Localization performance of coherent MIMO radar systems subject to phase synchronization errors

Localization advantages of coherent Multiple-Input Multiple-Output (MIMO) radar systems are reliant on full phase synchronization among all participating radars. Phase synchronization errors are practically inevitable, reflecting on the system localization performance. In this paper quantitative tools to asses this effect are provided. The lower bound on the mean-square error (MSE) is set by the hybrid Cramer-Rao bound (HCRB) for the joint estimation of the target location and the phase synchronization offsets at the radars. The latter are modeled as random unknown. The HCRB of the unknown target location is shown to be equal to the CRB derived by discarding the random phase synchronization errors through marginalization, i.e., treating them as nuisance parameters. Therefore, the HCRB closed-form expression provides an asymptotically tight bound on target location estimation MSE at high signal-to-noise ratio (SNR). The bound is shown to follow the CRB in the absence of phase errors up to a threshold point, determined by the synchronization error variance, the SNR, and the number of mismatched transmitting and receiving sensors. Beyond this point, the HCRB asymptotically reaches a lower limit, proportional to the synchronization errors variance and independent of the SNR. The value of the threshold point and lower limit are determined for symmetrical radar deployments.