Article
Environmental Sciences
Wenzhe Li, Yanxin Yuan, Yuanpeng Zhang, Ying Luo
Summary: This paper proposes an unblurring ISAR imaging method based on an advanced Transformer structure for maneuvering targets. By generating pseudo-measured data and adopting a locally-enhanced window Transformer block to construct a Uformer-based GAN, deblurred ISAR images with rich details and textures can be obtained.
Article
Geochemistry & Geophysics
Yu Ying Dou, Yu Mao Wu, Han Qi Jin, Ya-Qiu Jin, Jun Hu, Jin Cheng
Summary: This study focuses on the sparse reconstruction of ISAR images using compressed sensing method. The results show that the L-1/2 regularization framework outperforms the L-1 regularization framework in terms of recovery accuracy and computational efficiency.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xinhang Zhu, Yicheng Jiang, Zitao Liu, Ruida Chen, Xin Qi
Summary: A novel ISAR imaging algorithm based on the cubic phase function and changing sampling rate (CPF-CSR) is proposed in this letter to eliminate high-order phase terms, resulting in a well-focused ISAR image.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Hongyin Shi, Long Zhang, Da Liu, Ting Yang, Zhijun Qiao
Summary: In this paper, a motion compensation-driven high-resolution imaging algorithm based on Fourier ptychographic microscopy (FPM) technology is proposed to address the issues of low resolution and poor noise suppression in maneuvering target imaging with traditional inverse synthetic aperture radar (ISAR) algorithms. The effectiveness of the proposed algorithm is verified through the use of simulation data and real radar data.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Haobo Wang, Yijun Chen, Yuanpeng Zhang, Ying Luo, Qun Zhang
Summary: Range instantaneous Doppler (RID) methods outperform traditional range-Doppler (RD) algorithms in imaging performance for maneuvering targets. However, traditional RID methods suffer from noise and sparse aperture, resulting in low resolution and degraded quality. To address this issue, the study proposes an RID high-resolution (HR) imaging framework that incorporates a hybrid transformer-based HR network (HTHRNet) into the RID method. HTHRNet is trained to map low-resolution time-frequency distribution (TFD) images to high-resolution TFD images under different conditions. Experimental results demonstrate the superior performance of the proposed method, suggesting the effective application of complex deep neural networks in ISAR HR imaging.
IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION
(2023)
Article
Engineering, Electrical & Electronic
Penghui Huang, Muyang Zhan, Zhiwei Yang, Xingzhao Liu, Guisheng Liao, Zhiling Liu, Wentao Du
Summary: This paper proposes a novel ISAR imaging algorithm using RID technique to achieve high-resolution imaging for maneuvering targets with complex phase fluctuation. The algorithm includes motion compensation and high-resolution time-frequency domain transformation based on TFD, which effectively solves the performance degradation and resolution limitations of traditional methods.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2021)
Article
Geochemistry & Geophysics
Shunjun Wei, Jiadian Liang, Mou Wang, Jun Shi, Xiaoling Zhang, Jinhe Ran
Summary: A novel compressive sensing-based imaging and autofocusing framework, AF-AMP, is proposed to tackle ISAR imaging and autofocusing challenges under sparse aperture conditions. By incorporating phase error estimation into the CS framework and iteratively solving the compound CS problem with approximate message-passing, the approach offers well-focused imaging results. Furthermore, a deep learning-based AF-AMPNet is introduced to enhance efficiency and performance, demonstrating superior results in simulated and measured experiments compared to other methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Fengkai Liu, Darong Huang, Xinrong Guo, Cunqian Feng
Summary: This paper proposes an effective method for translational motion compensation of maneuvering targets with sparse aperture. The method models the translational and rotational motions as polynomials and uses optimization and gradient descent algorithms to estimate parameters and achieve compensation. Experimental results demonstrate the effectiveness and robustness of the proposed method in various scenarios.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Chen Chen, Sirui Tian, Zhiyong Xu
Summary: In this letter, an efficient ISAR motion compensation method is proposed based on fast parameter estimation. The image entropy is derived using a compensation matrix and sinc function interpolation to eliminate the high-order phase term, transforming the parameter estimation into an optimization problem. The gradient descent algorithm (GDA) is employed to accelerate the computation speed. Our method outperforms other recently proposed methods in terms of robustness and computing cost.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Geochemistry & Geophysics
