Submit Manuscript
Peer Review
Editor Work2022 Vol. 11, No. 4
Previous Issue
2022, 11(4): 517-529.
Abstract
1706KB
This paper proposes a novel task-driven flexible array beampattern synthesis model for self-organized drone swarms according to their task requirements so that they can adjust their positions appropriately and point in a specific direction. First, we formulate the novel beampattern synthesis model using the drone-swarm antenna position and weight vector as the optimization variables and the maximum driving distance as the constraint. Then, the Lawson criterion is used to simplify the objective function, and the two kinds of optimization variables of antenna position and weight vector are reduced to a single kind of variable optimization problem of antenna position, alleviating the optimization difficulty caused by the usage of coupled variables. Simultaneously, auxiliary variables are introduced to separate the constraints from the complex objective function, and the Alternating Direction Method of Multipliers (ADMM）is used to slove the problem, which reduces the difficulty of solving a highly nonlinear optimization problem with constraints,. In addition, we extend this method to a scenario in which the provided Direction Of Arrival (DOA）of interest is imprecise. Simulation results show that the proposed method can obtain lower sidelobe levels than previous methods.
This paper proposes a novel task-driven flexible array beampattern synthesis model for self-organized drone swarms according to their task requirements so that they can adjust their positions appropriately and point in a specific direction. First, we formulate the novel beampattern synthesis model using the drone-swarm antenna position and weight vector as the optimization variables and the maximum driving distance as the constraint. Then, the Lawson criterion is used to simplify the objective function, and the two kinds of optimization variables of antenna position and weight vector are reduced to a single kind of variable optimization problem of antenna position, alleviating the optimization difficulty caused by the usage of coupled variables. Simultaneously, auxiliary variables are introduced to separate the constraints from the complex objective function, and the Alternating Direction Method of Multipliers (ADMM）is used to slove the problem, which reduces the difficulty of solving a highly nonlinear optimization problem with constraints,. In addition, we extend this method to a scenario in which the provided Direction Of Arrival (DOA）of interest is imprecise. Simulation results show that the proposed method can obtain lower sidelobe levels than previous methods.
2022, 11(4): 530-542.
In this study, under the Peak-to-Average Power Ratio (PAPR), energy, and binary (for antenna position selection) constraints, we proposed an antenna position selection and beam scanning method for colocated Multiple-Input Multiple-Output (MIMO) radar system using the min-max beampattern amplitude matching criterion. In our design, antenna positions and a set of probing waveforms were jointly determined to match a set of beampattern masks, and hence realize the beam scan. The resultant problem was large-scale, nonconvex, nonsmooth, and typical nondeterministic hard, because of the PAPR and nonconvex binary constraints, and the max and modulus operations in the objective function. To address these issues, we first transformed the min-max problem into the Iterative weighted Least Squares (ILS) problem using the Lawson algorithm, replaced the nonsmooth nonconvex objective function with the convex majorization function, and finally applied the alternating direction method of multipliers to solve the majorized ILS problem. Finally, several numerical examples were given to show the effectiveness of the proposed algorithms.
In this study, under the Peak-to-Average Power Ratio (PAPR), energy, and binary (for antenna position selection) constraints, we proposed an antenna position selection and beam scanning method for colocated Multiple-Input Multiple-Output (MIMO) radar system using the min-max beampattern amplitude matching criterion. In our design, antenna positions and a set of probing waveforms were jointly determined to match a set of beampattern masks, and hence realize the beam scan. The resultant problem was large-scale, nonconvex, nonsmooth, and typical nondeterministic hard, because of the PAPR and nonconvex binary constraints, and the max and modulus operations in the objective function. To address these issues, we first transformed the min-max problem into the Iterative weighted Least Squares (ILS) problem using the Lawson algorithm, replaced the nonsmooth nonconvex objective function with the convex majorization function, and finally applied the alternating direction method of multipliers to solve the majorized ILS problem. Finally, several numerical examples were given to show the effectiveness of the proposed algorithms.
2022, 11(4): 543-556.
