Submit Manuscript
Peer Review
Editor Work2024 Vol. 13, No. 5
Previous Issue
2024, 13(5): 941-954.
Abstract
26021KB
Three-Dimensional (3D) Synthetic Aperture Radar (SAR) holds great potential for applications in fields such as mapping and disaster management, making it an important research focus in SAR technology. To advance the application and development of 3D SAR, especially by reducing the number of observations or antenna array elements, the Aerospace Information Research Institute, Chinese Academy of Sciences, (AIRCAS) has pioneered the development of the full-polarimetric Microwave Vision 3D SAR (MV3DSAR) experimental system. This system is designed to serve as an experimental platform and a source of data for microwave vision SAR 3D imaging studies. This study introduces the MV3DSAR experimental system along with its full-polarimetric SAR data set. It also proposes a set of full-polarimetric data processing scheme that covers essential steps such as polarization correction, polarization coherent enhancement, microwave vision 3D imaging, and 3D fusion visualization. The results from the 3D imaging data set confirm the full-polarimetric capabilities of the MV3DSAR experimental system and validate the effectiveness of the proposed processing method. The full-polarimetric unmanned aerial vehicle -borne array interferometric SAR data set, released through this study, offers enhanced data resources for advancing 3D SAR imaging research.
Three-Dimensional (3D) Synthetic Aperture Radar (SAR) holds great potential for applications in fields such as mapping and disaster management, making it an important research focus in SAR technology. To advance the application and development of 3D SAR, especially by reducing the number of observations or antenna array elements, the Aerospace Information Research Institute, Chinese Academy of Sciences, (AIRCAS) has pioneered the development of the full-polarimetric Microwave Vision 3D SAR (MV3DSAR) experimental system. This system is designed to serve as an experimental platform and a source of data for microwave vision SAR 3D imaging studies. This study introduces the MV3DSAR experimental system along with its full-polarimetric SAR data set. It also proposes a set of full-polarimetric data processing scheme that covers essential steps such as polarization correction, polarization coherent enhancement, microwave vision 3D imaging, and 3D fusion visualization. The results from the 3D imaging data set confirm the full-polarimetric capabilities of the MV3DSAR experimental system and validate the effectiveness of the proposed processing method. The full-polarimetric unmanned aerial vehicle -borne array interferometric SAR data set, released through this study, offers enhanced data resources for advancing 3D SAR imaging research.
2024, 13(5): 955-973.
Range Cell Migration Correction (RCMC) represents an important advancement in the estimation of moving target parameters and imaging of targets in high-resolution Synthetic Aperture Radar (SAR) systems. When the motion of a target or platform becomes complex, the traditional low-order RCMC method may no longer be suitable. Meanwhile, the existing high-order RCMC method based on parameterization is susceptible to issues such as model mismatch and high computational complexity. Additionally, its performance may decrease significantly under a low Signal-to-Noise Ratio (SNR). This research utilizes Extended Kalman Filter (EKF) to track the phase responsible for RCM and develop a phase compensation function to achieve RCMC. The proposed approach is model-independent and can track high-order components in the phase, thereby enabling high-order RCMC of moving targets in SAR. In addition, EKF can filter signals during phase tracking to effectively lower the SNR threshold of the proposed method. Thus, this method offers broad applicability, moderate computational complexity, and the ability to correct non-negligible high-order residual range cell migrations, thereby distinguishing it from traditional methods. This study thoroughly explains the principles and mathematical model behind the proposed method, demonstrating its effectiveness and superiority through multiple sets of simulations and measured data processing.