Shuai Shao, Lei Zhang, Hongwei Liu, Penghui Wang, Qianqian Chen
Summary: This article addresses the impact of noncooperative targets' 3-D rotational motion on InISAR imaging, as well as the challenges of fine image registration and sparse reconstruction with SFB-SA signals. A novel SVWPD signal model and JWPDC algorithm are proposed for high-quality InISAR image reconstruction. The JMC-2-D-JSR ISAR imaging algorithm based on the BCS theory outperforms other frameworks in 2-D imaging, 3-D imaging, and motion compensation.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Bangjie Zhang, Gang Xu, Hanwen Yu, Hui Wang, Hao Pei, Wei Hong
Summary: This article proposes a novel robust gridless compressed sensing (RGLCS) algorithm for high-resolution 3-D imaging. The algorithm uses atomic norm minimization to model the joint-sparsity pattern on elevation distribution between adjacent pixels, and models outliers and disturbances as sparsely distributed spike noise in the image domain. Experimental results validate the effectiveness of the proposed algorithm.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Zitao Liu, Yicheng Jiang, Yong Wang, Yuhan Du, Jinxiang Wang
Summary: This article introduces a novel ISAR technique, the Range Instantaneous Doppler Derivative (RIDD) algorithm, which can obtain two different types of well-focused ISAR images. In order to improve the accuracy of target classification and recognition, a new imaging and scaling approach is proposed, and its effectiveness is demonstrated through experiments.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xuejun Huang, Jinshan Ding, Zhong Xu
Summary: This letter presents an unsupervised CNN-based framework for super-resolution ISAR imaging, which can directly produce high-resolution ISAR images in real time and is suitable for practical applications.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Penghui Huang, Xiang-Gen Xia, Muyang Zhan, Xingzhao Liu, Guisheng Liao, Xue Jiang
Summary: A novel ISAR imaging algorithm is proposed in this article for maneuvering targets with moderate reflection intensity, effectively handling the impact of target motion on ISAR image quality, and the effectiveness and superiority of the algorithm are validated through the imaging results of simulated and real measured data.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Pei Ye, Meng-Dao Xing, Xiang-Gen Xia, Guang-Cai Sun, Yachao Li, Yuexin Gao
Summary: The paper introduces a new method for reconstructing high-resolution ISAR images using Kalman filtering, which corrects the state vector through a two-step KF process of prediction and update to achieve a well-focused high-resolution image in a short observation time. The proposed method demonstrates good performance in both simulated and real data, addressing the conflict between short observation time and high resolution requirements.
Article
Geochemistry & Geophysics
Jiyu Guo, Weike Feng, Jean Michel Friedt, Qing Zhao, Motoyuki Sato
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2020)
Article
Geochemistry & Geophysics
Suyun Wang, Weike Feng, Kazutaka Kikuta, Motoyuki Sato
Summary: A ground-based synthetic aperture radar system using an optical electric field sensor as the bistatic receiving unit has been designed for target imaging and displacement estimation purposes. The system is capable of acquiring monostatic and bistatic SAR images simultaneously from different angles, with displacement estimation accuracies reaching up to millimeter level in both x and y directions.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Environmental Sciences
Xiaowei Hu, Weike Feng, Yiduo Guo, Qiang Wang
Summary: This paper proposes a deep generation and recognition model based on the CVAE-GAN model to address the issues of transparency and unknown target classes in SAR-ATR. The feature space allows for generating clear SAR images and classifying targets.
Article
Geochemistry & Geophysics
Xiaowei Hu, Feng Xu, Yiduo Guo, Weike Feng, Ya-Qiu Jin
Summary: Microwave imaging with large rotating angle and sparse sampling is challenging for traditional imaging methods, but a new model-driven learning imaging network (MDLI-Net) is proposed in this article to efficiently output high-resolution target images. The network utilizes electromagnetic scattering model to generate training data and sparse microwave imaging theory to guide the network design, showing effective performance in both simulated and real data experiments.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Shanshan Ding, Ningning Tong, Weike Feng, Pengcheng Wan, Yongshun Zhang
Summary: A time-frequency decoupling algorithm based on matrix enhancement and matrix pencil (MEMP) is proposed for high-resolution 3-D imaging using wideband two-dimensional MIMO radar. This algorithm eliminates the time-frequency coupling term and improves radar imaging accuracy.
