Pub Date : 2024-12-09DOI: 10.1109/TRS.2024.3513293
Chandra S. Pappu;Sonny Grooms;Dmitriy Garmatyuk;Thomas L. Carroll;Aubrey N. Beal;Saba Mudaliar
The increased usage of wireless services in the congested electromagnetic spectrum has caused communication systems to contend with the existing operational radar frequency bands. Integrated sensing and communication (ISAC) systems that share the same frequency band and signaling strategies, such as a single radio frequency emission, address these congestion issues. In this work, we propose novel chaotic signal processing techniques and waveform design methods for ISAC systems. First, we consider a family of chaotic oscillators and use their output to encode the information. Next, we multiplex the information carrying chaotic signals to improve the data rates significantly and further use it for ISAC transmission. We show that a simple correlator can accurately decode the information with low bit-error rates. The performance of the multiplexed waveform is robust in the Rician multipath channel. Using correlation and ambiguity function analysis, we claim that the proposed waveforms are excellent candidates for high-resolution radar imaging. We generate synthetic aperture radar (SAR) images using the backprojection algorithm (BPA). The SAR images generated using multiplexed chaos-based waveforms are of similar quality compared to traditionally used linear frequency-modulated waveforms. The most important feature of the proposed multiplexed chaos-based waveforms is their inherent resilience to intentional and nonintentional interference.
{"title":"Interference Resilient Integrated Sensing and Communication Using Multiplexed Chaos","authors":"Chandra S. Pappu;Sonny Grooms;Dmitriy Garmatyuk;Thomas L. Carroll;Aubrey N. Beal;Saba Mudaliar","doi":"10.1109/TRS.2024.3513293","DOIUrl":"https://doi.org/10.1109/TRS.2024.3513293","url":null,"abstract":"The increased usage of wireless services in the congested electromagnetic spectrum has caused communication systems to contend with the existing operational radar frequency bands. Integrated sensing and communication (ISAC) systems that share the same frequency band and signaling strategies, such as a single radio frequency emission, address these congestion issues. In this work, we propose novel chaotic signal processing techniques and waveform design methods for ISAC systems. First, we consider a family of chaotic oscillators and use their output to encode the information. Next, we multiplex the information carrying chaotic signals to improve the data rates significantly and further use it for ISAC transmission. We show that a simple correlator can accurately decode the information with low bit-error rates. The performance of the multiplexed waveform is robust in the Rician multipath channel. Using correlation and ambiguity function analysis, we claim that the proposed waveforms are excellent candidates for high-resolution radar imaging. We generate synthetic aperture radar (SAR) images using the backprojection algorithm (BPA). The SAR images generated using multiplexed chaos-based waveforms are of similar quality compared to traditionally used linear frequency-modulated waveforms. The most important feature of the proposed multiplexed chaos-based waveforms is their inherent resilience to intentional and nonintentional interference.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"3 ","pages":"26-43"},"PeriodicalIF":0.0,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10781422","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142880292","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}
Pub Date : 2024-12-02DOI: 10.1109/TRS.2024.3509775
Nicolas C. Kruse;Ronny G. Guendel;Francesco Fioranelli;Alexander Yarovoy
The problem of human activity classification using a distributed network of radar sensors has been considered. A novel sensor fusion method has been proposed that processes data from a network of radar sensors and yields 3-D representations of both reflection intensity and velocity distribution. The formulated method has been verified in an experimental case study, where activity classification was performed using data collected with 14 participants moving in diverse, unconstrained trajectories and executing nine activities. The classification performance of the proposed method has been compared to alternative fusion methods on the same dataset, and a test accuracy and macro $F1$ -score of, respectively, 87.4% and 81.9% have been demonstrated. A feasibility study has also been performed to demonstrate the ability of the proposed method to generate 3-D distributions of intensity and target velocity.
