Vehicle recognition is a fundamental problem in SAR image interpretation. However, robustly recognizing vehicle targets is a challenging task in SAR due to the large intraclass variations and small interclass variations. Additionally, the lack of large datasets further complicates the task. Inspired by the analysis of target signature variations and deep learning explainability, this paper proposes a novel domain alignment framework named the Hierarchical Disentanglement-Alignment Network (HDANet) to achieve robustness under various operating conditions. Concisely, HDANet integrates feature disentanglement and alignment into a unified framework with three modules: domain data generation, multitask-assisted mask disentanglement, and domain alignment of target features. The first module generates diverse data for alignment, and three simple but effective data augmentation methods are designed to simulate target signature variations. The second module disentangles the target features from background clutter using the multitask-assisted mask to prevent clutter from interfering with subsequent alignment. The third module employs a contrastive loss for domain alignment to extract robust target features from generated diverse data and disentangled features. Lastly, the proposed method demonstrates impressive robustness across nine operating conditions in the MSTAR dataset, and extensive qualitative and quantitative analyses validate the effectiveness of our framework.

In recent years, deep learning has been widely used in SAR ATR and achieved excellent performance on the MSTAR dataset. However, due to constrained imaging conditions, MSTAR has data biases such as background correlation, i.e ., background clutter properties have a spurious correlation with target classes. Deep learning can overfit clutter to reduce training errors. Therefore, the degree of overfitting for clutter reflects the non-causality of deep learning in SAR ATR. Existing methods only qualitatively analyze this phenomenon. In this paper, we quantify the contributions of different regions to target recognition based on the Shapley value. The Shapley value of clutter measures the degree of overfitting. Moreover, we explain how data bias and model bias contribute to non-causality. Concisely, data bias leads to comparable signal-to-clutter ratios and clutter textures in training and test sets. And various model structures have different degrees of overfitting for these biases. The experimental results of various models under standard operating conditions on the MSTAR dataset support our conclusions. Our code is available at https://github.com/waterdisappear/Data-Bias-in-MSTAR.

Micro‐range (m‐R) signatures which are induced by micro‐motion dynamics can be observed from range profiles, providing that the range resolution of radar is high enough. For real scenarios, micro‐motion is often mixed with macro‐motion (translation). To extract the micro‐motion signatures, it is required to remove the macro‐motion component. The widely employed range alignment technique fails for rigid‐body targets with micro‐motion, since the relative distances between different scattering centres on a rigid‐body target are varying and it is unable to obtain a stable reference range profile. Thus, the extracted m‐R signatures will be accompanied with residual macro‐motion, which may lead to the degradation. However, this issue is often ignored in the research of m‐R signatures extraction. In this work, by modelling the motions of scattering centres as the superimposition of a polynomial signal (represents macro‐motion) and a sinusoidal signal (represents micro‐motion), a micro‐motion period estimation method based on high‐order difference sequence is proposed. The property that the difference operation can decrease the order of polynomial signals while preserve sinusoidal signals with the same frequency enables the proposed method to extract m‐R signatures in the presence of macro‐motion. The effectiveness of the proposed method is validated by synthetic and measured radar data.

The combination of frequency modulation continuous wave (FMCW) technology and circular synthetic aperture radar (CSAR) would pave the way for the development of the microminiature unmanned aerial vehicle (UAV) earth observation application. Regarding the FMCW CSAR imaging process, the stop-and-go approximation is no longer effective due to the longer pulse duration. In addition, the signal characteristics of FMCW CSAR with higher frequency band and higher resolution need to be further studied. Aiming at millimeter-wave FMCW CSAR imaging, the signal models of FMCW CSAR in the time domain and wavenumber domain are first derived in this work. Then, the signal characteristics including the two-dimensional (2-D) wavenumber spectrum, the spectrum support region, the point spread function, and the resolution are analyzed. Moreover, the effects of intra-pulse movement are analyzed, including the range offset error, the main lobe broadening, and the additional phase error in the range dimension, as well as the extra first-order and second-order phase errors in the 2-D wavenumber domain. Based on the intra-pulse motion analysis, two criterions are provided to determine whether the intra-pulse movement of FMCW CSAR is negligible. An improved back projection (BP) algorithm compensating the intra-pulse motion in the range domain and another algorithm combining the traditional BP and compensation the intra-pulse motion in the wavenumber domain are proposed. Simulation results verify the correctness of the FMCW CSAR signal characteristics analysis and the effectiveness of the intra-pulse movement compensation methods.

Micro-Doppler (m-D) period is an important feature of micro-motion targets, playing a vital role in detecting and recognizing moving targets like ground vehicles, aircraft, and space debris. Existing m-D period estimation algorithms mostly exploit the periodicity of the global radar echo signals, and extra compensation steps are needed under high-order translations. To solve that problem, this work focuses on the periodicity of the local structure of radar signals, and a short-time similarity statistic framework is introduced. It is a standard m-D period estimation procedure where various statistics can be flexibly incorporated. Under this framework three novel short-time similarity statistics – short-time ambiguity function (AF) entropy, scatter feature correlation coefficient, and normalized time-frequency (TF) Wasserstein distance are presented. Our proposed framework and specific statistics are proved to outperform conventional global statistic methods under low signal-noise ratio (SNR), high-order translations, and signal-missing scenarios through Monte Carlo simulations. The proposed methods in this work are also validated through the SAR experiment.