This paper proposes a joint design method to optimize the transmit waveforms and receive filter bank in Multiple-Input Multiple-Output (MIMO) structure ensuring spectral compatibility with the surrounding communication service network. Considering the signal-dependent clutter interference, under the constraints of transmission energy, waveform similarity and spectrum compatibility, the formulated optimization problem of the output Signal-to-Interference-plus-Noise Ratio (SINR) maximization is NP-hard. Toward this end, an auxiliary variable is first introduced to modify the original problem, and then a primal-dual algorithm based on the Alternating Block Successive Upper-bound Minimization (ABSUM) method is developed to deal with the resulting problem. Furthermore, an interior point method is used to handle the quadratic programming problem involved in each update procedure of the devised ABSUM method. Finally, numerical simulations are performed to demonstrate the superiority of the proposed method over state-of-the-art methods in terms of the optimized SINR, beampattern, computational complexity and ambiguity properties.
This paper proposes a joint design method to optimize the transmit waveforms and receive filter bank in Multiple-Input Multiple-Output (MIMO) structure ensuring spectral compatibility with the surrounding communication service network. Considering the signal-dependent clutter interference, under the constraints of transmission energy, waveform similarity and spectrum compatibility, the formulated optimization problem of the output Signal-to-Interference-plus-Noise Ratio (SINR) maximization is NP-hard. Toward this end, an auxiliary variable is first introduced to modify the original problem, and then a primal-dual algorithm based on the Alternating Block Successive Upper-bound Minimization (ABSUM) method is developed to deal with the resulting problem. Furthermore, an interior point method is used to handle the quadratic programming problem involved in each update procedure of the devised ABSUM method. Finally, numerical simulations are performed to demonstrate the superiority of the proposed method over state-of-the-art methods in terms of the optimized SINR, beampattern, computational complexity and ambiguity properties.
2022, 11(4): 557-569.
Outstanding beamforming performance of the Multiple-Input Multiple-Output (MIMO) radar can be achieved by deploying a large number of active antenna elements. Nonetheless, this will significantly increase power consumption, circuit complexity and hardware cost. These problems can be overcome by utilizing low-resolution Digital-to-Analog Converter (DAC) components. However, MIMO radar waveforms designed under the condition of infinite-resolution DACs are usually inapplicable to systems with low-resolution DACs. Therefore, under the constraints of discrete phases, this paper proposes a MIMO radar constant modulus waveform design method based on Integrated Sidelobe-to-Mainlobe Ratio (ISMR) minimization. The Dinkelbach algorithm is first used to convert the objective function with quadratic fractional form into a subtraction form. Then, the alternating direction penalty method is employed to solve the nonconvex constant modulus discrete phase constraint problem. Finally, by comparison with other methods through numerical simulations, the behavior of the transmit beampattern and the performance of ISMR are analyzed, and the effectiveness of the method is verified.
Outstanding beamforming performance of the Multiple-Input Multiple-Output (MIMO) radar can be achieved by deploying a large number of active antenna elements. Nonetheless, this will significantly increase power consumption, circuit complexity and hardware cost. These problems can be overcome by utilizing low-resolution Digital-to-Analog Converter (DAC) components. However, MIMO radar waveforms designed under the condition of infinite-resolution DACs are usually inapplicable to systems with low-resolution DACs. Therefore, under the constraints of discrete phases, this paper proposes a MIMO radar constant modulus waveform design method based on Integrated Sidelobe-to-Mainlobe Ratio (ISMR) minimization. The Dinkelbach algorithm is first used to convert the objective function with quadratic fractional form into a subtraction form. Then, the alternating direction penalty method is employed to solve the nonconvex constant modulus discrete phase constraint problem. Finally, by comparison with other methods through numerical simulations, the behavior of the transmit beampattern and the performance of ISMR are analyzed, and the effectiveness of the method is verified.
2022, 11(4): 570-580.