{"title":"Reconstruction of Extended Target Intensity Maps and Velocity Distribution for Human Activity Classification","authors":"Nicolas C. Kruse;Ronny G. Guendel;Francesco Fioranelli;Alexander Yarovoy","doi":"10.1109/TRS.2024.3509775","DOIUrl":"https://doi.org/10.1109/TRS.2024.3509775","url":null,"abstract":"The problem of human activity classification using a distributed network of radar sensors has been considered. A novel sensor fusion method has been proposed that processes data from a network of radar sensors and yields 3-D representations of both reflection intensity and velocity distribution. The formulated method has been verified in an experimental case study, where activity classification was performed using data collected with 14 participants moving in diverse, unconstrained trajectories and executing nine activities. The classification performance of the proposed method has been compared to alternative fusion methods on the same dataset, and a test accuracy and macro \u0000<inline-formula> <tex-math>$F1$ </tex-math></inline-formula>\u0000-score of, respectively, 87.4% and 81.9% have been demonstrated. A feasibility study has also been performed to demonstrate the ability of the proposed method to generate 3-D distributions of intensity and target velocity.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"3 ","pages":"14-25"},"PeriodicalIF":0.0,"publicationDate":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142798013","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}
In synthetic aperture radar (SAR) information acquisition, target detection is often performed on the basis of the acquired radar images. Under low signal-to-clutter ratio (SCR) or low signal-to-noise ratio (SNR) conditions, detection by images is likely to cause loss of targets. To address this problem, we propose a joint imaging and target detection network based on Bayesian maximum a posteriori (MAP) estimation. The imaging and detection results are, respectively, defined as scene magnitude and detection label, and their joint probability distribution is used in place of the distribution of scene magnitudes. In the MAP estimation, the continuity feature of the detection label is merged into the optimization process, and the imaging and detection results are optimized alternately to get an iterative solution. The iterative solution is then unrolled into a network, which consists of three modules. We first utilize the unrolled fast iterative shrinkage thresholding algorithm (FISTA) method for the image formation module and then incorporate the detection label estimation module and distribution parameter updating module to learn the detection label and the function of distribution parameters. This approach applies prior information for both imaging and detection processes and enables automatic learning of parameters that are difficult to fit. Simulation experiments demonstrate that the method can simultaneously achieve imaging and target detection under strong clutter and strong noise conditions, showing superior performance in both aspects.
{"title":"A Learning Bayesian MAP Framework for Joint SAR Imaging and Target Detection","authors":"Hongyang An;Jianyu Yang;Yuping Xiao;Min Li;Haowen Zuo;Zhongyu Li;Wei Pu;Junjie Wu","doi":"10.1109/TRS.2024.3497057","DOIUrl":"https://doi.org/10.1109/TRS.2024.3497057","url":null,"abstract":"In synthetic aperture radar (SAR) information acquisition, target detection is often performed on the basis of the acquired radar images. Under low signal-to-clutter ratio (SCR) or low signal-to-noise ratio (SNR) conditions, detection by images is likely to cause loss of targets. To address this problem, we propose a joint imaging and target detection network based on Bayesian maximum a posteriori (MAP) estimation. The imaging and detection results are, respectively, defined as scene magnitude and detection label, and their joint probability distribution is used in place of the distribution of scene magnitudes. In the MAP estimation, the continuity feature of the detection label is merged into the optimization process, and the imaging and detection results are optimized alternately to get an iterative solution. The iterative solution is then unrolled into a network, which consists of three modules. We first utilize the unrolled fast iterative shrinkage thresholding algorithm (FISTA) method for the image formation module and then incorporate the detection label estimation module and distribution parameter updating module to learn the detection label and the function of distribution parameters. This approach applies prior information for both imaging and detection processes and enables automatic learning of parameters that are difficult to fit. Simulation experiments demonstrate that the method can simultaneously achieve imaging and target detection under strong clutter and strong noise conditions, showing superior performance in both aspects.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1214-1228"},"PeriodicalIF":0.0,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142797930","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}
Pub Date : 2024-11-08DOI: 10.1109/TRS.2024.3493992
Sen Yuan;Francesco Fioranelli;Alexander G. Yarovoy
The problem of 3-D high-resolution imaging in automotive multiple-input multiple-output (MIMO) side-looking radar using a 1-D array is considered. The concept of motion-enhanced snapshots is introduced to generate larger apertures in the azimuth dimension. For the first time, 3-D imaging capabilities can be achieved with high angular resolution using a 1-D MIMO antenna array, which can alleviate the requirement for large radar systems in autonomous vehicles. The robustness to variations in the vehicle’s movement trajectory is also considered and addressed with relevant compensations in the steering vector. The available degrees of freedom, as well as the signal-to-noise ratio (SNR), are shown to increase with the proposed method compared to conventional imaging approaches. The performance of the algorithm has been studied in simulations, and validated with experimental data collected in a realistic driving scenario.
