IEEE Transactions on Radar Systems最新文献

{"title":"On-Demand Beam-Widening Reflector Antenna for a J-Band Imaging Radar With Micro-Doppler Measurement Capability","authors":"Abdullah Alburadi;Adib Y. Nashashibi;Hussein N. Shaman;Kamal Sarabandi","doi":"10.1109/TRS.2026.3654524","DOIUrl":"https://doi.org/10.1109/TRS.2026.3654524","url":null,"abstract":"In this article, a novel technique is proposed to widen the beamwidth of a millimeter-wave (MMW) imaging radar system that employs a mechanically scanning elliptic toroidal reflector antenna. For applications requiring Doppler measurements, increased beamwidth enhances Doppler resolution while maintaining scanning speed, enabling improved extraction of micro-Doppler features. The reflector antenna’s beamwidth is widened by employing a feed horn with an adjustable phase center. Relying on the fact that the horn’s phase center is near its aperture, the design incorporates a hollow cylindrical sleeve that extends the aperture forward, thereby shifting the phase center. This phase center displacement de-focuses the reflector and results in a wider radar beam. The design of the feed horn and its electrically actuated sleeve is presented, along with fabrication details. The phase center position of the horn is determined using two experimental techniques. In the first, the horn is placed behind a dielectric lens illuminated by a plane wave, and the phase center is localized by identifying the position that maximizes the received signal power. In the second method, coherent signal measurements on the plane perpendicular to the horn’s boresight are collected using a network analyzer and processed to localize its phase center. In both techniques, the phase center is shown to shift when the sleeve is extended. Beam widening is verified by measuring the reflector’s radiation pattern with the sleeve both extended and retracted, showing good agreement with simulations. A beamwidth increase by a factor of four is demonstrated using a 20-mm sleeve extension. The technique’s effectiveness is further validated through Doppler measurements of a walking person, where Doppler spread is observed using the defocused beam at scanning speeds of up to 5 cycles/s.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"309-320"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148486810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint Power/Amplitude Spectral Estimation for SAR Image Stabilization","authors":"Alex Batts;Brian Rigling","doi":"10.1109/TRS.2026.3718337","DOIUrl":"https://doi.org/10.1109/TRS.2026.3718337","url":null,"abstract":"Due to the Fourier relationship between the phase history data and the scene reflectivity function, synthetic aperture radar (SAR) imaging can be viewed as a spectral estimation problem. Classical Fourier methods of SAR imaging have several drawbacks with respect to image quality, which modern spectral estimation techniques can mitigate. However, these modern imaging techniques are based on power spectral density (PSD) estimation, and therefore phase information is usually lost. This article presents a novel method that combines images formed from PSD and amplitude spectrum (AS) estimates to produce an image that has stabilized magnitude and phase. The benefits of the technique are demonstrated through simulations and measured data experiments.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1516-1526"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimized Feature Correlation for Quantum SVM Detection of Robotic Dogs From Radar Micro-Doppler Signatures","authors":"Minseo Seong;Insoo Choi;Youngwook Kim","doi":"10.1109/TRS.2026.3725673","DOIUrl":"https://doi.org/10.1109/TRS.2026.3725673","url":null,"abstract":"This study proposes optimizing correlated features to effectively classify radar micro-Doppler signatures of quadrupedal robotic dogs using quantum support vector machines (QSVMs). Detecting robotic dogs is crucial for ensuring safety in human–robot interactions and strengthening military surveillance and strategic operations. Rather than a conventional SVM, we apply a ZZFeatureMap-based QSVM that utilizes quantum circuit-based kernel computation to improve classification accuracy. To efficiently utilize QSVM, we suggest intentionally preserving correlations between features during dimensionality reduction, which is expected to leverage characteristics of quantum entanglement. On the measured dataset, correlated features outperform traditional feature extraction methods that eliminate feature correlation. Further, we investigate the optimal value of correlation by controlling it using a shallow convolutional neural network (CNN). In our experiments, QSVM with features of moderate correlations (0.5) achieved higher accuracy than classical SVM when paired with a shallow, low-FLOPs backbones. These findings suggest that introducing correlated features for QSVM paired with a low-FLOPs backbone can lead to improved performance of micro-Doppler classification.