A transmission-reflection-integrated multifunctional metasurface (TRMM) has been proposed, capable of switching between transmission and reflection modes. The switchable characteristics is achieved by integrating a broadened orthogonal grating structure with PIN diodes. By incorporating an artificial magnetic conductor (AMC) with varactors, the polarization of incident waves can be controlled as altering the surface impedance distribution. Simulation and experimental results indicate that in the frequency range of 3.67-4.44 GHz, incident waves are reflected in the form of cross-polarization, with a maximum reflection amplitude of -1.27 dB and a relative bandwidth of 19%. The frequency range shifts as the capacitance value of the varactor is adjusted. When the metasurface operates in the transmissive mode, the outgoing wave maintains polarization along the y-direction, regardless of the polarization direction of the incident electromagnetic wave along the x or y-direction. The center frequencies are 4.26 GHz (x polarization) and 4.89 GHz (y polarization), with maximum transmission coefficients of -2.23 dB and -3.0 dB, respectively. This study overcomes the limitations associated with half-space electromagnetic (EM) waves by regulating the transmission and reflection of EM waves throughout the entire space, as well as effectively controls the polarization parameters within specific frequency bands. This advancement significantly expands the coverage range of communication systems and offers a novel technical solution for high-performance radar detection, intelligent antenna systems, and various other applications.

Deep learning has significantly enhanced the performance of automatic target recognition (ATR) in synthetic aperture radar (SAR). However, the concept of model overinterpretation, characterized by classifiers discerning strong class evidence within image regions that lack semantically salient features related to target (e.g., background clutter in SAR images), has undermined confidence in the reliability of deep learning models. Previous studies predominantly relied solely on one interpretability method to qualitatively identify the key input pixels, without assessing the efficiency of these features in decision-making process, posing a significant hurdle in evaluating the model overinterpretation. In this paper, we propose necessity-sufficiency index (NSI) to select models’ decision-making basis among all the key regions identified by multiple interpretability methods and segmentation algorithm. Furthermore, we propose weighted composition ratio statistic (WCRS) method to quantitatively analyze the model overinterpretation by incorporating the NSI as weighted average weights. The experimental results indicate that our methods are capable of accurately identifying the decision-making features and quantitatively analyzing the models’ tendency towards overinterpretation.

High-resolution range profile (HRRP) target recognition has garnered significant attention in radar automatic target recognition (RATR) research for its rich structural information and low computational costs. With the rapid advancements in deep learning, methods for HRRP target recognition that leverage deep neural networks have emerged as the dominant approaches. Nevertheless, these traditional closed-set recognition methods require labeled data for every class in training, while in reality, seen classes and unseen classes coexist. Therefore, it is necessary to explore methods that can identify both seen and unseen classes simultaneously. To this end, a semantic- and physical-guided generative network (SPGGN) was innovatively proposed for HRRP generalized zero-shot recognition; it combines a constructed knowledge graph with attribute vectors to comprehensively represent semantics and reconstructs strong scattering points to introduce physical constraints. Specifically, to boost the robustness, we reconstructed the strong scattering points from deep features of HRRPs, where class-aware contrastive learning in the middle layer effectively mitigates the influence of target-aspect variations. In the classification stage, discriminative features are produced through attention-based feature fusion to capture multi-faceted information, while the design of balancing loss abates the bias towards seen classes. Experiments on two measured aircraft HRRP datasets validated the superior recognition performance of our method.

This paper presents a maritime target detection framework termed Multi-Domain Features Guided Supervised Contrastive Learning (MDFG_SCL), which integrates physically interpretable statistical features with deep feature representations within a supervised contrastive learning framework. To further enhance the discriminative capability of the shallow features, the Gini Index is calculated for different candidate features to evaluate their separability between clutter and target echoes, and these features are then adaptively combined through a weighted strategy. Specifically, in the wavelet domain, two effective features are designed, where radar echoes are processed through wavelet decomposition to extract the Wavelet Approximation Energy Ratio and Wavelet Entropy, which characterize signal energy distribution and complexity. These features, along with additional domain-specific components, are incorporated as structured priors to guide the representation learning of a convolutional neural network. This fusion enhances the model's ability to distinguish between sea clutter and target echoes. Experimental results on both simulated and real radar datasets demonstrate that the proposed method significantly improves detection probability and enhances generalization performance across diverse maritime environments.

Deep learning-based synthetic aperture radar (SAR) automatic target recognition (ATR) methods exhibit a tendency to overfit specific operating conditions—such as radar parameters and background clutter—which frequently leads to high sensitivity against variations in these conditions. A novel electromagnetic reconstruction feature alignment (ERFA) method is proposed in this paper, which integrates electromagnetic reconstruction with feature alignment into a fully convolutional network, forming the ERFA-FVGGNet. The ERFA-FVGGNet comprises three modules: electromagnetic reconstruction using our proposed orthogonal matching pursuit with image-domain cropping-optimization (OMP-IC) algorithm for efficient, high-precision attributed scattering center (ASC) reconstruction and extraction; the designed FVGGNet combining transfer learning with a lightweight fully convolutional network to enhance feature extraction and generalization; and feature alignment employing a dual-loss to suppress background clutter while improving robustness and interpretability. Experimental results demonstrate that ERFA-FVGGNet boosts trustworthiness by enhancing robustness, generalization and interpretability.