{"title":"3-D High-Resolution Imaging Algorithm Using 1-D MIMO Array for Autonomous Driving Application","authors":"Sen Yuan;Francesco Fioranelli;Alexander G. Yarovoy","doi":"10.1109/TRS.2024.3493992","DOIUrl":"https://doi.org/10.1109/TRS.2024.3493992","url":null,"abstract":"The problem of 3-D high-resolution imaging in automotive multiple-input multiple-output (MIMO) side-looking radar using a 1-D array is considered. The concept of motion-enhanced snapshots is introduced to generate larger apertures in the azimuth dimension. For the first time, 3-D imaging capabilities can be achieved with high angular resolution using a 1-D MIMO antenna array, which can alleviate the requirement for large radar systems in autonomous vehicles. The robustness to variations in the vehicle’s movement trajectory is also considered and addressed with relevant compensations in the steering vector. The available degrees of freedom, as well as the signal-to-noise ratio (SNR), are shown to increase with the proposed method compared to conventional imaging approaches. The performance of the algorithm has been studied in simulations, and validated with experimental data collected in a realistic driving scenario.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1186-1199"},"PeriodicalIF":0.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142679409","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}
Tremor is one of the most prevalent movement disorders, which is especially observed in patients with Parkinson’s disease (PD) and other conditions common in palliative care (PC). Effective treatment and monitoring of disease progression are crucial in the context of PC for patients suffering from movement disorders. To this aim, accurate and continuous detection and assessment of tremor characteristics, such as the tremor frequency, is required. Current evaluations by clinicians conducted during sporadic consultations are subjective and intermittent. Radar sensors provide continuous, objective evaluations of tremor motion in patient monitoring, offering a contactless, light-independent, and privacy-preserving method that directly measures tremor’s radial motion through the Doppler effect. As previous radar-based research lacks continuous tremor monitoring in realistic scenarios, this study uses a frequency-modulated continuous-wave (FMCW) radar to detect subtle tremor motions and estimates their frequencies amid challenges such as large body motion interference in a clinical setting. Seventeen healthy participants were instructed to mimic tremors in their right hand while performing three diagnostics movements frequently used in tremor assessment, and two activities that were inspired by common daily tasks encountered in PC settings. Tremor detection and frequency estimation was enabled using suitable radar signal preprocessing followed by a neural network comprising convolutional and recurrent layers. Reference frequencies were obtained from an inertial measurement unit (IMU) attached to the participants’ right hands. Cross-validation revealed a mean absolute error (MAE) of 1.47 Hz in radar-based frequency estimation compared with the reference and a 90% accuracy in distinguishing the presence or absence of tremor, highlighting the proposed approach’s high potential for future tremor assessment.
震颤是最常见的运动障碍之一，尤其见于帕金森病（PD）患者和姑息治疗（PC）中常见的其他疾病。有效治疗和监测疾病进展对于运动障碍患者的姑息治疗至关重要。为此，需要对震颤特征（如震颤频率）进行准确、持续的检测和评估。目前临床医生在零星会诊时进行的评估是主观和间歇性的。雷达传感器可在患者监测过程中对震颤运动进行连续、客观的评估，它提供了一种非接触、不受光线影响、保护隐私的方法，可通过多普勒效应直接测量震颤的径向运动。由于以往基于雷达的研究缺乏在现实场景中的连续震颤监测，本研究使用频率调制连续波（FMCW）雷达来检测微妙的震颤运动，并估算其频率，以应对临床环境中的大型体动干扰等挑战。17 名健康的参与者在进行震颤评估中常用的三个诊断动作和两个受 PC 环境中常见日常任务启发的活动时，被要求模仿右手的震颤。震颤检测和频率估算是通过适当的雷达信号预处理，然后利用由卷积层和递归层组成的神经网络实现的。参考频率由连接在参与者右手上的惯性测量单元（IMU）获得。交叉验证显示，与参考频率相比，基于雷达的频率估计平均绝对误差（MAE）为 1.47 Hz，区分是否存在震颤的准确率为 90%，这凸显了所提出的方法在未来震颤评估中的巨大潜力。