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1457-1466"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CompactRADNet: Embeddable Temporal Road-User Detection From Radar","authors":"Fahed Hassanat;Robert Laganière","doi":"10.1109/TRS.2026.3673162","DOIUrl":"https://doi.org/10.1109/TRS.2026.3673162","url":null,"abstract":"Achieving accurate road-user detection using automotive radar while maintaining computational efficiency for embedded deployment remains challenging. This work presents CompactRADNet, a hybrid convolutional neural network (CNN)–MetaFormer architecture that processes temporal radar sequences for efficient detection of dynamic road users. The MetaFormer-based temporal stem replaces computationally expensive self-attention mechanisms with efficient token mixing strategies, including pooling-based, convolutional, and shift-based temporal fusion, with the pooling mixer achieving best performance. This approach achieves superior performance with dramatically reduced computational requirements compared to transformer variants. In addition, we introduce the adaptive quadratic rectified linear unit (ReLU) [AQR] activation function, specifically designed for radar signal processing. AQR incorporates a learned gating mechanism that provides input-dependent modulation for handling varying radar signal strengths within the target radar configuration. Extensive experiments on the Camera-Radar of the University of Washington (CRUW) dataset demonstrate that the MetaFormer configuration achieves 81.10% average precision (AP) with three-frame sequences, outperforming transformer variants (79.51% AP) while requiring approximately 60% fewer operations. The optimal 13-frame configuration achieves 88.15% AP with only 1.65M parameters and 15.85 giga floating-point operations per second (GFLOPs), representing 73% fewer parameters and 52% less computational cost compared with previous methods. The low parameter count (1.65M) and reduced computational requirements make CompactRADNet suitable for deployment on embedded automotive platforms such as NVIDIA Jetson. Signal processing analysis reveals that AQR achieves 2.59-dB improvement in target detection signal-to-noise ratio compared with conventional activation functions. These results establish that domain-specific architectural innovations can achieve state-of-the-art performance while maintaining computational efficiency suitable for widespread deployment in autonomous driving systems.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"650-660"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of Proactive Uncoordinated Techniques to Mitigate Interference in FMCW Automotive Radars","authors":"Alessandro Bazzi;Francesco Miccoli;Fabrizio Cuccoli;Luca Facheris;Vincent Martinez","doi":"10.1109/TRS.2026.3670939","DOIUrl":"https://doi.org/10.1109/TRS.2026.3670939","url":null,"abstract":"Modern vehicles increasingly rely on advanced driver-assistance systems (ADAS), with radars playing a key role due to their cost-effectiveness and reliable performance. However, the growing number of radars operating in the same spectrum raises concerns about mutual interference, which could lead to system malfunctions and potential safety risks. This study focuses on a scenario in which all vehicles are equipped with frequency-modulated continuous-wave (FMCW) radars, and it assesses the impact of interference on radar functionality—expressed in terms of probability of failure—by considering both direct and reflected signals. The radars may employ one of the following proactive mitigation methods to reduce the impact of interference, all of which require no intervehicle coordination but differ in complexity: 1) random carrier-frequency hopping on a frame-by-frame basis; 2) random carrier-frequency hopping on a chirp-by-chirp basis; and 3) a directional, compass-based method specifically addressing interference from opposite directions, which can be combined with either of the two previous methods. In this work, we assume realistic simulated road-traffic scenarios and develop a novel model that captures correlated interference and accounts for the main radar setting parameters. Results reveal that dense scenarios pose a high risk of radar malfunctions. Among the analyzed methods, chirp-by-chirp frequency hopping emerges as the most effective approach to mitigate interference and ensure system reliability, but only when combined with a sufficiently large bandwidth. The compass-based method, on the other hand, shows limited effectiveness and appears not worth the additional system complexity.