IET RADAR SONAR AND NAVIGATION
(2022)
Article
Environmental Sciences
Weike Feng, Jean-Michel Friedt, Pengcheng Wan
Summary: In this study, a fixed-receiver mobile-transmitter passive bistatic synthetic aperture radar (MF-PB-SAR) system was developed using embedded software-defined radio (SDR) hardware and utilizing the Sentinel-1 SAR satellite as the emitting source. It enables high-resolution imaging of targets in a local area.
Article
Engineering, Electrical & Electronic
Chunguang Lu, Weike Feng, Wenling Li, Yongshun Zhang, Yiduo Guo
Summary: This paper proposes a Kullback-Leibler average (KLA) based adaptive interacting multiple model (IMM) filter for jump Markov systems with inaccurate noise covariances and missing measurements. The switching error model (SEM) is constructed to accurately model the probability density function of measurement likelihood and a variational Bayesian (VB) technique is employed to jointly estimate the state, noise covariances, and binary indicator. The adaptive IMM filter is derived by using the KLA fusion scheme to combine the conditioned estimates from every mode. Simulation results demonstrate the superior performance of the proposed filter compared to existing typical algorithms.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Environmental Sciences
Tao Pu, Ningning Tong, Weike Feng, Pengcheng Wan, Xiaowei Hu
Summary: In this study, a sparse recovery imaging method with wideband interference (WBI) prediction based on the predictive recurrent neural network (PredRNN) and tensor-based smooth L0 (TSL0) algorithm is proposed. The method extracts the time-frequency (TF) feature of historical measured WBI and uses PredRNN to predict the future TF feature. It then adaptively designs random sparse stepped frequency waveform based on the predicted WBI TF feature and reconstructs the 3D high-resolution target image using the TSL0 algorithm.
Article
Environmental Sciences
Bo Zou, Xin Wang, Weike Feng, Hangui Zhu, Fuyu Lu
Summary: In this paper, a space-time two-dimensional (2D)-decoupled SR network, called 2DMA-Net, is constructed to achieve a fast clutter spectrum estimation without complicated parameter tuning. Through unsupervised training methods, the method overcomes the interferences caused by non-ideal factors. Simulation results demonstrate that the proposed method can simultaneously improve clutter suppression performance and reduce computational complexity.
Article
Environmental Sciences
Weike Feng, Pengcheng Wan, Xiaowei Hu, Yiduo Guo, Hangui Zhu
Summary: This paper proposes a cognitive method to achieve 3D high-resolution target imaging with reduced sampling cost in the presence of wideband interference. Through sparse sampling, interference removal, and a smoothed L0 algorithm, a high-resolution 3D target image is obtained. The cognitive sparse imaging loop for MIMO radar under wideband interference situations is formed, demonstrating the effectiveness and advantage of the proposed methods through simulation and experiment results.
Article
Geochemistry & Geophysics
Peng Li, Xiaowei Hu, Cunqian Feng, Xiaozhen Shi, Yiduo Guo, Weike Feng
Summary: Although DNNs have achieved good results in SAR automatic target recognition (ATR), their opacity and difficulty in interpretation limit practical application. Many interpretable models have been proposed, but they only represent transparent network structures and cannot be considered as real interpretable models. We propose SAR-AD-BagNet, a new SAR image recognition model based on adversarial defense, with a transparent decision-making process and a reasonable basis for decision-making. Moreover, the model exhibits high SAR image recognition accuracy and strong adversarial robustness.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Environmental Sciences
Hangui Zhu, Weike Feng, Cunqian Feng, Teng Ma, Bo Zou
Summary: This paper proposes two deep unfolded gridless DOA estimation networks to solve the problem that traditional methods cannot handle. By training these networks, higher DOA estimation performance can be achieved with lower computational expenditure.
