The Radon-Fourier transform (RFT) with a long coherent integration time has recently been proposed for detecting a moving target with an across range cell (ARC) effect. However, without effective clutter suppression, clutter will also be integrated via the RFT, which may affect weak target detection. Based on the maximal signal-to-clutter-plus-noise ratio (SCNR) criteria, a novel adaptive RFT (ARFT) is proposed in this paper to effectively detect a “low-observable” target in a clutter background. The proposed ARFT can combine RFT and adaptive clutter suppression by introducing an optimal filter weight, which is determined from the clutter's covariance matrix as well as a steering vector for a moving target with the ARC effect. In the transformed range-velocity space, the proposed ARFT can suppress background clutter and optimally integrate the target's energy. Nevertheless, with the increase in the integration time, the ARFT needs to address two difficulties in its real implementation. One is the lack of independently and identically distributed (i.i.d.) training samples in a heterogeneous clutter background, and the other is that the computational complexity is too high due to the large number of pulse samples. Therefore, a subaperture ARFT (SA-ARFT) is further proposed in this paper. It divides all coherent pulse samples into several subapertures and accomplishes adaptive clutter suppression in each subaperture. Subsequently, SA-ARFT implements coherent integration among the outputs of different subapertures. The proposed SA-ARFT method can obtain a similar SCNR improvement factor (SCNR IF) with a large number of i.i.d. training samples, while it can obtain a much higher SCNR IF than the ARFT with limited i.i.d. training samples and much lower computational complexity in a heterogeneous clutter background. Finally, some numerical results are provided to demonstrate the effectiveness of the two proposed methods.

For modern synthetic aperture radar (SAR), it has much more urgent demands on ground moving target indication (GMTI), which includes not only the point moving targets like cars, truck or tanks but also the distributed moving targets like river or ocean surfaces. Among the existing GMTI methods, displaced phase center antenna (DPCA) can effectively cancel the strong ground clutter and has been widely used. However, its detection performance is closely related to the target’s signal-to-clutter ratio (SCR) as well as radial velocity, and it cannot effectively detect the weak large-sized river surfaces in strong ground clutter due to their low SCR caused by specular scattering. This paper proposes a novel method called relative residue of DPCA (RR-DPCA), which jointly utilizes the DPCA cancellation outputs and the multi-look images to improve the detection performance of weak river surfaces. Furthermore, based on the statistics analysis of the RR-DPCA outputs on the homogenous background, the cell average (CA) method can be well applied for subsequent constant false alarm rate (CFAR) detection. The proposed RR-DPCA method can well detect the point moving targets and distributed moving targets simultaneously. Finally, the results of both simulated and real data are provided to demonstrate the effectiveness of the proposed SAR/GMTI method.

In this letter, a robust multi-component secure transmission scheme based on general multi-fractional Fourier transform (GMFRFT) with imperfect channel state information (CSI) is proposed and analyzed. The scheme combines signal and spatial domains for transmitted components design, aiming at degrading the wiretap channel utilizing the energy distribution characteristics of GMFRFT. The inter-component interference is only formed at the eavesdropper because of the broken GMFRFT signal, and its effect is similar to artificial noise (AN). Considering the problem of AN leakage in traditional AN-based scheme under imperfect CSI, the proposed scheme does not consume extra power and can further improve the secrecy performance when the available spatial degrees of freedom are limited. Numerical results are carried out to demonstrate the secrecy performance of the proposed scheme.

To improve the detection performance in the presence of target‐steering vector mismatches, a novel robust adaptive matched filter (AMF) detector for compound Gaussian clutter is proposed. First, the multivariate inverse Gaussian distribution is first introduced to compound Gaussian clutter model. Second, the texture value is estimated with maximum a posteriori (MAP) estimator in spatial domain, which has a closed form and is not affected by the target's steering vector mismatch. A robust AMF detector is derived based on this estimation. Simulation results demonstrate that the proposed detector can achieve better performance in the presence of target‐steering vector mismatches.

Millimeter-wave (MMW) radar, as a widely used sensor, is increasingly employed in intelligent driving systems. Particularly in challenging driving conditions, such as adverse weather or varying light levels, radar often outperforms other sensors like camera and LiDAR. However, due to the relatively low resolution of radar sensors and its original complex data being difficult to process, deep learning approaches have been less explored in radar target detection compared to camera and LiDAR. Furthermore, we observe that existing methods in the literature often suffer from slow processing speeds and relatively low precision, which poses significant challenges for meeting the detection accuracy and real-time performance requirements of autonomous driving systems. To address these challenges, this paper proposes a lightweight radar target detection network designed to efficiently detect targets from FFT-processed radar frequency data in Range-Doppler spectrum format. The proposed model draws inspiration from YOLOv7-tiny and is further optimized to accommodate the specific characteristics of radar data. A novel, lightweight feature extraction backbone and a multi-scale feature enhancement network are introduced, significantly improving detection accuracy while reducing computational complexity. Additionally, to fully exploit radar data, various data augmentation strategies are explored, and a more efficient augmentation method is developed. Extensive experimental evaluations demonstrate that our model achieves superior detection accuracy and faster inference speed compared to state-of-the-art methods on the RADDet and CARRADA datasets.