| MIMO radar is a new type of radar that has been proposed recently. It transmits orthogonal signals via different antennas. This waveform diversity provides MIMO radar improved performance in DOA estimation and target identification compared with conventional radars.In MIMO radar, a filled array of virtual elements can be formed from relative few actual elements. This dissertation explains the virtual element concept and extends it to the synthesis of various especial antenna array configurations. Then the DOF in MIMO radar system and the maximum number of targets that can be uniquely identified by MIMO radar is analyzed. Based on the introduction of MIMO radar concept, a model for monostatic multi-pulse integrated MIMO radar is established. Then some commonly used DOA estimation algorithms are introduced and applied to MIMO radar system. It is shown that Capon and MUSIC algorithms are applicable to MIMO radars with arbitrary antenna arrays,but suffer from high computational complexity. While ESPRIT algorithm offers much lower computational complexity for extracting directions of targets from the generalized eigenvalues directly. Considering the limitation of ESPRIT algorithm that it can only be used in arrays with spatial shift configuration, the implementations of the ESPRIT algorithm in MIMO radar with ULA and L shaped array are derived respectively. Finally, some numerical simulations are presented to demonstrate the validity and the feasibility of the algorithms considered herein. |
PDF Full Text Request