advantages of high spatial resolution and low sidelobe level.The TMVDR algorithm based on constructing time analytic signal effectively improves the stability of the optimal weight vector estimation.But for large aperture active sonar, the output power of TMVDR algorithm will be distorted due to the estimation error of optimal weight vector caused by too few accumulated snapshots or array mismatch.Based on TMVDR, this paper proposes a Double constraint Time-domain MVDR adaptive beamforming algorithm.This algorithm can effectively improve the stability of the optimal weight vector estimation, improve the output power, and realize the optimal weight vector estimation in one signal snapshot, avoiding the defect that MVDR algorithm needs a lot of snapshots to maintain stability, and is suitable for actiove sonar.Numerical simulation and experimental data analysis verify the effectiveness of the algorithm.
2021,43(7): 127-130 收稿日期：2021-03-04
DOI：10.3404/j.issn.1672-7649.2021.07.026
分类号：TB566
作者简介：董晋(1992-),男,工程师,研究方向为主动声呐的自适应杂波抑制技术
参考文献：
[1] 何振亚. 自适应信号处理[M]. 北京:科学出版社, 2002, 21−117.
[2] KROLIK J, SWINGLER D. Multiple broadBand source location using steered covariance matrices[J]. IEEE, Acoustics Speech and Signal Processing, 1989, 37(10):1481-1494
[3] 王良, 宋志杰, 华洋. 时域解析信号的MVDR自适应波束形成方法[J]. 数据采集与处理, 2009(3):318-322
[4] J. LI, P. STOICA, WANG Z S On robust Capon beamforming and diagonal loading[J]. IEEE, Signal Processing, 2003, 51(7):1702-1715
[5] J. LI, P. STOICA robust adaptive beamforming[J]. Wiley, New York, 2005
[6] Implementation of adaptive and synthetic-aperture processing schemes in integrated active-passive sonar systems[J]. IEEE, Signal Processing, 1998, 86(2):358−376.