Article
Geochemistry & Geophysics
Xixi Chen, Hao Wu, Yongqiang Cheng, Weike Feng, Yifan Guo
Summary: This study focuses on the joint design of transmit sequence and receive filter for MIMO radar to improve target detection performance in non-Gaussian backgrounds. By approximating the probability density function of observed non-Gaussian data with GMM, a Riemannian manifold of Gaussian mixture distribution is developed. The proposed method, which maximizes the geometric distance on manifolds, shows advantages in detection performance compared with competitive methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Pengcheng Wan, Weike Feng, Ningning Tong, Xiaowei Hu, Guimei Zheng
Summary: A deep learning method is proposed in this study to predict the time-frequency feature of wideband interference (WBI) and applied to cognitive radar high-resolution range profile (HRRP) estimation. The spatiotemporal correlation of the WBI time-frequency figures is learned using a long short-time memory (LSTM) network, reducing the influence of WBI on target estimation.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Electrical & Electronic
Weike Feng, Jean-Michel Friedt, Pengcheng Wan
Summary: In this article, general software-defined radio hardware is used for ground-based interferometric radar system development to achieve static target imaging and displacement estimation. The proposed system synchronization approach, frequency-domain bandwidth synthesis method, and data preprocessing techniques have successfully improved range resolution and enhanced target image quality. Various experiments demonstrate the high-resolution target image and accurate displacement measurement capabilities of the developed SDR-GBIR systems.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Zou Deyue, Gao Siyu, Li Xinyue, Zhao Wanlong
Summary: In this paper, an integrated navigation/communication signal is proposed by combining Cyclic Code Shift Keying (CCSK) as a communication signal with traditional Global Navigation Satellite System (GNSS) signal. A fuzzy judging algorithm and segmenting correlation domain are used to improve the transmission robustness of the integrated signal. Additionally, a strategy is proposed to solve the incompatibility problem between the integrated signal and binary communication system using interleaving coding and dual-thresholds optimization. The proposed algorithm significantly improves the fault-tolerant rate and reduces the Bit Error Rate (BER), enhancing the reliability of the communication system.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Guoxing Huang, Yunfei Xiang, Shibiao Deng, Yu Zhang, Jingwen Wang
Summary: This paper proposes a dual-channel cooperative sub-Nyquist sampling system for LFM-BPSK hybrid modulated signal. The system can solve the frequency ambiguity problem caused by low sampling rate and estimate the phases of binary symbols and the positions of discontinuities through self-mixing technology.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
A. Yousefian Darani, Y. Khedmati Yengejeh, G. Navarro, H. Pakmanesh, J. Sharafi
Summary: The development and use of network technology have brought society into the internet world, where communication through digital data takes place. However, the protection of digital resources has become a significant challenge, which steganography is seen as a possible solution for. This paper presents an image steganography algorithm that hides a grayscale secret image within an RGB color cover image, with a genetic algorithm used to optimize the selection of suitable hiding places. The proposed method is empirically demonstrated to be effective in terms of transparency, security, and resistance.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Xiao Tan, Zhiwei Yang, Xianghai Li, Yongfei Mao, Di Jiang
Summary: A novel gridless SR-STAP algorithm applicable to a non-ULA with AAPEs is proposed in this study. By establishing the ANM-STAP model and iteratively updating the clutter subspace and AAPEs, the performance of STAP is improved.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Irfan Manisali, Okyanus Oral, Figen S. Oktem
Summary: In this paper, a novel non-iterative deep learning-based reconstruction method for real-time near-field MIMO imaging is proposed. The method achieves high image quality with low computational cost at compressive settings through two stages of processing. A large-scale dataset is also developed for training the neural networks. The effectiveness of the method is demonstrated through experimental data and extensive simulations.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Haibin Li, Dengchao Wu, Wenming Zhang, Cunjun Xiao
Summary: Workplace safety accidents are a pervasive issue worldwide, with 67.95% of construction accidents attributed to the lack of helmet-wearing. Existing helmet detection algorithms suffer from underperformance in real-world scenarios due to various challenges. This study introduces a lightweight helmet detection algorithm, YOLO-PL, which achieves state-of-the-art performance by optimizing network structure and incorporating lightweight modules.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Fen Liu, Jianfeng Chen, Kemeng Li, Jisheng Bai, Weijie Tan, Chang Cai, Muhammad Saad Ayub