Range Cell Migration Correction (RCMC) represents an important advancement in the estimation of moving target parameters and imaging of targets in high-resolution Synthetic Aperture Radar (SAR) systems. When the motion of a target or platform becomes complex, the traditional low-order RCMC method may no longer be suitable. Meanwhile, the existing high-order RCMC method based on parameterization is susceptible to issues such as model mismatch and high computational complexity. Additionally, its performance may decrease significantly under a low Signal-to-Noise Ratio (SNR). This research utilizes Extended Kalman Filter (EKF) to track the phase responsible for RCM and develop a phase compensation function to achieve RCMC. The proposed approach is model-independent and can track high-order components in the phase, thereby enabling high-order RCMC of moving targets in SAR. In addition, EKF can filter signals during phase tracking to effectively lower the SNR threshold of the proposed method. Thus, this method offers broad applicability, moderate computational complexity, and the ability to correct non-negligible high-order residual range cell migrations, thereby distinguishing it from traditional methods. This study thoroughly explains the principles and mathematical model behind the proposed method, demonstrating its effectiveness and superiority through multiple sets of simulations and measured data processing.
2024, 13(5): 974-984.
The success of deep supervised learning in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) relies on a large number of labeled samples. However, label noise often exists in large-scale datasets, which highly influence network training. This study proposes loss curve fitting-based label noise uncertainty modeling and a noise uncertainty-based correction method. The loss curve is a discriminative feature to model label noise uncertainty using an unsupervised fuzzy clustering algorithm. Then, according to this uncertainty, the sample set is divided into different subsets: the noisy-label set, clean-label set, and fuzzy-label set, which are further used in training loss with different weights to correct label noise. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset prove that our method can deal with varying ratios of label noise during network training and correct label noise effectively. When the training dataset contains a small ratio of label noise (40%), the proposed method corrects 98.6% of these labels and trains the network with 98.7% classification accuracy. Even when the proportion of label noise is large (80%), the proposed method corrects 87.8% of label noise and trains the network with 82.3% classification accuracy.
The success of deep supervised learning in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) relies on a large number of labeled samples. However, label noise often exists in large-scale datasets, which highly influence network training. This study proposes loss curve fitting-based label noise uncertainty modeling and a noise uncertainty-based correction method. The loss curve is a discriminative feature to model label noise uncertainty using an unsupervised fuzzy clustering algorithm. Then, according to this uncertainty, the sample set is divided into different subsets: the noisy-label set, clean-label set, and fuzzy-label set, which are further used in training loss with different weights to correct label noise. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset prove that our method can deal with varying ratios of label noise during network training and correct label noise effectively. When the training dataset contains a small ratio of label noise (40%), the proposed method corrects 98.6% of these labels and trains the network with 98.7% classification accuracy. Even when the proportion of label noise is large (80%), the proposed method corrects 87.8% of label noise and trains the network with 82.3% classification accuracy.
2024, 13(5): 985-1003.
Spaceborne Synthetic Aperture Radar (SAR) systems are often subject to strong electromagnetic interference, resulting in imaging quality degradation. However, existing image domain-based interference suppression methods are prone to image distortion and loss of texture detail information, among other difficulties. To address these problems, this paper proposes a method for suppressing active suppression interferences inspaceborne SAR images based on perceptual learning of regional feature refinement. First, an active suppression interference signal and image model is established in the spaceborne SAR image domain. Second, a high-precision interference recognition network based on regional feature perception is designed to extract the active suppression interference pattern features of the involved SAR image using an efficient channel attention mechanism, consequently resulting in effective recognition of the interference region of the SAR image. Third, a multivariate regional feature refinement interference suppression network is constructed based on the joint learning of the SAR image and suppression interference features, which are combined to form the SAR image and suppression interference pattern. A feature refinement interference suppression network is then constructed based on the joint learning of the SAR image and suppression interference feature. The network slices the SAR image into multivariate regions, and adopts multi-module collaborative processing of suppression interference features on the multivariate regions to realize refined suppression of the active suppression interference of the SAR image under complex conditions. Finally, a simulation dataset of SAR image active suppression interference is constructed, and the evaluated Sentinel-1 data are used for experimental verification and analysis. The experimental results show that the proposed method can effectively recognize and suppress various typical active suppression interferences in spaceborne SAR images.
