Abstract: For the purpose of improving the direction of arrival (DOA) and direction of departure (DOD) estimation performance for a bistatic multiple-input multiple-output (MIMO) radar by making the best use of the target temporal structures, this article investigates multi-dimensional parameter joint estimation for bistatic MIMO radar based on a 2-dimension temporal-spatial model. For slowly moving targets, a rotational invariant subspace is formed which is based on the multi-dimensional estimation of signal parameters by rotational invariance techniques (ESPRIT) by means of the rotational factors produced by the phase shifts due to the adjacent delays in the time domain and the adjacent transmitting and receiving of array elements respectively, and the parameter matching is solved according to the tripartite relationships among the matrix, the eigenvalue of the matrix and the eigenvector of the corresponding eigenvalue. Then the 3D parameter joint estimation is implemented. The method outperforms the conventional ESPRIT method which does not make use of signal temporal structure in the case of less accumulative pulses and low singal to noise ratio (SNR). Simulation results prove the validity of the method.

Key words: bistatic MIMO radar, temparal-spatial model, ESPRIT, direction of arrival, direction of departure