Summary: In this paper, a semi-tensor product-based multi-modal factorized multilinear (STP-MFM) pooling method is proposed for information fusion in sentiment analysis. Experimental results demonstrate that the proposed method outperforms baselines in terms of accuracy, training speed, and model complexity.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Xiaoyu Zhao, Xiaoqian Chen, Zhiqiang Gong, Wen Yao, Yunyang Zhang, Xiaohu Zheng
Summary: In this paper, a contrastive enhancement approach is proposed to mine latent prototypes from background regions and leverage latent classes to improve the utilization of similarity information between prototype and query features. The proposed modules, including a latent prototype sampling module and a contrastive enhancement module, significantly improve the performance of state-of-the-art methods for few-shot segmentation.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Xiaoyong Lyu, Baojin Liu, Wenbing Fan, Zhi Quan
Summary: This paper provides an in-depth analysis of the channel estimate model (CEM) in passive radar using orthogonal frequency division multiplexing (OFDM) waveforms. The authors propose a new CEM that takes into consideration the influence of inter-carriers interference (ICI). The theoretical analysis and simulation results support the effectiveness of the new CEM and the proposed method for canceling ICI.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Ruisheng Rana, Jinping Wang, Bin Fang, Weiming Yang
Summary: This paper introduces an improved neighborhood preserving embedding method (NPEAE), which utilizes a linear autoencoder to achieve more accurate and effective data projection from high-dimensional space to low-dimensional space. NPEAE performs better in recognition accuracy compared to other comparative methods.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Yuehao Guo, Xianpeng Wang, Jinmei Shi, Lu Sun, Xiang Lan
Summary: This article presents a fast real-valued tensor propagator method for FDA-MIMO radar. The method improves estimation accuracy by utilizing the original structural information of multidimensional data and eliminates the high computational complexity of high-order singular value decomposition. The proposed algorithm achieves parameter estimation at low snapshots and has lower computational complexity than other algorithms at high snapshots.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Bang Huang, Wen-Qin Wang, Weijian Liu, Shunsheng Zhang, Yizhen Jia
Summary: This paper studies the robust moving target detection problem in FDA-MIMO radar with an unknown covariance matrix in Gaussian clutter. The proposed approach adopts the subspace method and proposes three robust adaptive detectors, which are validated to be effective.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Asutosh Kar
Summary: In this paper, a class of variable step size adaptive algorithms is developed for hybrid narrow-band active noise control (HNANC) systems and compared with the existing state-of-the-art methods. Two algorithms are proposed for HNANC systems operating in multiple noise environments, and their performance is analyzed. The results demonstrate significant improvement in noise reduction compared to the counterparts.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Shouqi Wang, Zhigang Feng
Summary: With the rapid development of artificial intelligence and sensor technology, intelligent fault diagnosis methods based on deep learning are widely used in industrial production. In practical applications, complex noise environments and huge model parameters affect the performance and cost-effectiveness of diagnostic models. In this paper, a lightweight intelligent fault diagnosis model using multi-sensor data fusion is proposed to address these issues. By processing vibration signals from different sensors of rolling bearings using variational mode decomposition (VMD) and designing unique grayscale feature maps based on each intrinsic modal function (IMF) component, the proposed model achieves high accuracy diagnosis in noisy environments while meeting the requirements of small, light, and fast production. The ultra-lightweight GoogLeNet model (ULGoogLeNet) is constructed by adjusting the traditional GoogLeNet structure, and the ultra-lightweight subspace attention module (ULSAM) is introduced to reduce model parameters and enhance feature extraction capability. Experimental results on two datasets demonstrate the effectiveness and superiority of the proposed method.
DIGITAL SIGNAL PROCESSING
(2024)
Article
Engineering, Electrical & Electronic
Jichen Yang, Fangfan Chen, Yu Cheng, Pei Lin
Summary: Recently, there has been significant attention on multi-speaker multimedia speaker recognition (MMSR). This study explores innovative techniques for integrating audio and visual cues from the front-end representations of both speaker's voice and face. Experimental results show that these techniques achieve considerable improvements in the performance of MMSR.
DIGITAL SIGNAL PROCESSING
(2024)