Spaceborne Synthetic Aperture Radar (SAR) systems are often subject to strong electromagnetic interference, resulting in imaging quality degradation. However, existing image domain-based interference suppression methods are prone to image distortion and loss of texture detail information, among other difficulties. To address these problems, this paper proposes a method for suppressing active suppression interferences inspaceborne SAR images based on perceptual learning of regional feature refinement. First, an active suppression interference signal and image model is established in the spaceborne SAR image domain. Second, a high-precision interference recognition network based on regional feature perception is designed to extract the active suppression interference pattern features of the involved SAR image using an efficient channel attention mechanism, consequently resulting in effective recognition of the interference region of the SAR image. Third, a multivariate regional feature refinement interference suppression network is constructed based on the joint learning of the SAR image and suppression interference features, which are combined to form the SAR image and suppression interference pattern. A feature refinement interference suppression network is then constructed based on the joint learning of the SAR image and suppression interference feature. The network slices the SAR image into multivariate regions, and adopts multi-module collaborative processing of suppression interference features on the multivariate regions to realize refined suppression of the active suppression interference of the SAR image under complex conditions. Finally, a simulation dataset of SAR image active suppression interference is constructed, and the evaluated Sentinel-1 data are used for experimental verification and analysis. The experimental results show that the proposed method can effectively recognize and suppress various typical active suppression interferences in spaceborne SAR images.
Multidomain Characteristic-guided Multimodal Contrastive Recognition Method for Active Radar Jamming
2024, 13(5): 1004-1018.
Achieving robust joint utilization of multidomain characteristics and deep-network features while maintaining a high jamming-recognition accuracy with limited samples is challenging. To address this issue, this paper proposes a multidomain characteristic-guided multimodal contrastive recognition method for active radar jamming. This method involves first thoroughly extracting the multidomain characteristics of active jamming and then designing an optimization unit to automatically select effective characteristics and generate a text modality imbued with implicit expert knowledge. The text modality and involved time-frequency transformation image are separately fed into text and image encoders to construct multimodal-feature pairs and map them to a high-dimensional space for modal alignment. The text features are used as anchors and a guide to time-frequency image features for aggregation around the anchors through contrastive learning, optimizing the image encoder’s representation capability, achieving tight intraclass and separated interclass distributions of active jamming. Experiments show that compared to existing methods, which involve directly combining multidomain characteristics and deep-network features, the proposed guided-joint method can achieve differential feature processing, thereby enhancing the discriminative and generalization capabilities of recognition features. Moreover, under extremely small-sample conditions (2～3 training samples for each type of jamming), the accuracy of our method is 9.84% higher than those of comparative methods, proving the effectiveness and robustness of the proposed method.
Achieving robust joint utilization of multidomain characteristics and deep-network features while maintaining a high jamming-recognition accuracy with limited samples is challenging. To address this issue, this paper proposes a multidomain characteristic-guided multimodal contrastive recognition method for active radar jamming. This method involves first thoroughly extracting the multidomain characteristics of active jamming and then designing an optimization unit to automatically select effective characteristics and generate a text modality imbued with implicit expert knowledge. The text modality and involved time-frequency transformation image are separately fed into text and image encoders to construct multimodal-feature pairs and map them to a high-dimensional space for modal alignment. The text features are used as anchors and a guide to time-frequency image features for aggregation around the anchors through contrastive learning, optimizing the image encoder’s representation capability, achieving tight intraclass and separated interclass distributions of active jamming. Experiments show that compared to existing methods, which involve directly combining multidomain characteristics and deep-network features, the proposed guided-joint method can achieve differential feature processing, thereby enhancing the discriminative and generalization capabilities of recognition features. Moreover, under extremely small-sample conditions (2～3 training samples for each type of jamming), the accuracy of our method is 9.84% higher than those of comparative methods, proving the effectiveness and robustness of the proposed method.
2024, 13(5): 1019-1036.