{"title":"Radar-Based Tremor Quantification Using Deep Learning for Improved Parkinson’s and Palliative Care Assessment","authors":"Desar Mejdani;Johanna Bräunig;Stefan G. GrießHammer;Daniel Krauss;Tobias Steigleder;Lukas Engel;Jelena Jukic;Anna Rozhdestvenskaya;Jürgen Winkler;Bjoern Eskofier;Christoph Ostgathe;Martin Vossiek","doi":"10.1109/TRS.2024.3494473","DOIUrl":"https://doi.org/10.1109/TRS.2024.3494473","url":null,"abstract":"Tremor is one of the most prevalent movement disorders, which is especially observed in patients with Parkinson’s disease (PD) and other conditions common in palliative care (PC). Effective treatment and monitoring of disease progression are crucial in the context of PC for patients suffering from movement disorders. To this aim, accurate and continuous detection and assessment of tremor characteristics, such as the tremor frequency, is required. Current evaluations by clinicians conducted during sporadic consultations are subjective and intermittent. Radar sensors provide continuous, objective evaluations of tremor motion in patient monitoring, offering a contactless, light-independent, and privacy-preserving method that directly measures tremor’s radial motion through the Doppler effect. As previous radar-based research lacks continuous tremor monitoring in realistic scenarios, this study uses a frequency-modulated continuous-wave (FMCW) radar to detect subtle tremor motions and estimates their frequencies amid challenges such as large body motion interference in a clinical setting. Seventeen healthy participants were instructed to mimic tremors in their right hand while performing three diagnostics movements frequently used in tremor assessment, and two activities that were inspired by common daily tasks encountered in PC settings. Tremor detection and frequency estimation was enabled using suitable radar signal preprocessing followed by a neural network comprising convolutional and recurrent layers. Reference frequencies were obtained from an inertial measurement unit (IMU) attached to the participants’ right hands. Cross-validation revealed a mean absolute error (MAE) of 1.47 Hz in radar-based frequency estimation compared with the reference and a 90% accuracy in distinguishing the presence or absence of tremor, highlighting the proposed approach’s high potential for future tremor assessment.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1174-1185"},"PeriodicalIF":0.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672059","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}
Pub Date : 2024-11-06DOI: 10.1109/TRS.2024.3493037
Yi Li;Weijie Xia;Lingzhi Zhu;Jianjiang Zhou;Yongyan Chu;Wogong Zhang;Jie Zhang
In striving for optimal performance in distributed radar networks tailored for short-range applications, conventional direction-of-arrival (DOA) estimation often proves inadequate. The presence of close-in targets introduces a mismatch in the radar echo model, challenging the validity of far-field (FF) assumptions. To address this problem, we have developed a misspecified Cramér-Rao bound (MCRB) for DOA estimation in distributed radar networks influenced by near-field (NF) effects. The derivation aids in understanding potential performance degradations associated with the mean-squared error (mse) of a misspecified maximum-likelihood estimator. Through comprehensive analysis, we explore the interaction between the usual Cramér-Rao bound (CRB) and the MCRB. Moreover, we conduct a meticulous investigation into the relationship between these bounds, target parameters, and system architecture. Our examination significantly advances radar performance in practical scenarios, providing valuable insights to inform the design and configuration of distributed radar systems.