To prevent the degradation of the detection performance of Dual-Function Radar-Communication (DFRC) system in the presence of clutter, we propose the joint design of a transmit waveform and receiver filter to suppress the clutter and enhance the target detection performance. We use the Signal-to-Interference-plus-Noise Ratio (SINR) as the design criterion. Meanwhile, the Multi-User Interference (MUI) energy of the communication signals is constrained to maintain the quality of service for information transmission via DFRC systems. In addition, a similarity constraint is enforced to enable the transmitted waveform to have a good ambiguity function. To tackle the joint optimization problem, we present an iterative algorithm based on cyclic optimization and Semi-Definite Relaxation (SDR). The convergence of the algorithm is proved by a theoretical analysis. The simulation results show that the designed waveform can improve the target detection performance of a DFRC system in clutter and efficiently realize multi-user communication.
To prevent the degradation of the detection performance of Dual-Function Radar-Communication (DFRC) system in the presence of clutter, we propose the joint design of a transmit waveform and receiver filter to suppress the clutter and enhance the target detection performance. We use the Signal-to-Interference-plus-Noise Ratio (SINR) as the design criterion. Meanwhile, the Multi-User Interference (MUI) energy of the communication signals is constrained to maintain the quality of service for information transmission via DFRC systems. In addition, a similarity constraint is enforced to enable the transmitted waveform to have a good ambiguity function. To tackle the joint optimization problem, we present an iterative algorithm based on cyclic optimization and Semi-Definite Relaxation (SDR). The convergence of the algorithm is proved by a theoretical analysis. The simulation results show that the designed waveform can improve the target detection performance of a DFRC system in clutter and efficiently realize multi-user communication.
2022, 11(4): 581-599.
This paper releases a rotated SAR ship detection dataset, named Rotated Ship Detection Dataset in SAR Images (RSDD-SAR), to address the problem that the existing rotated SAR ship detection datasets are not enough to meet the requirements of algorithm development and practical application. This dataset consists of 84 scenes of GF-3 data slices, 41 scenes of TerraSAR-X data slices, and 2 scenes of large uncropped images, including 7,000 slices and 10,263 ship instances of multi-observing modes, multi-polarization modes, and multi-resolutions. This dataset is effectively annotated by automatic annotation with manual correction. Meanwhile, experiments were conducted for several popular rotated object detection algorithms in optical remote sensing images and rotated ship detection algorithms in SAR images, and the one-stage algorithm S2ANet achieved the highest average precision of 90.06%. When using this dataset, scholars can reference the experimental results, and corresponding analysis can be used. Finally, this paper conducts generalization ability testing experiments on other datasets and large uncropped images to analyze and discuss the performance of the model trained on RSDD-SAR. The experimental results show that the model trained on RSDD-SAR has decent performance and confirms the application value of this dataset. The RSDD-SAR dataset is available at https://radars.ac.cn/web/data/getData?dataType=SDD-SAR .
This paper releases a rotated SAR ship detection dataset, named Rotated Ship Detection Dataset in SAR Images (RSDD-SAR), to address the problem that the existing rotated SAR ship detection datasets are not enough to meet the requirements of algorithm development and practical application. This dataset consists of 84 scenes of GF-3 data slices, 41 scenes of TerraSAR-X data slices, and 2 scenes of large uncropped images, including 7,000 slices and 10,263 ship instances of multi-observing modes, multi-polarization modes, and multi-resolutions. This dataset is effectively annotated by automatic annotation with manual correction. Meanwhile, experiments were conducted for several popular rotated object detection algorithms in optical remote sensing images and rotated ship detection algorithms in SAR images, and the one-stage algorithm S2ANet achieved the highest average precision of 90.06%. When using this dataset, scholars can reference the experimental results, and corresponding analysis can be used. Finally, this paper conducts generalization ability testing experiments on other datasets and large uncropped images to analyze and discuss the performance of the model trained on RSDD-SAR. The experimental results show that the model trained on RSDD-SAR has decent performance and confirms the application value of this dataset. The RSDD-SAR dataset is available at https://radars.ac.cn/web/data/getData?dataType=SDD-SAR .
2022, 11(4): 600-617.