Interrupted Sampling Repeater Jamming (ISRJ) is a type of intrapulse coherent jamming that can form multiple realistic false targets that lead or lag behind the actual target, severely affecting radar detection. It is one of the hotspots of current research on electronic counter-countermeasures. To address this problem, an anti-ISRJ method based on an intrapulse frequency-coded joint Frequency Modulation (FM) slope agile waveform is proposed in this paper. In this method, the radar first transmits an intrapulse frequency-coded joint FM slope agile signal to improve the mutual coverability of subpulses by manipulating subpulse center frequency and FM slope agility. Next, the echo signal is divided into several slices according to the subpulse timing of the transmitted signal. Then, the Fuzzy C-Means (FCM) algorithm is used to classify the echo slices. Finally, the interference is suppressed via fractional-domain joint time domain filtering. Simulation results show that the FCM-based method can identify 100% of the interfered echo slices in a jammer synchronous sampling scenario when the Signal-to-Noise Ratio (SNR) is greater than −2.5 dB, and the Jamming-to-Signal Ratio (JSR) is greater than 5 dB. For high JSRs and low SNRs, the proposed method can effectively reduce the target energy loss and suppress the range sidelobes generated via residual interference. Moreover, the target detection probability after interference suppression exceeds 90% when JSR = 50 dB.
Interrupted Sampling Repeater Jamming (ISRJ) is a type of intrapulse coherent jamming that can form multiple realistic false targets that lead or lag behind the actual target, severely affecting radar detection. It is one of the hotspots of current research on electronic counter-countermeasures. To address this problem, an anti-ISRJ method based on an intrapulse frequency-coded joint Frequency Modulation (FM) slope agile waveform is proposed in this paper. In this method, the radar first transmits an intrapulse frequency-coded joint FM slope agile signal to improve the mutual coverability of subpulses by manipulating subpulse center frequency and FM slope agility. Next, the echo signal is divided into several slices according to the subpulse timing of the transmitted signal. Then, the Fuzzy C-Means (FCM) algorithm is used to classify the echo slices. Finally, the interference is suppressed via fractional-domain joint time domain filtering. Simulation results show that the FCM-based method can identify 100% of the interfered echo slices in a jammer synchronous sampling scenario when the Signal-to-Noise Ratio (SNR) is greater than −2.5 dB, and the Jamming-to-Signal Ratio (JSR) is greater than 5 dB. For high JSRs and low SNRs, the proposed method can effectively reduce the target energy loss and suppress the range sidelobes generated via residual interference. Moreover, the target detection probability after interference suppression exceeds 90% when JSR = 50 dB.
2024, 13(5): 1037-1048.
In the context of counter-reconnaissance against airborne interferometers, this study proposes a jamming method designed to disrupt the parameter measurement capabilities of interferometers by generating distributed signals based on an interrupted-sampling repeating technique. An emitter and a transmitting jammer are combined to form a distributed jamming system. The transmitting jammer samples the emitter signal and transmits the repeating signal to an interferometer. A quasi-synchronization constraint is established according to the change in the positional relation between the airborne interferometer and the jamming system. Additionally, a model for the superposition of distributed signals is provided. Then, the mathematical principle underlying distributed signal jamming is expounded according to the pulse spatial and temporal parameter measurement using the interferometer system. Moreover, the influence of various signal parameters on the jamming effect is analyzed to propose a principle for distributed signal design. Simulation and darkroom experiments show that the proposed method can effectively disrupt the accurate measurement of the pulse spatial domain and time domain parameters, such as azimuth-of-arrival, pulse width, and repetition interval.
In the context of counter-reconnaissance against airborne interferometers, this study proposes a jamming method designed to disrupt the parameter measurement capabilities of interferometers by generating distributed signals based on an interrupted-sampling repeating technique. An emitter and a transmitting jammer are combined to form a distributed jamming system. The transmitting jammer samples the emitter signal and transmits the repeating signal to an interferometer. A quasi-synchronization constraint is established according to the change in the positional relation between the airborne interferometer and the jamming system. Additionally, a model for the superposition of distributed signals is provided. Then, the mathematical principle underlying distributed signal jamming is expounded according to the pulse spatial and temporal parameter measurement using the interferometer system. Moreover, the influence of various signal parameters on the jamming effect is analyzed to propose a principle for distributed signal design. Simulation and darkroom experiments show that the proposed method can effectively disrupt the accurate measurement of the pulse spatial domain and time domain parameters, such as azimuth-of-arrival, pulse width, and repetition interval.