在为短程应用量身定制的分布式雷达网络中，传统的到达方向（DOA）估计往往无法达到最佳性能。近距离目标的存在给雷达回波模型带来了不匹配，对远场（FF）假设的有效性提出了挑战。为了解决这个问题，我们为受近场（NF）效应影响的分布式雷达网络中的 DOA 估测开发了一种误设克拉梅尔-拉奥约束（MCRB）。这一推导有助于理解与误设最大似然估计器的均方误差 (mse) 相关的潜在性能下降。通过综合分析，我们探讨了通常的克拉梅尔-拉奥约束（CRB）与 MCRB 之间的相互作用。此外，我们还对这些界限、目标参数和系统结构之间的关系进行了细致的研究。我们的研究大大提高了雷达在实际场景中的性能，为分布式雷达系统的设计和配置提供了有价值的见解。
{"title":"Performance Degradation of DOA Estimation in Distributed Radar Networks Under Near-Field Influence","authors":"Yi Li;Weijie Xia;Lingzhi Zhu;Jianjiang Zhou;Yongyan Chu;Wogong Zhang;Jie Zhang","doi":"10.1109/TRS.2024.3493037","DOIUrl":"https://doi.org/10.1109/TRS.2024.3493037","url":null,"abstract":"In striving for optimal performance in distributed radar networks tailored for short-range applications, conventional direction-of-arrival (DOA) estimation often proves inadequate. The presence of close-in targets introduces a mismatch in the radar echo model, challenging the validity of far-field (FF) assumptions. To address this problem, we have developed a misspecified Cramér-Rao bound (MCRB) for DOA estimation in distributed radar networks influenced by near-field (NF) effects. The derivation aids in understanding potential performance degradations associated with the mean-squared error (mse) of a misspecified maximum-likelihood estimator. Through comprehensive analysis, we explore the interaction between the usual Cramér-Rao bound (CRB) and the MCRB. Moreover, we conduct a meticulous investigation into the relationship between these bounds, target parameters, and system architecture. Our examination significantly advances radar performance in practical scenarios, providing valuable insights to inform the design and configuration of distributed radar systems.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1148-1159"},"PeriodicalIF":0.0,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142671121","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}
Most training set selection (TSS) methods are based on data processing methods. These methods have improved the state-of-the-art in clutter suppression under heterogeneous condition; however, TSS for heterogeneous and complex environments has rarely been investigated, especially for large outliers. This problem arises in situations such as isolated elevation points, spike effects of mountains, and urban-rural interfaces in actual radar operating environments. To address such a problem, this article proposes a novel enhanced outlier detection framework that deals with TSS in the presence of an unknown number of multiple outliers. First, the design of the overall structure of the TSS framework is proposed. We decompose the actual radar returns into four components and further integrate them into the TSS framework. The proposed framework uses the statistical characteristics of the returns from the range cells as a classification criterion. A deep neural network is devised to extract these statistical characteristics for outlier detection. The loss function and learning rate selection of the proposed TSS framework are, furthermore, specified. Then, the classification model for the four signal components is presented. To validate this framework, we use a real radar dataset sampled from heterogeneous environments and characterize signals in real radar scenarios. Experimental results demonstrate that the proposed framework significantly improves the accuracy of outlier detection in comparison with the traditional heterogeneous TSS method. In addition, our framework can further distinguish the interference outliers from the target echoes.
{"title":"Outlier Detection Enhancement in Heterogeneous Environments Through a Novel Training Set Selection Framework","authors":"Yongchan Gao;Kexuan Cui;Danilo Orlando;Chen Zhang;Guisheng Liao;Lei Zuo","doi":"10.1109/TRS.2024.3491795","DOIUrl":"https://doi.org/10.1109/TRS.2024.3491795","url":null,"abstract":"Most training set selection (TSS) methods are based on data processing methods. These methods have improved the state-of-the-art in clutter suppression under heterogeneous condition; however, TSS for heterogeneous and complex environments has rarely been investigated, especially for large outliers. This problem arises in situations such as isolated elevation points, spike effects of mountains, and urban-rural interfaces in actual radar operating environments. To address such a problem, this article proposes a novel enhanced outlier detection framework that deals with TSS in the presence of an unknown number of multiple outliers. First, the design of the overall structure of the TSS framework is proposed. We decompose the actual radar returns into four components and further integrate them into the TSS framework. The proposed framework uses the statistical characteristics of the returns from the range cells as a classification criterion. A deep neural network is devised to extract these statistical characteristics for outlier detection. The loss function and learning rate selection of the proposed TSS framework are, furthermore, specified. Then, the classification model for the four signal components is presented. To validate this framework, we use a real radar dataset sampled from heterogeneous environments and characterize signals in real radar scenarios. Experimental results demonstrate that the proposed framework significantly improves the accuracy of outlier detection in comparison with the traditional heterogeneous TSS method. In addition, our framework can further distinguish the interference outliers from the target echoes.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1160-1173"},"PeriodicalIF":0.0,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142671122","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}
Synthetic aperture radar (SAR) imaging provides a method for increasing the resolution of small and low-cost frequency-modulated continuous wave (FMCW) multiple-input multiple-output (MIMO) radar sensors. Using SAR images as an alternative to traditional point cloud-based representations of the environment may improve the performance of simultaneous localization and mapping (SLAM) algorithms for mobile robots. This article presents the details of an indoor mobile robot system that fuses inertial measurement unit (IMU) measurements and radar velocity estimates from an incoherent network of automotive radar sensors using an error-state Kalman filter (ESKF). This trajectory estimate is used to create surround-view SAR images of the robot’s operating environment. The obtained trajectory accuracy is compared against a laboratory reference system, and high-resolution SAR imaging results are presented. The measurement results provide insights into the challenges of robotic millimeter-wave imaging in indoor scenarios.
