| Multiple-input multiple-output(MIMO) radar is a new radar architecture, it brought MIMO technology into the radar system, which increase the degree of freedom and improve the resolution, thus it has broad application prospects. As a combination of bistatic radar and MIMO radar, bistatic MIMO radar has the potential advantages of both bistatic rada and of MIMO radar. Those advantages include reduced space loss, covert operation, and reduced susceptibility of jamming.In radar signal processing the estimation of angle and Doppler of multi-targets is of great significance. Scholars normally estimate the angle and Doppler separately, then apply the pairing method to accomplish the estimation of targets' parameters. Considering the high complexity, the paring method is not applicable to engineering practice. Therefore bistatic MIMO radar in angles and Doppler estimation has arien widespread concern. In bistatic MIMO radar, the traditional MUSIC and Capon angle estimation method requires a two-dimensional spectral peak searching, which results in high computational complexity, and need to pair the angle and Doppler together. ESPRIT algorithm can realize the estimation of the target angle and Doppler together. But under the circumstance that both the transmit and receive array are ULAs, we can only obtain the target's information of two angles with respect to each transmitter and receiver, thus the three-dimensional target localization is impossible. The accuracy of angle estimation varies with different angles when using ULAs. The UCAs can achieve target three-dimensional localization with high accuracy. But the corresponding steering vectors of transmitter and receiver do not have rotational invariance. Therefore ESPRIT algorithm cannot be used directly.In order to solve the above problems, this thesis studies the multi-targets localization and velocity measurement problem in bistatic UCA MIMO radar, and proposes an angle and Doppler joint estimation algorithm based on Parallel factor. The algorithm formulates the received signals with parallel factor modal, then applies the COMFAC to solve the modal, meanwhile we use the geometry of the steering vectors to obtain transmit and receive angles and Doppler of the targets, furthermore we obtain the location and velocity of the targets. The algorithm we proposed does not require spectral peak search, can accomplish angle and Doppler estimation jointly with parameters automatically paired.Using computer simulation we analyzes and compares the performance of the algorithm we proposed, since the Cramer-Rao bound is a minimum variance for the parameter estimation in theory, reflecting the best performance that the parameter estimation algorithm can reach. Therefore, in this article we derive the Cramer-Rao bound of angles, Doppler, and localization accuracy of multi-targets in bistatic UCA MIMO radar, as a comparison with the parameter estimation error of the proposed algorithm. Later in this thesis, under different SNRs, different sizes of the array, we evaluate the algorithm we proposed. Through the analysis of simulation results, the algorithm we proposed can jointly estimate the angle and Doppler, distinguish multi-targets, avoid the space synchronization problems in traditional bistatic radar, conquer the high computational complexity, eliminate spectral peak search. It also automatically pairs parameters of multiple parameters in the presence of multiple targets and improves the accuracy of localization and velocity measurement. |
PDF Full Text Request