Motion Compensation for Multiple-Input-Multiple-Output Inverse Synthetic Aperture Imaging of Automotive Targets

TL;DR

This work tackles ISAR imaging of automotive targets, which is degraded by road clutter and complex motion. It proposes a MIMO-ISAR processing pipeline with coarse translational compensation followed by fine motion compensation using three algorithms, evaluated on monostatic mmWave data from a system with $P$ transmit and $Q$ receive elements to form $P\times Q$ ISAR frames. The study provides a quantitative and qualitative comparison of entropy minimization, cross-correlation, and phase gradient autofocus, finding cross-correlation MOCOMP to yield the best overall improvement (about $36\%$) in measured data, though frame-dependent. The results highlight the practical benefits of MIMO-ISAR for automotive radar and point to the need for more robust MOCOMP strategies in real-world conditions.

Abstract

Inverse synthetic aperture radar (ISAR) images generated from single-channel automotive radar data provide critical information about the shape and size of automotive targets. However, the quality of ISAR images degrades due to road clutter and when translational and higher order rotational motions of the targets are not suitably compensated. One method to enhance the signal-to-clutter-and-noise ratio (SCNR) of the systems is to leverage the advantages of the multiple-input-multiple-output (MIMO) framework available in commercial automotive radars to generate MIMO-ISAR images. While substantial research has been devoted to motion compensation of single-channel ISAR images, the effectiveness of these methods for MIMO-ISAR has not been studied extensively. This paper analyzes the performance of three popular motion compensation techniques - entropy minimization, cross-correlation, and phase gradient autofocus - on MIMO-ISAR. The algorithms are evaluated on the measurement data collected using Texas Instruments millimeter-wave MIMO radar. The results indicate that the cross-correlation MOCOMP performs better than the other two MOCOMP algorithms in the MIMO configuration, with an overall improvement of 36%.

Figures (3)

• Figure 1: Measurement setup with TI AWR1843 radar in MIMO configuration, the camera (for ground truth information), and a mid-size car as a target.
• Figure 2: Range-Doppler ambiguity diagrams of the mid-size car taken from radar in SISO configuration where five columns of each row correspond to different time frames: 8.1s, 8.4s, 8.6s, 8.7 and 9.0s. This first row (a-e), second row (f-j), third row (k-o), and fourth row (p-t) includes the plots without MOCOMP, after coarse MOCOMP along with entropy minimization MOCOMP, PGA MOCOMP, and CCR MOCOMP, respectively. The range along the vertical axis spans from 0 to 34.4m with a range resolution of 0.13m, while the Doppler index spans along the horizontal axis spans from 1 to 128.
• Figure 3: Range-Doppler ambiguity diagrams of the mid-size car taken from radar in MIMO configuration where five columns of each row correspond to different time frames: 8.1s, 8.4s, 8.6s, 8.7 and 9.0s. This first row (a-e), second row (f-j), third row (k-o), and fourth row (p-t) includes the plots without MOCOMP, after coarse MOCOMP along with entropy minimization MOCOMP, PGA MOCOMP, and CCR MOCOMP, respectively. The range along the vertical axis spans from 0 to 34.4m with a range resolution of 0.13m, while the Doppler index spans along the horizontal axis spans from 1 to 128.