{"title":"RIO-SAR: Synthetic Aperture Radar Imaging of Indoor Scenes Based on Radar-Inertial Odometry Using a Mobile Robot","authors":"Yuma Elia Ritterbusch;Johannes Fink;Christian Waldschmidt","doi":"10.1109/TRS.2024.3488474","DOIUrl":"https://doi.org/10.1109/TRS.2024.3488474","url":null,"abstract":"Synthetic aperture radar (SAR) imaging provides a method for increasing the resolution of small and low-cost frequency-modulated continuous wave (FMCW) multiple-input multiple-output (MIMO) radar sensors. Using SAR images as an alternative to traditional point cloud-based representations of the environment may improve the performance of simultaneous localization and mapping (SLAM) algorithms for mobile robots. This article presents the details of an indoor mobile robot system that fuses inertial measurement unit (IMU) measurements and radar velocity estimates from an incoherent network of automotive radar sensors using an error-state Kalman filter (ESKF). This trajectory estimate is used to create surround-view SAR images of the robot’s operating environment. The obtained trajectory accuracy is compared against a laboratory reference system, and high-resolution SAR imaging results are presented. The measurement results provide insights into the challenges of robotic millimeter-wave imaging in indoor scenarios.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1200-1213"},"PeriodicalIF":0.0,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142777885","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}
Autonomous driving vehicles are being more and more popular in the community with the rise of artificial intelligence systems. However, in the context of airborne navigation, it remains a challenge, especially during landing maneuver. In order to operate in all conditions (weather, day, and night) and in all airports, we propose a runway localization method based on images acquired by an onboard radar. The proposed algorithm is a radar data segmentation method designed for use by an aircraft, as an on-board system, to provide the pilot, whether human or automatic, with a runway location prediction to facilitate and secure the landing maneuver. This article describes the acquisition and labeling of a large-scale real dataset over 18 airports in France and Switzerland, and the proposition of an attention-based deep recurrent neural network (RNN) for semantic segmentation of 4-D radar data acquired during a landing maneuver. This end-to-end trainable neural network combines attention mechanisms adapted to the geometry of an approach scene, with the exploitation of spatial-temporal information via recursive cells, all being associated with a convolutional segmentation model (patent pending). This article proposes a sensitivity analysis of Lyon’s airport to tune the hyperparameters, demonstrating the interest in adapting the attention sequence, especially through the shape of patches. The experimental results have shown the benefit of each block in the model. Extensive experiments on the other available airports have allowed validating the potential of the proposed network. Experiments have shown a considerable gain of about 0.17 on the DICE score associated with the exploitation of attention mechanisms and recursive cells and a gain of 0.1 compared to the SegFormer-B0 model.
{"title":"Attention-Based Deep Recurrent Neural Network for Semantic Segmentation of 4-D Radar Data Acquired During Landing Maneuver","authors":"Solène Vilfroy;Thierry Urruty;Philippe Carré;Jean-Philippe Lebrat;Lionel Bombrun","doi":"10.1109/TRS.2024.3488475","DOIUrl":"https://doi.org/10.1109/TRS.2024.3488475","url":null,"abstract":"Autonomous driving vehicles are being more and more popular in the community with the rise of artificial intelligence systems. However, in the context of airborne navigation, it remains a challenge, especially during landing maneuver. In order to operate in all conditions (weather, day, and night) and in all airports, we propose a runway localization method based on images acquired by an onboard radar. The proposed algorithm is a radar data segmentation method designed for use by an aircraft, as an on-board system, to provide the pilot, whether human or automatic, with a runway location prediction to facilitate and secure the landing maneuver. This article describes the acquisition and labeling of a large-scale real dataset over 18 airports in France and Switzerland, and the proposition of an attention-based deep recurrent neural network (RNN) for semantic segmentation of 4-D radar data acquired during a landing maneuver. This end-to-end trainable neural network combines attention mechanisms adapted to the geometry of an approach scene, with the exploitation of spatial-temporal information via recursive cells, all being associated with a convolutional segmentation model (patent pending). This article proposes a sensitivity analysis of Lyon’s airport to tune the hyperparameters, demonstrating the interest in adapting the attention sequence, especially through the shape of patches. The experimental results have shown the benefit of each block in the model. Extensive experiments on the other available airports have allowed validating the potential of the proposed network. Experiments have shown a considerable gain of about 0.17 on the DICE score associated with the exploitation of attention mechanisms and recursive cells and a gain of 0.1 compared to the SegFormer-B0 model.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1135-1147"},"PeriodicalIF":0.0,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636292","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}
