Abstract: A new approach to high resolution estimation of four-dimensional (4-D) angle of multiple narrowband far-field signals is presented for bistatic multiple-input multiple-output (MIMO) radar with double L-shaped arrays. The spatial noise can be eliminated and the angle estimation performance is improved by constructing the matched array data and making full use of the cross-correlation matrix. The proposed method decouples the two-dimensional (2-D) direction-of-arrival (DOA) or 2-D direction-of-departure (DOD) estimation problem into two successive one-dimensional (1-D) estimation problems, and only requires eigenvalue decomposition of lower dimensional matrix, which can reduce the computational burden. The received or transmitted elevation angles are estimated using the eigenvalues, and the corresponding received or transmitted azimuth angles are estimated through the virtual array response matrix constructed by the corresponding eigenvectors. Numerically simulated results show that the proposed method can accurately estimate the 4-D angle of multiple spatial targets with lower computational complexity, and 2-D DOA or 2-D DOD could be paired automatically. The proposed method has better angle estimation performance in the situation of low signal-to-noise ratio and short data length. Key

Key words: ordnancescienceandtechnology, MIMOradar, 4-Dangle, cross-correlationmatrix, doubleL-shapedarray