In the community of automotive millimeter wave radar, the recently developed concept of four-dimensional (4D) radar can provide high-resolution point clouds image with enhanced imaging performance. Currently, the density of point clouds for single-frame image is usually too sparse to satisfy the demands of target classification and recognition due to the limitation of Doppler and angle resolutions. To address the aforementioned issues, a novel algorithm is proposed for 4D high-resolution imagery generation of point clouds with extremely high Doppler and angle resolutions in this paper. For high Doppler resolution with high-dynamic, a novel velocity ambiguity resolution algorithm is proposed using a dual pulse repetition frequency (dual-PRF) waveform design embedded in an innovative time-division multiplexing & Doppler-division multiplexing MIMO (TDM-DDM-MIMO) framework. Meanwhile, an attractive complex-valued deep convolutional network (CV-DCN) of super-resolution direction-of-arrival (DOA) estimation is proposed only using single-frame data. To be specific, a spatial smoothing operator on array data is applied as input of the network, and a CV-DCN is designed to learn the transformation of the spatial spectrum from the end-to-end to effectively protect the spectrum extraction. Furthermore, experimental analysis is performed to confirm the effectiveness of the proposed super-resolution DOA estimation algorithm. Finally, the 4D high-resolution imagery of point clouds is obtained by experiments in the parking lot.

Forward-looking multi-input multi-output synthetic aperture radar (FL-MIMO-SAR) is the state-of-the-art technique to achieve the high-resolution image of the front area. However, the performance of FL-MIMO-SAR is primary limited by left-right Doppler ambiguity resolution. To address this issue, a novel space-time cascade imaging framework is proposed. Concretely, the Doppler-division multiplexing MIMO (DDM-MIMO) scheme is adopted to achieve the waveform orthogonality and the back-projection (BP) time-domain algorithm is applied to obtain high-resolution SAR images. Furthermore, the effect of array errors is analyzed and calibrated, which is essential for subsequent Doppler ambiguity resolution using the linearly constrained minimum variance (LCMV) beamforming. Finally, experimental analysis is performed to confirm the effectiveness of the proposal.

Accurately estimating the direction of arrival (DOA) of wideband signals with a sensor array is critical in communications, radar, and the Internet of Things. This paper proposes two single-source DOA estimation methods for wideband linear frequency modulation signals: time-delay mixing multiple signal classification (TDM-MUSIC) and enhanced self-mixing MUSIC (ESM-MUSIC). TDM-MUSIC employs time-delay mixing of the received signal to construct an equivalent single-frequency signal model, thereby enhancing estimation accuracy while maintaining reasonable computational efficiency. ESM-MUSIC improves the conventional self-mixing model by adding frequency correction steps, resulting in excellent DOA estimation performance at the expense of computational complexity. Unlike conventional methods that rely on approximate models, our methods establish more accurate equivalent models. A key advantage of our methods is that they allow flexible adjustment of the optimal sensor inter-element spacing in arrays based on the equivalent signal model rather than the actual signal model, simplifying engineering fabrication and reducing mutual coupling between sensors. The paper establishes the Cramér–Rao bounds for both proposed methods and demonstrates their superiority over existing methods through comprehensive numerical simulations. Further, the experiment using a TI-AWR2243 multi-sensor array radar system confirms that our methods are feasible for practical engineering applications.

Human activity recognition (HAR) technology using millimeter-wave radar has been applied in many fields, such as health monitoring and security. In this study, a HAR network using contrastive learning is proposed, named as CLHAR-Net, which can take good generalization. The network first uses unlabeled samples for self-supervised training to obtain deep representations of target feature, and then fine-tune the downstream classifier using labeled samples to finally achieve high-precision classification of human activity. In the experiments, compared with the traditional network architecture, CLHAR-Net has better classification performance and reliability.

Benefiting from the characteristics of low-cost, small-size, and high-resolution, the millimeter-wave (mmWave) radar has been gradually applied to automotive parking assistance. In this article, a novel algorithm of automotive synthetic aperture radar (SAR) imaging is proposed for the mapping of parking places. To deal with the motion error from the inaccurate speed of the radar platform, a parametric method of sparse Bayesian learning (SBL) is presented for well-focused and high-resolution SAR imaging. Then, a watershed-based SAR image segmentation algorithm is applied to detect the vehicles, which can indicate the locations of free parking spaces. Finally, the experimental analysis using 77-GHz automotive radar data is performed to confirm the effectiveness of the proposal.