Radio frequency (RF) sensing applications such as RF waveform classification and human activity recognition (HAR) demand real-time processing capabilities. Current state-of-the-art techniques often require a two-stage process for classification: first, computing a time-frequency (TF) transform, and then applying machine learning (ML) using the TF domain as the input for classification. This process hinders the opportunities for real-time classification. Consequently, there is a growing interest in direct classification from raw IQ-RF data streams. Applying existing deep learning (DL) techniques directly to the raw IQ radar data has shown limited accuracy for various applications. To address this, this article proposes to learn the parameters of structured functions as filterbanks within complex-valued (CV) neural network architectures. The initial layer of the proposed architecture features CV parameterized learnable filters (PLFs) that directly work on the raw data and generate frequency-related features based on the structured function of the filter. This work presents four different PLFs: Sinc, Gaussian, Gammatone, and Ricker functions, which demonstrate different types of frequency-domain bandpass filtering to show their effectiveness in RF data classification directly from raw IQ radar data. Learning structured filters also enhances interpretability and understanding of the network. The proposed approach was tested on both experimental and synthetic datasets for sign and modulation recognition. The PLF-based models achieved an average of 47% improvement in classification accuracy compared with a 1-D convolutional neural network (CNN) on raw RF data and an average 7% improvement over CNNs with real-valued learnable filters for the experimental dataset. It also matched the accuracy of a 2-D CNN applied to micro-Doppler ( $mu $ D) spectrograms while reducing computational latency by around 75%. These results demonstrate the potential of the proposed model for a range of RF sensing applications with enhanced accuracy and computational efficiency.
{"title":"PLFNets: Interpretable Complex-Valued Parameterized Learnable Filters for Computationally Efficient RF Classification","authors":"Sabyasachi Biswas;Cemre Omer Ayna;Ali Cafer Gurbuz","doi":"10.1109/TRS.2024.3486183","DOIUrl":"https://doi.org/10.1109/TRS.2024.3486183","url":null,"abstract":"Radio frequency (RF) sensing applications such as RF waveform classification and human activity recognition (HAR) demand real-time processing capabilities. Current state-of-the-art techniques often require a two-stage process for classification: first, computing a time-frequency (TF) transform, and then applying machine learning (ML) using the TF domain as the input for classification. This process hinders the opportunities for real-time classification. Consequently, there is a growing interest in direct classification from raw IQ-RF data streams. Applying existing deep learning (DL) techniques directly to the raw IQ radar data has shown limited accuracy for various applications. To address this, this article proposes to learn the parameters of structured functions as filterbanks within complex-valued (CV) neural network architectures. The initial layer of the proposed architecture features CV parameterized learnable filters (PLFs) that directly work on the raw data and generate frequency-related features based on the structured function of the filter. This work presents four different PLFs: Sinc, Gaussian, Gammatone, and Ricker functions, which demonstrate different types of frequency-domain bandpass filtering to show their effectiveness in RF data classification directly from raw IQ radar data. Learning structured filters also enhances interpretability and understanding of the network. The proposed approach was tested on both experimental and synthetic datasets for sign and modulation recognition. The PLF-based models achieved an average of 47% improvement in classification accuracy compared with a 1-D convolutional neural network (CNN) on raw RF data and an average 7% improvement over CNNs with real-valued learnable filters for the experimental dataset. It also matched the accuracy of a 2-D CNN applied to micro-Doppler (\u0000<inline-formula> <tex-math>$mu $ </tex-math></inline-formula>\u0000D) spectrograms while reducing computational latency by around 75%. These results demonstrate the potential of the proposed model for a range of RF sensing applications with enhanced accuracy and computational efficiency.","PeriodicalId":100645,"journal":{"name":"IEEE Transactions on Radar Systems","volume":"2 ","pages":"1102-1111"},"PeriodicalIF":0.0,"publicationDate":"2024-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142595067","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：