Polarimetric Interferometric Synthetic Aperture Radar (PolInSAR) simultaneously has interferometric height measurement and full-polarized detection capabilities, which can better reflect the structural properties of feature targets. Therefore, its potential for application in complex scenarios, such as urban areas, has attracted increasing attention. In urban areas, the processing mainly includes three modes: using interferometry to extract height based on polarimetric optimal coherence, using interferometry based on polarized decomposition, and associating polarimetric interferometric observation equations to retrieve the heights of different scattering mechanisms. The analysis of error factors and effects on Interferometric SAR (InSAR) and polarized SAR is almost complete, but the analysis of error effects under different processing modes of PolInSAR is insufficient. Based on the PolInSAR error model, our paper proposes a method for solving the scattering mechanism under the simultaneous polarization observation equation. Moreover, we derive the model including each error under different processing modes in PolInSAR from the aspect of polarized errors, interferometric errors, and the Signal-to-Noise Ratio (SNR). Furthermore, the model is verified through simulations, and we provide height inversion results through three processing modes after compensating for polarized errors and interferometric errors. After the error compensation, we obtain a Root Mean Squared Error (RMSE) in building areas of 2.77 m through polarimetric optimal coherence. Finally, the simulations provide the error impact curves under different processing modes of PolInSAR and compare the degree of different processing methods affected by errors, which provides a reasonable explanation for the design of the PolInSAR system, selection of processing methods, and data application.
Polarimetric Interferometric Synthetic Aperture Radar (PolInSAR) simultaneously has interferometric height measurement and full-polarized detection capabilities, which can better reflect the structural properties of feature targets. Therefore, its potential for application in complex scenarios, such as urban areas, has attracted increasing attention. In urban areas, the processing mainly includes three modes: using interferometry to extract height based on polarimetric optimal coherence, using interferometry based on polarized decomposition, and associating polarimetric interferometric observation equations to retrieve the heights of different scattering mechanisms. The analysis of error factors and effects on Interferometric SAR (InSAR) and polarized SAR is almost complete, but the analysis of error effects under different processing modes of PolInSAR is insufficient. Based on the PolInSAR error model, our paper proposes a method for solving the scattering mechanism under the simultaneous polarization observation equation. Moreover, we derive the model including each error under different processing modes in PolInSAR from the aspect of polarized errors, interferometric errors, and the Signal-to-Noise Ratio (SNR). Furthermore, the model is verified through simulations, and we provide height inversion results through three processing modes after compensating for polarized errors and interferometric errors. After the error compensation, we obtain a Root Mean Squared Error (RMSE) in building areas of 2.77 m through polarimetric optimal coherence. Finally, the simulations provide the error impact curves under different processing modes of PolInSAR and compare the degree of different processing methods affected by errors, which provides a reasonable explanation for the design of the PolInSAR system, selection of processing methods, and data application.
2022, 11(4): 618-636.
In the practical application of Tomographic Synthetic Aperture Radar (TomoSAR), the number of acquisitions is usually restricted due to their expensive cost. A coprime TomoSAR technique was proposed to reduce the required number of acquisitions by sparsely distributing acquisitions and elongating baseline aperture. To guarantee a reliable tomogram, this study aims to determine the minimum number of acquisitions for the coprime TomoSAR when adopting a subspace method for performing the tomographic reconstruction. However, the performance of the subspace method depends on multiple parameters. In light of this, the selection of acquisition times has to comprehensively weigh the effects of all these parameters on the reconstruction performance. To this end, a prerequisite for reliable reconstruction is established by quantifying the relationship between the sample eigenvalues and all related parameters. Compared with conventional estimation approaches for the minimum number of acquisitions, the proposed approach has twofold advantages of containing all related parameters and of having a closed-form expression. Finally, the simulation experiments verify that the number of acquisitions estimated by our approach is close to the minimum and can guarantee reconstruction reliability.
