与最大化信干噪比波束相比,最小均方误差(minimum mean squared error,MMSE)波束能够获得更好的信号波形估计,因此非常适合超宽带穿墙雷达目标成像。通过分析MMSE最优权矢量中标量因子和用于旁瓣抑制的相干因子(coherence fac-tor,CF)之间的关系,建立MMSE算法与CF加权最小方差无失真响应(minimum variance distortionless response,MVDR)算法之间的统一形式,提出尺度MMSE算法。考虑到墙体介质的非均匀性、天线定位误差等的影响,给出了最优权矢量中未知参数的稳健估计方法。最后,利用时域有限差分(finite difference time domain,FDTD)数值仿真和实验数据分析了不同尺度的成像性能。结果表明尺度越大,成像质量越好,目标杂波比(target-clutter-ratio,TCR)提高约25 dB。
Compared with maximizing the signal-to-interference-pulse-noise ratio beamforming,minimum mean squared error(MMSE) beamforming could result in better signal estimation;therefore,it is suitable for ultrawideband through-the-wall imaging.Through analyzing the relation between the scalar factor of MMSE beamformer weights and the coherence factor(CF) used for sidelobe suppression,the connection between the minimum variance distortionless response(MVDR) beamforming algorithm combined with CF and MMSE beamfoming algorithm is found,and a scaled MMSE beamforming algorithm is proposed.Considering the effect of the wall medium inhomogeneity,antenna position error and etc,the robust estimation methods of unknown parameters are presented.FDTD numerical simulation data and experimentally measured data are used to demonstrate the effectiveness of the proposed beamformer in though-the-wall imaging.The results show that the larger the scale coefficient,the better the image quality;and this algorithm improves the target-clutter-ratio(TCR) by about 25dB.