2024, 13(5): 1049-1060.
In multichannel adaptive radar target detection, diverse nonhomogeneous background factors can cause considerable outlier interference, making it challenging to meet the requirements of independent and identically distributed training data. Current methods for screening training data rely on prior knowledge of the number of outliers, often leading to poor performance in real-world scenarios where this number is usually unknown. This paper addresses these issues by focusing on adaptive training data screening when the number of outliers is unknown. First, the outlier set is estimated using maximum likelihood estimation, assuming known covariance matrices of clutter and noise. In particular, the training data is initially ranked based on the generalized inner product of each range cell data, approximately transforming the maximum likelihood estimation of the outlier set to the estimation of the number of outliers. Second, a fast maximum likelihood estimation algorithm is employed to calculate the unknown covariance matrix, and an adaptive screening approach is designed for scenarios with an unspecified number of outliers. Furthermore, to address the adverse effects of outliers on ranking performance, a normalized generalized inner product form is devised utilizing the normalized sampling covariance matrix. This form is subsequently incorporated into an iterative estimation procedure to improve the adaptive screening accuracy of training data. Simulation results demonstrate that the screening accuracy of the normalized generalized inner product exceeds that of the generalized inner product. Moreover, through even a small number of reiterations, maintaining a consistent enhancement in terms of the Normalized Signal-to-Interference Ratio (NSIR) is still possible. Compared with existing methods, the proposed algorithm considerably improves screening performance, especially when the number of outliers is unknown.
In multichannel adaptive radar target detection, diverse nonhomogeneous background factors can cause considerable outlier interference, making it challenging to meet the requirements of independent and identically distributed training data. Current methods for screening training data rely on prior knowledge of the number of outliers, often leading to poor performance in real-world scenarios where this number is usually unknown. This paper addresses these issues by focusing on adaptive training data screening when the number of outliers is unknown. First, the outlier set is estimated using maximum likelihood estimation, assuming known covariance matrices of clutter and noise. In particular, the training data is initially ranked based on the generalized inner product of each range cell data, approximately transforming the maximum likelihood estimation of the outlier set to the estimation of the number of outliers. Second, a fast maximum likelihood estimation algorithm is employed to calculate the unknown covariance matrix, and an adaptive screening approach is designed for scenarios with an unspecified number of outliers. Furthermore, to address the adverse effects of outliers on ranking performance, a normalized generalized inner product form is devised utilizing the normalized sampling covariance matrix. This form is subsequently incorporated into an iterative estimation procedure to improve the adaptive screening accuracy of training data. Simulation results demonstrate that the screening accuracy of the normalized generalized inner product exceeds that of the generalized inner product. Moreover, through even a small number of reiterations, maintaining a consistent enhancement in terms of the Normalized Signal-to-Interference Ratio (NSIR) is still possible. Compared with existing methods, the proposed algorithm considerably improves screening performance, especially when the number of outliers is unknown.
2024, 13(5): 1061-1072.
Airborne radar receivers that utilize subarray processing face challenges owing to the complex space-time coupling distribution caused by grating-lobe clutter. This results in multiple performance notches in the main beam, which severely affects target detection performance. To address this issue, we analyze the characteristics of grating-lobe clutter distribution in subarray processing and propose an approach for space-time clutter suppression based on the design of a receiving subarray beam pattern. Our approach leverages an overlapping subarray scheme to form wide nulls in the regions between subarrays where grating-lobe clutter is prevalent through beam pattern design. This design facilitates grating-lobe clutter pre-filtering between subarrays. Furthermore, we develop a subarray-level space-time processor that avoids the grating-lobe clutter coupling diffusion in the space-time two-dimensional plane by performing clutter pre-filtering within each subarray. This strategy enhances clutter suppression and moving-target-detection capabilities. Simulation results verify that the proposed method can remarkably improve the output signal to clutter plus noise ratio loss performance in grating-lobe clutter regions.
