The issue of two-dimensional (2D) direction-of-departure and direction-of-arrival estimation for bistatic multiple-input multiple-output (MIMO) radar with a coprime electromagnetic vector sensor (EMVS) is addressed in this paper, and a tensor-based subspace algorithm is proposed. Firstly, the covariance measurement of the received data is arranged into a fourth-order tensor, which can maintain the multi-dimensional characteristic of the received data. Then, the higher-order singular value decomposition is followed to get an accurate signal subspace. By utilizing the uniformity of the subarrays in coprime EMVS–MIMO radar, the rotation invariant technique is adopted to achieve ambiguous elevation angle estimation. Thereafter, the unambiguous elevation angles are recovered by exploring the coprime characteristic of the subarrays. Finally, all azimuth angles are achieved by using the vector cross-product strategy. The tensor nature inherited from the array measurement is fully explored, and the coprime geometry enables EMVS–MIMO radar to achieve larger array aperture than the existing uniform linear configuration; thus, the proposed method offers better estimation performance than current state-of-the-art algorithms. Several computer simulations validate the effectiveness of the proposed algorithm.

本文研究了带有互质电磁矢量传感器（EMVS）的双基地多输入多输出（MIMO）雷达的二维（2D）出发方向和到达方向估计问题，提出了基于张量的子空间算法。首先，将接收到的数据的协方差测量安排在一个四阶张量中，该张量可以保持接收到的数据的多维特性。然后，进行高阶奇异值分解以获得准确的信号子空间。通过利用互质EMVS–MIMO雷达中子阵列的均匀性，采用旋转不变技术实现模棱两可的仰角估计。此后，通过探索子阵列的互质特性来恢复明确的仰角。最后，所有方位角都是通过使用矢量叉积策略实现的。充分探索了从阵列测量中继承的张量性质，并且互质几何使EMVS–MIMO雷达能够获得比现有均匀线性配置更大的阵列孔径。因此，与目前的最新算法相比，该方法提供了更好的估计性能。若干计算机仿真验证了所提出算法的有效性。