In the practical application of Tomographic Synthetic Aperture Radar (TomoSAR), the number of acquisitions is usually restricted due to their expensive cost. A coprime TomoSAR technique was proposed to reduce the required number of acquisitions by sparsely distributing acquisitions and elongating baseline aperture. To guarantee a reliable tomogram, this study aims to determine the minimum number of acquisitions for the coprime TomoSAR when adopting a subspace method for performing the tomographic reconstruction. However, the performance of the subspace method depends on multiple parameters. In light of this, the selection of acquisition times has to comprehensively weigh the effects of all these parameters on the reconstruction performance. To this end, a prerequisite for reliable reconstruction is established by quantifying the relationship between the sample eigenvalues and all related parameters. Compared with conventional estimation approaches for the minimum number of acquisitions, the proposed approach has twofold advantages of containing all related parameters and of having a closed-form expression. Finally, the simulation experiments verify that the number of acquisitions estimated by our approach is close to the minimum and can guarantee reconstruction reliability.
2022, 11(4): 637-651.
With the expansion of Synthetic Aperture Radar (SAR) applications and the development of SAR data acquisition technology, multiangle SAR datasets of various typical targets need to be constructed. Presently, a comprehensive multiangle SAR image dataset for aircraft targets is still lacking. This study explores a method of dataset construction based on the acquisition of actual data and intelligent simulation. Multiangle SAR images of aircraft targets are collected through flight tests, and the interpolation simulations of SAR images of specific angles are realized based on scattering analysis and self-attention generative adversarial network, which provide a new solution for dataset construction and expansion. Finally, under the assumption that some data are missing, the similarities between the simulated and actual images are evaluated using six evaluation indices, which verify the effectiveness of the proposed method.
With the expansion of Synthetic Aperture Radar (SAR) applications and the development of SAR data acquisition technology, multiangle SAR datasets of various typical targets need to be constructed. Presently, a comprehensive multiangle SAR image dataset for aircraft targets is still lacking. This study explores a method of dataset construction based on the acquisition of actual data and intelligent simulation. Multiangle SAR images of aircraft targets are collected through flight tests, and the interpolation simulations of SAR images of specific angles are realized based on scattering analysis and self-attention generative adversarial network, which provide a new solution for dataset construction and expansion. Finally, under the assumption that some data are missing, the similarities between the simulated and actual images are evaluated using six evaluation indices, which verify the effectiveness of the proposed method.
2022, 11(4): 652-665.
The sample scarcity issue is still challenged for SAR images interpretation. The number of geospatial targets related images is constrained of the SAR images interpretation ability of data acquisition, sample labeling, and the lack of target coverage. Our SAR-ATR method is demonstrated based on scattering information and meta-learning. First, the discrete distribution of the spatial structure of different types of aircraft is quite different in SAR images. An associated scattering classifier is designed to guide the network to learn more discriminative intra-class and inter-class feature descriptions. Our proposed classifier facilitates the modeling of discrete degree of the aircraft target quantitatively and balance the weights of sample pairs dynamically through the differentiated analysis of different target discrete distributions. In addition, an adaptive feature refinement module is designed to optimize the network cohesion for the key parts of the aircraft and reduce the interference of background noise. The proposed method integrates the target scattering distribution properties to the network learning process. On 5-way 1-shot emerging categorized recognition task involved only few samples, our experimental results demonstrate that the recognition accuracy of this method is 59.90%, which is 3.85% higher than the benchmark. After reducing the amount of training data by half, the proposed method is still competitive on the new category of few-shot recognition tasks.
The sample scarcity issue is still challenged for SAR images interpretation. The number of geospatial targets related images is constrained of the SAR images interpretation ability of data acquisition, sample labeling, and the lack of target coverage. Our SAR-ATR method is demonstrated based on scattering information and meta-learning. First, the discrete distribution of the spatial structure of different types of aircraft is quite different in SAR images. An associated scattering classifier is designed to guide the network to learn more discriminative intra-class and inter-class feature descriptions. Our proposed classifier facilitates the modeling of discrete degree of the aircraft target quantitatively and balance the weights of sample pairs dynamically through the differentiated analysis of different target discrete distributions. In addition, an adaptive feature refinement module is designed to optimize the network cohesion for the key parts of the aircraft and reduce the interference of background noise. The proposed method integrates the target scattering distribution properties to the network learning process. On 5-way 1-shot emerging categorized recognition task involved only few samples, our experimental results demonstrate that the recognition accuracy of this method is 59.90%, which is 3.85% higher than the benchmark. After reducing the amount of training data by half, the proposed method is still competitive on the new category of few-shot recognition tasks.