Airborne radar receivers that utilize subarray processing face challenges owing to the complex space-time coupling distribution caused by grating-lobe clutter. This results in multiple performance notches in the main beam, which severely affects target detection performance. To address this issue, we analyze the characteristics of grating-lobe clutter distribution in subarray processing and propose an approach for space-time clutter suppression based on the design of a receiving subarray beam pattern. Our approach leverages an overlapping subarray scheme to form wide nulls in the regions between subarrays where grating-lobe clutter is prevalent through beam pattern design. This design facilitates grating-lobe clutter pre-filtering between subarrays. Furthermore, we develop a subarray-level space-time processor that avoids the grating-lobe clutter coupling diffusion in the space-time two-dimensional plane by performing clutter pre-filtering within each subarray. This strategy enhances clutter suppression and moving-target-detection capabilities. Simulation results verify that the proposed method can remarkably improve the output signal to clutter plus noise ratio loss performance in grating-lobe clutter regions.
2024, 13(5): 1073-1091.
Real Aperture Radar (RAR) observes wide-scope target information by scanning its antenna. However, because of the limited antenna size, the angular resolution of RAR is much lower than the range resolution. Angular super-resolution methods can be applied to enhance the angular resolution of RAR by inverting the low-rank steering matrix based on the convolution relationship between the antenna pattern and target scatterings. Because of the low-rank characteristics of the antenna steering matrix, traditional angular super-resolution methods suffer from manual parameter selection and high computational complexity. In particular, these methods exhibit poor super-resolution angular resolution at low signal-to-noise ratios. To address these problems, an iterative adaptive approach for angular super-resolution imaging of scanning RAR is proposed by combining the traditional Iterative Adaptive Approach (IAA) with a deep network framework, namely IAA-Net. First, the angular super-resolution problem for RAR is transformed into an echo autocorrelation matrix inversion problem to mitigate the ill-posed condition of the inverse matrix. Second, a learnable repairing matrix is introduced into the IAA procedure to combine the IAA algorithm with the deep network framework. Finally, the echo autocorrelation matrix is updated via iterative learning to improve the angular resolution. Simulation and experimental results demonstrate that the proposed method avoids manual parameter selection and reduces computational complexity. The proposed method provides high angular resolution under a low signal-to-noise ratio because of the learning ability of the deep network.
Real Aperture Radar (RAR) observes wide-scope target information by scanning its antenna. However, because of the limited antenna size, the angular resolution of RAR is much lower than the range resolution. Angular super-resolution methods can be applied to enhance the angular resolution of RAR by inverting the low-rank steering matrix based on the convolution relationship between the antenna pattern and target scatterings. Because of the low-rank characteristics of the antenna steering matrix, traditional angular super-resolution methods suffer from manual parameter selection and high computational complexity. In particular, these methods exhibit poor super-resolution angular resolution at low signal-to-noise ratios. To address these problems, an iterative adaptive approach for angular super-resolution imaging of scanning RAR is proposed by combining the traditional Iterative Adaptive Approach (IAA) with a deep network framework, namely IAA-Net. First, the angular super-resolution problem for RAR is transformed into an echo autocorrelation matrix inversion problem to mitigate the ill-posed condition of the inverse matrix. Second, a learnable repairing matrix is introduced into the IAA procedure to combine the IAA algorithm with the deep network framework. Finally, the echo autocorrelation matrix is updated via iterative learning to improve the angular resolution. Simulation and experimental results demonstrate that the proposed method avoids manual parameter selection and reduces computational complexity. The proposed method provides high angular resolution under a low signal-to-noise ratio because of the learning ability of the deep network.
2024, 13(5): 1092-1108.
