MIMO radar with dense or random pattern: analysis of phase and positioning error sensitivity

Multiple Input Multiple Output (MIMO) radar allows to reduce the number of antennas maintaining performances comparable to a phased array radar. The Compressive Sensing (CS) approach is a processing technique that promises to reduce further the number of antennas. The strict control of the unwanted phase shift and of the position on each antenna is a critical condition in any MIMO array for radar application. So an open question is if CS approach makes the MIMO more o less sensitive to possible phase errors and positioning error. The aim of this paper is just a compared analysis of the effects of phase/positioning error for conventional MIMO radar (with dense pattern of antennas) and CS-MIMO (with random pattern of antennas). The results of the simulations show that both MIMOs are rather sensible to phase error and positioning error along y (the direction in front of the radar), but rather robust to positioning error along x (the direction of the synthetic aperture) and z practically. Furthermore, C-MIMO seems a bit more robust to phase error then dense MIMO.