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"597-612"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Multitarget Tracking Approach for UAV-Borne Video SAR","authors":"Libin Han;Xuejun Huang;Liwu Wen;Jinshan Ding","doi":"10.1109/TRS.2026.3651555","DOIUrl":"https://doi.org/10.1109/TRS.2026.3651555","url":null,"abstract":"Shadow-based moving target tracking is essential for uncrewed aerial vehicle (UAV)-borne video synthetic aperture radar (ViSAR). Traditional detect-before-track (DBT) algorithms are ineffective for dim shadows, while track-before-detect (TBD) algorithms incur prohibitive computational costs. Moreover, improving tracking accuracy for dim shadows often increases false alarms in cluttered areas. This article presents an adaptive shadow-aware tracking (ASAT) algorithm and its implementation for low-power edge deployment. The algorithm adaptively extracts the moving target shadow aggregation area (MSAA) from current frames using prior tracking results. Shadows within the MSAA are classified into strong and dim categories using multivalue constant false alarm rate (CFAR) detection and morphological processing. Dim shadow traces are then processed by a TBD-based algorithm, while strong shadow traces are handled by a DBT-based algorithm. Together, adaptive MSAA extraction and subsequent classification-based processing form a joint strategy that enhances tracking accuracy and improves real-time performance on low-power edge platforms. The ASAT algorithm is deployed on a low-power edge system-on-module (SoM) featuring a multicore central processing unit (CPU) and an embedded graphics processing unit (GPU). To enable low-power edge-oriented algorithm design and lightweight implementation, an algorithm–hardware codesign methodology is introduced. Experiments on UAV-borne ViSAR data demonstrate that the proposed approach effectively outperforms traditional approaches in both tracking accuracy and real-time performance on UAV-borne low-power edge platforms.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"280-295"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11333891","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148486594","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feature-to-Image RGB Encoding of Physical Tri-Features for Maritime Radar Small Target Detection","authors":"Shiji Fan;Pedro Pinho;Feng Luo;Ming Shen;Daniel Albuquerque","doi":"10.1109/TRS.2026.3707962","DOIUrl":"https://doi.org/10.1109/TRS.2026.3707962","url":null,"abstract":"Detection of slow-moving small maritime targets under strong sea clutter remains a challenging problem, particularly in low signal-to-clutter ratio conditions where target energy leakage produces ambiguous secondary cells. Conventional tri-feature detectors, such as those based on relative average amplitude (RAA), relative Doppler peak height (RDPH), and relative vector entropy (RVE), improve target–clutter separability but typically rely on point-wise decision strategies and often exclude secondary cells during evaluation, which limits precise localization capability in realistic scenarios. This article proposes a physically informed feature-to-image encoding framework that enables joint spatial interpretation of multiple radar-derived features without modifying standard deep learning (DL) detection architectures. Instead of processing features independently or employing multistream fusion networks, the proposed method maps RAA, RDPH, and RVE onto orthogonal channels of a unified three-channel representation, preserving spatial continuity along the range dimension and facilitating structured context learning. Based on this encoding, lightweight You Only Look Once (YOLO) detectors are adapted for radar target detection with a scale-consistent head design and a Neyman–Pearson-based statistical thresholding strategy to ensure controllable false alarm performance. Experiments conducted on the measured IPIX maritime radar dataset demonstrate that the proposed approach achieves superior detection probability under low signal-clutter ratio (SCR) conditions while maintaining strict false alarm control, particularly in scenarios where secondary cells are retained. The results indicate that appropriate multifeature representation and spatially aware decision mechanisms can substantially enhance maritime radar small target detection without increasing system complexity.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1265-1281"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"m2rDM: A Map-Conditioned Diffusion Model for Maritime X-Band Radar Synthesis and Dynamic Object Suppression","authors":"Jinbum Park;Dongje Lee;Yejin Kang;Hanguen Kim;Hyesu Jang","doi":"10.1109/TRS.2026.3714779","DOIUrl":"https://doi.org/10.1109/TRS.2026.3714779","url":null,"abstract":"Autonomous vessel operation relies on X-band radar for navigation because it offers greater robustness than optical sensors under adverse weather and sea conditions. Place recognition and map-based navigation algorithms assume a static world, yet terrain, vessel, and sea-clutter returns share overlapping intensities and morphologies in real radar scans with unreliable separation. Terrain-only radar data are therefore required for training and evaluating maritime autonomy algorithms, but public datasets remain unavailable, and electronic navigational charts (ENCs) cannot reproduce real radar scattering characteristics. We propose m2rDM, a map-conditioned diffusion model with two complementary applications: m2rDM-RS generates synthetic terrain-accurate X-band radar images from ENC map masks without at-sea data collection, and m2rDM-DS suppresses dynamic-object returns in real radar scans to yield terrain-only outputs for downstream perception. The model is conditioned on geolocation-aligned map masks, radar-specific embeddings from a pretrained segmentation encoder, and temporal context from consecutive radar frames. On the MOANA dataset, m2rDM outperforms existing generative baselines in pixel-level accuracy [peak signal-to-noise ratio (PSNR), root-mean-square error (RMSE)] and gray-level co-occurrence matrix (GLCM)-based texture fidelity, with m2rDM-DS achieving the highest reconstruction accuracy and m2rDM-RS producing the closest texture distributions from map masks alone. m2rDM-DS also improves place recognition accuracy on dense-traffic sequences.