Due to the short wavelength of millimeter-wave, active electrical scanning millimeter-wave imaging system requires large imaging scenarios and high resolutions in practical applications. These requirements lead to a large uniform array size and high complexity of the feed network that satisfies the Nyquist sampling theorem. Accordingly, the system faces contradictions among imaging accuracy, imaging speed, and system cost. To this end, a novel, Credible Bayesian Inference of near-field Sparse Array Synthesis (CBI-SAS) algorithm is proposed under the framework of sparse Bayesian learning. The algorithm optimizes the complex-valued excitation weights based on Bayesian inference in a sparse manner. Therefore, it obtains the full statistical posterior Probability Density Function (PDF) of these weights. This enables the algorithm to utilize higher-order statistical information to obtain the optimal values, confidence intervals, and confidence levels of the excitation weights. In Bayesian inference, to achieve a small number of array elements to synthesize the desired beam orientation pattern, a heavy-tailed Laplace sparse prior is introduced to the excitation weights. However, considering that the prior probability model is not conjugated with the reference pattern data probability, the prior model is encoded in a hierarchical Bayesian manner so that the full posterior distribution can be represented in closed-form solutions. To avoid the high-dimensional integral in the full posterior distribution, a variational Bayesian expectation maximization method is employed to calculate the posterior PDF of the excitation weights, enabling reliable Bayesian inference. Simulation results show that compared with conventional sparse array synthesis algorithms, the proposed algorithm achieves lower element sparsity, a smaller normalized mean square error, and higher accuracy for matching the desired directional pattern. In addition, based on the measured raw data from near-field 1D electrical scanning and 2D plane electrical scanning, an improved 3D time domain algorithm is applied for 3D image reconstruction. Results verify that the proposed CBI-SAS algorithm can guarantee imaging results and reduce the complexity of the system.
Due to the short wavelength of millimeter-wave, active electrical scanning millimeter-wave imaging system requires large imaging scenarios and high resolutions in practical applications. These requirements lead to a large uniform array size and high complexity of the feed network that satisfies the Nyquist sampling theorem. Accordingly, the system faces contradictions among imaging accuracy, imaging speed, and system cost. To this end, a novel, Credible Bayesian Inference of near-field Sparse Array Synthesis (CBI-SAS) algorithm is proposed under the framework of sparse Bayesian learning. The algorithm optimizes the complex-valued excitation weights based on Bayesian inference in a sparse manner. Therefore, it obtains the full statistical posterior Probability Density Function (PDF) of these weights. This enables the algorithm to utilize higher-order statistical information to obtain the optimal values, confidence intervals, and confidence levels of the excitation weights. In Bayesian inference, to achieve a small number of array elements to synthesize the desired beam orientation pattern, a heavy-tailed Laplace sparse prior is introduced to the excitation weights. However, considering that the prior probability model is not conjugated with the reference pattern data probability, the prior model is encoded in a hierarchical Bayesian manner so that the full posterior distribution can be represented in closed-form solutions. To avoid the high-dimensional integral in the full posterior distribution, a variational Bayesian expectation maximization method is employed to calculate the posterior PDF of the excitation weights, enabling reliable Bayesian inference. Simulation results show that compared with conventional sparse array synthesis algorithms, the proposed algorithm achieves lower element sparsity, a smaller normalized mean square error, and higher accuracy for matching the desired directional pattern. In addition, based on the measured raw data from near-field 1D electrical scanning and 2D plane electrical scanning, an improved 3D time domain algorithm is applied for 3D image reconstruction. Results verify that the proposed CBI-SAS algorithm can guarantee imaging results and reduce the complexity of the system.
2024, 13(5): 1109-1122.
