基于图信号处理的大口径阵面测量数据恢复

针对热载荷和器件老化等因素造成空间大口径阵列上的测量传感器失效问题,提出大口径阵面测量数据恢复方法。根据图信号模型得到测量数据的图拉普拉斯矩阵,并基于矩阵填充理论建立了数据恢复模型,采用交替迭代方法求解对应的正则化优化问题。利用阵列的图结构特性,能够在存在单元失效和非均匀误差情况下高精度恢复测量数据。仿真验证了所提方法的有效性以及在失效率较大情况下的稳健性。

Large aperture array measured data recovery method based on graph signal

The thermal load and the element aging can lead to the failure of the measurement sensors in the large aperture array. This paper proposes a method for recovering measurement data. We first obtain the graph signal model and the graph Laplacian matrix. Then the data recovery model is established based on the matrix completion method. Finally, the regularization optimization problem is solved by alternating iteration. The proposed method can make use of the characteristics of the array graph structure and reach a high precision for the restored results in case of element failure and non uniform errors. The simulation verifies the effectiveness and the robustness in the condition of a large failure rate.