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1305-1317"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physics-Guided Learnable Wavelet Framework for Micro-Doppler Characterization of Low-Altitude Radar Targets","authors":"Xiaolong Chen;Jinhao Wang;Zhijian Zhao;Jian Guan;Xiaolin Du;Yifeng Wu;Hai Zhang","doi":"10.1109/TRS.2026.3738487","DOIUrl":"https://doi.org/10.1109/TRS.2026.3738487","url":null,"abstract":"Reliable micro-Doppler characterization is relevant to low-altitude radar sensing, where measured echoes may contain weak returns. Fixed-parameter time–frequency methods can produce smeared or fragmented time–frequency representations (TFRs), which can affect weak-component observability and Doppler-ridge traceability. This article presents WConV-TF, a physics-guided learnable wavelet front end for radar micro-motion analysis. WConV-TF converts preprocessed target-range-cell echoes into micro-Doppler TFRs by embedding parameterized Morlet bases in a learnable complex wavelet convolution layer. The normalized center frequencies and bandwidths are optimized from data while retaining a time–frequency interpretation. A multiscale residual module then refines the magnitude-domain TFR. The method is evaluated using simulated data, measured frequency-modulated continuous-wave (FMCW) radar data, and measured S-band array radar data. Across four measured target categories, WConV-TF reduces Rényi entropy by approximately 0.4%–46.3% relative to the strongest competing method for each category and reduces the controlled-rotor mean absolute localization error (MALE) by approximately 66.9% relative to the best baseline. On the common four-category sequence-level split, it improves downstream classification accuracy by 4.66 percentage points over short-time Fourier transform (STFT). The results across simulated signals, two measured radar systems, and four target types indicate applicability to low-altitude radar signal analysis under the evaluated protocols.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1678-1691"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149133407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and Evaluation of Multibeam Transmit and Nulled-Receive Array Patterns for Polarimetric Weather Radar Surveillance","authors":"David Schvartzman;Robert D. Palmer;Matthew Herndon;Dušan S. Zrnić","doi":"10.1109/TRS.2026.3729549","DOIUrl":"https://doi.org/10.1109/TRS.2026.3729549","url":null,"abstract":"Simultaneous transmit beams offer a direct way to increase temporal sampling of phased-array weather radar while retaining narrow angular illumination. With <inline-formula> <tex-math>$N_{b}$ </tex-math></inline-formula> directions observed in one dwell, the number of dwells required to sample a fixed angular set can be reduced by a factor proportional to <inline-formula> <tex-math>$N_{b}$ </tex-math></inline-formula>. Herein, we design and experimentally evaluate an end-to-end multibeam transmit/receive scan mode on the Horus fully digital polarimetric phased-array radar. Measured phase-only transmit patterns containing two through seven simultaneous mainlobes are paired with one precomputed constrained receive beam per mainlobe. The receive coefficients use established regularized linearly constrained minimum-norm synthesis to suppress returns from the other known transmit directions. An end-to-end scan mode is proposed, and scans with two through seven peaks are demonstrated. Polarimetric measurements of the patterns are used to quantify two-way performance. Relative to conventional digital beamforming (DBF), the measured nulled-receive beams reduce average copolar crosstalk integrated sidelobe level (ISL) by 13.6–18.9 dB over 3-dB mainlobe regions and by 9.9–13.8 dB over 10-dB mainlobe regions. The case-average receive white-noise-gain loss is 0.03–0.49 dB, with a worst-beam loss of 0.59 dB. Average intended-beam integrated cross-polar levels remain between −40.0 and −37.8 dB (as with conventional DBF), and the average H/V centroid separation remains below 0.084°. These results establish a concept of operations for rotating, polarimetric phased-array radar, capable of meeting temporal resolution, spatial sampling, and data quality requirements simultaneously.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"4 ","pages":"1527-1543"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11675937","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148985605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}