联合平移不变子空间的压缩采样及应用

Compressed Sampling of Signals in Union of Shift-Invariant Subspaces and Application

针对有效核函数(active kernel function)未知的联合平移不变子空间(Union of Shift-InvariantSubspaces,USI),提出了一种压缩采样模型,基于稀疏重构理论,该采样模型能够有效降低信号的采样率。首先建立一个多脉冲雷达回波信号模型,在信号的延时-多普勒平面上对延时轴离散化,将回波信号表示为USI信号;然后在根据构建的压缩采样模型降低信号采样率的同时,利用稀疏贝叶斯学习和ESPRIT算法由信号样本值估计出雷达回波信号的延时、多普勒频移和反射系数等参数;最后仿真验证了研究结论的有效性。

For the signals in union of shift-invariant subspaces （USI） when the active kernel functions are unknown, a concrete compressed sampling scheme is proposed which can reduce the sampling rate effectively based on the sparse reconstruction. A signal model of multiple-pulse radar echo signal is estabilished firstly, the echo signal corresponds to a signal in union of shift-invariant subspaces by discretizing the delay of delay-doppler space. Furthermore, the parameters of echo signal are estimat- ed from the samples by sparse Bayesian learning and ESPRIT algorithm. Finally, simulation are car- ried out to prove the validity of the research result.