Automotive radar is an essential part of self-driving and advanced driver assistant systems. Multi-input multi-output (MIMO) millimeter wave (MMW) radar has gained more and more attention due to its advantages of low hardware cost, high resolution and small size. Among MMW MIMO radar, time division multiple access (TDMA) is one of the most widely used modes due to its simple implementation, where the transmitting antennas are required to transmit waveforms at different time. Due to the existence of target velocity, phase errors tends to be introduced between subarray of different transmitters during the coherence MIMO processing. Meanwhile, the maximum unambiguous velocity obtained by TDMA MIMO is greatly reduced and limited, so that the phase errors cannot be completely eliminated by directly using the Doppler speed. This paper proposes two solutions to coherent error correction for TDMA MIMO automotive radar. The first solution uses part-overlap MIMO array by having at least two virtual array elements at the same position. By comparing the phase difference of the two virtual array elements in the same pulse repetition period, the phase 7error caused by the target ambiguous velocity can be effectively estimated. The second solution is to apply non-overlap MIMO array. In this case, the explicit formulation of phase errors is deducted and the objective function of phase error compensation is constructed. After performing all possible phase compensations on MIMO array, the phase errors can be estimated and corrected using the objective function. Finally, an example MIMO configuration and a simulation experiments are designed to demonstrate the proposed algorithms. The simulation results illustrate the robustness of the solutions.

Automotive radar is a class of important and necessary sensor for Advanced Driver Assistance System (ADAS) due to its recognized advantages of small size, low hardware cost, all-weather working, high-resolution and etc. However, the limitation of low angular resolution with low imaging performance can hardly satisfy the need of next-stage ADAS. The emerging 4D imaging radar (4D-radar), adopting the multi-chip cascaded multiple-input multiple-output (MIMO) technology, can achieve high resolution in the azimuth and elevation dimensions with providing high-quality three-dimensional point clouds images. In this paper, a novel algorithm is proposed by integrating the high-resolution MIMO radar point clouds imaging and processing. First, we have symmetrically studied the MIMO radar technologies, classing into three main modes, TDM-MIMO, Phase coding MIMO, DDM-MIMO. In particular, we have designed a mixed TDM-DDM-MIMO framework for point clouds imaging. Finally, the experimental analysis of the simulation is provided to confirm the effectiveness of the proposal.

Synthetic aperture radar (SAR) has been widely applied to terrain observation as an enhanced sensor for all-weather and all-day functions. In this paper, a ground-based millimeter-wave multi-channel interferometric SAR (InSAR) system design and high-resolution imaging algorithm are proposed, which is characterized by small-size, low-cost, and high-precision. Firstly, the frequency-domain imaging algorithm of chirp-scaling is adopted to realize the range cell migration correction (RCMC) and two-dimensional focusing. Then, multi-channel interferometric processing is used to obtain the angle estimation of the imaging scene. Finally, experiments using measured data are performed to confirm the effectiveness of ground-based SAR imaging for terrain observation.

Aiming at the acceleration of the maneuvering target changes uniformly, an adaptive Current Statistical (CS) model based on the estimation of the target motion parameters is proposed. Firstly, the improved CS model is derived by adding the jerk and jerk rate into the state vector, to get a more accurate estimation of the target motion state. Secondly, the distance, speed, acceleration, and jerk obtained by motion parameters estimation are also added to the model measurement. Furthermore, the changes of a jerk in adjacent cycles are compensated by adaptive updating of maneuver frequency. Finally, the filter equation of the improved CS model is introduced. The simulation results show that the proposed model can improve the accuracy of target tracking compared with the model that only measures distance and speed.

Foreign object debris (FOD) detection can be considered a kind of classification that distinguishes the measured signal as either containing FOD targets or only corresponding to ground clutter. In this paper, we propose a support vector domain description (SVDD) classifier with the particle swarm optimization (PSO) algorithm for FOD detection. The echo features of FOD and ground clutter received by the millimeter-wave radar are first extracted in the power spectrum domain as input eigenvectors of the classifier, followed with the parameters optimized by the PSO algorithm, and lastly, a PSO-SVDD classifier is established. However, since only ground clutter samples are utilized to train the SVDD classifier, overfitting inevitably occurs. Thus, a small number of samples with FOD are added in the training stage to further construct a PSO-NSVDD (NSVDD: SVDD with negative examples) classifier to achieve better classification performance. Experimental results based on measured data showed that the proposed methods could not only achieve a good detection performance but also significantly reduce the false alarm rate.

This letter proposes a blind phase compensation method for the phase errors in the Multi-Carrier Multiple-input multiple-output (MIMO) radar, which decouples the range and DOA coupling. The phase errors under the Linear Frequency Modulated Continuous Waveform (LFMCW) scheme are firstly derived, followed with the signal processing steps. Further, multiple targets with certain velocities can be handled uniformly without pre-knowledge of the actual range information of the targets. The evaluations of the DOA estimation performance are carried out through simulations, which validate the effectiveness of the proposed method.