Abstract:

Both the receiver and transmitter arraies are divided into two subarraies, and a bistatic multiple-input multiple-out (MIMO) radar array model where both the transmitter and receiver are non-uniform linear array is established. And two new joint DOAs and DODs estimation algorithms are presented for it, including the spectral search based ESPRIT and the polynomial rooting-based ESPRIT. Moreover, the maximum likelihood method is used to solve the pairing problem.The strongpoint is that the angles can be validly estimated without requiring to exploiting the transmitted and received rotation invariant matrix, which extends the ESPRIT to a much more general class array geometries. The simulation results show that the polynomial rooting-based ESPRIT can perform with higher resolution than the spectral searchbased ESPRIT in the low SNR, and both of them are close to the Cramer-Rao bound in the high SNR. Furthermore, the polynomial rooting-based ESPRIT can avoid the spectral peaking searching, which has less computational complexity and more advantages in practice.