Fast Target Localization Method for FMCW MIMO Radar

The conventional two-dimensional multiple signal classification (2D-MUSIC) algorithm for frequency-modulated continuous wave (FMCW) multiple-input multiple-output (MIMO) radar target localization is lack of real-time. In this paper, a forward-backward spatial smoothing (SS) based linear prediction-orthogonal propagator method (LP-OPM) is proposed for target localization. Direction-of-arrival (DOA) and range are estimated independently by SS-LP-OPM algorithm, and then the parameters are matched by least squares (LS). Among them, SS technique improves the Toeplitz and Hermitian characteristics of the covariance matrix of the received data, thereby improving the accuracy of using LP to reconstruct the diagonal elements of the covariance, and reducing the impact of noise. At the same time, because OPM neither needs to increase the data dimension or calculate eigenvalue decomposition, the real-time localization is guaranteed. The superiority of the proposed method is demonstrated by simulation and experiment results.