基于RPCA的小孔径垂直阵辐射噪声测量方法

The performance of radiated noise measurement by small-aperture vertical arrays is poor under a low signal-to-noise ratio (SNR) because the array gain is so small. To solve this problem, we propose a radiated noise measurement method based on robust principal component analysis (RPCA) that uses a small-aperture vertical array under low SNR conditions. This method decomposes the data covariance matrix into a low-rank received-signal covariance matrix and a sparse noise covariance matrix by RPCA when the array-received environment noise is mainly uncorrelated. The received-signal covariance matrix is then used to estimate the source level of the radiated signal, which reduces the effect of ambient noise. The numerical simulation results show that the effect of uncorrelated noise can be removed entirely by RPCA when the number of snapshots is sufficiently large, and even if the number of snapshots is small, some of the environment noise can be eliminated. As such, the performance of the proposed measurement method is better than that based on the direct data covariance matrix.