Vortex Electromagnetic Waves (VEMWs) have unique wavefront phase modulation characteristics. As a new degree of freedom in the diversity of radar transmitters, the VEMW Radar (VEMWR) provides Radar Cross-Section (RCS) diversity and improves signal and information processing dimensions and performances. The detection and imaging performances of VEMWR have been verified in various radar systems. This article focuses on the applying background of forward-looking radar imaging and proposes a time-division multiplemode scanning imaging method based on a Uniform Circular Array (UCA) system with multiple transmitters and a single receiver at the UCA center. First, we establish the forward-looking VEMWR imaging mode and corresponding signal mode. Next, an improved three-Dimensional (3D) back-projection and range-Doppler algorithm is proposed, which utilizes the magnitude difference at various elevation angles of multimode VEMW, phase difference at different azimuth angles, and Doppler effect resulting from the relative motion of the radar and target to achieve 3D imaging of the target. As the elevation angle increases, the beam pattern gain of the high-mode VEMW decreases sharply due to the energy divergence of the VEMW. The proposed method can maintain stability at low or high elevation angles using the energy distribution of multiple modes in the spatial domain. Imaging results of point targets revealed that the normalized gain of target-imaging results is equivalent either at low or high elevation angles within the multimode VEMW field of view. The proposed method is validated through experiments with an aircraft target. Based on the imaging results, it is verified that the proposed method can accurately reconstruct the 3D structure of complex targets.
Vortex Electromagnetic Waves (VEMWs) have unique wavefront phase modulation characteristics. As a new degree of freedom in the diversity of radar transmitters, the VEMW Radar (VEMWR) provides Radar Cross-Section (RCS) diversity and improves signal and information processing dimensions and performances. The detection and imaging performances of VEMWR have been verified in various radar systems. This article focuses on the applying background of forward-looking radar imaging and proposes a time-division multiplemode scanning imaging method based on a Uniform Circular Array (UCA) system with multiple transmitters and a single receiver at the UCA center. First, we establish the forward-looking VEMWR imaging mode and corresponding signal mode. Next, an improved three-Dimensional (3D) back-projection and range-Doppler algorithm is proposed, which utilizes the magnitude difference at various elevation angles of multimode VEMW, phase difference at different azimuth angles, and Doppler effect resulting from the relative motion of the radar and target to achieve 3D imaging of the target. As the elevation angle increases, the beam pattern gain of the high-mode VEMW decreases sharply due to the energy divergence of the VEMW. The proposed method can maintain stability at low or high elevation angles using the energy distribution of multiple modes in the spatial domain. Imaging results of point targets revealed that the normalized gain of target-imaging results is equivalent either at low or high elevation angles within the multimode VEMW field of view. The proposed method is validated through experiments with an aircraft target. Based on the imaging results, it is verified that the proposed method can accurately reconstruct the 3D structure of complex targets.
2024, 13(5): 1123-1133.
To improve the accuracy of Direction Of Arrival (DOA) estimation in Multiple Input Multiple Output (MIMO) radar systems under unknown mutual coupling, we propose a mutual coupling calibration and DOA estimation algorithm based on Sparse Learning via Iterative Minimization (SLIM). The proposed algorithm utilizes the spatial sparsity of target signals and estimates the spatial pseudo-spectra and the mutual coupling matrices of MIMO arrays through cyclic optimization. Moreover, it is hyperparameter-free and guarantees convergence. Numerical examples demonstrate that for MIMO radar systems under unknown mutual coupling conditions, the proposed algorithm can accurately estimate the DOA of targets with small angle separations and relatively high Signal-to-Noise Ratios (SNRs), even with a limited number of samples. In addition, low DOA estimation errors are achieved for targets with large angle separations and small sample sizes, even under low-SNR conditions.
To improve the accuracy of Direction Of Arrival (DOA) estimation in Multiple Input Multiple Output (MIMO) radar systems under unknown mutual coupling, we propose a mutual coupling calibration and DOA estimation algorithm based on Sparse Learning via Iterative Minimization (SLIM). The proposed algorithm utilizes the spatial sparsity of target signals and estimates the spatial pseudo-spectra and the mutual coupling matrices of MIMO arrays through cyclic optimization. Moreover, it is hyperparameter-free and guarantees convergence. Numerical examples demonstrate that for MIMO radar systems under unknown mutual coupling conditions, the proposed algorithm can accurately estimate the DOA of targets with small angle separations and relatively high Signal-to-Noise Ratios (SNRs), even with a limited number of samples. In addition, low DOA estimation errors are achieved for targets with large angle separations and small sample sizes, even under low-SNR conditions.
微信 | 公众平台 