2022, 11(4): 666-675.
When Frequency Diverse Array and Multiple-Input Multiple-Output (FDA-MIMO) radar detects moving targets, the frequency offset between transmitting arrays is coupled with the target velocity, resulting in severe Doppler spread in the slow time dimension. Further, the signal energy of each receiving channel cannot be coherently accumulated, thereby greatly reducing the detection performance of the system. To address this problem, a method of FDA-MIMO radar moving target detection based on Doppler spread compensation is proposed. Firstly, the moving target echo model of FDA-MIMO radar is established, and the Doppler spread caused by the frequency offset is analyzed. After establishing the maximum likelihood receiver model, a resampling method based on an interpolation filter is proposed to compensate for the Doppler spread caused by FDA-MIMO radar during the detection of moving targets. Simulation results reveal that the proposed method can suppress the Doppler spread, compensate for the cross-cell migration of subtarget echo in the range dimension, and realize the coherent accumulation of signal energy.
When Frequency Diverse Array and Multiple-Input Multiple-Output (FDA-MIMO) radar detects moving targets, the frequency offset between transmitting arrays is coupled with the target velocity, resulting in severe Doppler spread in the slow time dimension. Further, the signal energy of each receiving channel cannot be coherently accumulated, thereby greatly reducing the detection performance of the system. To address this problem, a method of FDA-MIMO radar moving target detection based on Doppler spread compensation is proposed. Firstly, the moving target echo model of FDA-MIMO radar is established, and the Doppler spread caused by the frequency offset is analyzed. After establishing the maximum likelihood receiver model, a resampling method based on an interpolation filter is proposed to compensate for the Doppler spread caused by FDA-MIMO radar during the detection of moving targets. Simulation results reveal that the proposed method can suppress the Doppler spread, compensate for the cross-cell migration of subtarget echo in the range dimension, and realize the coherent accumulation of signal energy.
2022, 11(4): 676-691.
The Sparse Recovery Space-Time Adaptive Processing (SR-STAP) method can use a small number of training range cells to effectively suppress the clutter of airborne radar. The SR-STAP approach may successfully eliminate airborne radar clutter using a limited number of training range cells. However, present SR-STAP approaches are all model-driven, limiting their practical applicability due to parameter adjustment difficulties and high computational cost. To address these problems, this study, for the first time, introduces the Deep Unfolding/Unrolling (DU) method to airborne radar clutter reduction and target recognition by merging the model-driven SR method and the data-driven deep learning method. Firstly, a combined estimation model for clutter space-time spectrum and Array Error (AE) parameters is established and solved using the Alternating Direction Method of Multipliers (ADMM) algorithm. Secondly, the ADMM algorithm is unfolded to a deep neural network, named AE-ADMM-Net, to optimize all iteration parameters using a complete training dataset. Finally, the training range cell data is processed by the trained AE-ADMM-Net, jointly estimating the clutter space-time spectrum and the radar AE parameters efficiently and accurately. Simulation results show that the proposed DU-STAP method can achieve higher clutter suppression performance with lower computational cost compared to typical SR-STAP methods.
The Sparse Recovery Space-Time Adaptive Processing (SR-STAP) method can use a small number of training range cells to effectively suppress the clutter of airborne radar. The SR-STAP approach may successfully eliminate airborne radar clutter using a limited number of training range cells. However, present SR-STAP approaches are all model-driven, limiting their practical applicability due to parameter adjustment difficulties and high computational cost. To address these problems, this study, for the first time, introduces the Deep Unfolding/Unrolling (DU) method to airborne radar clutter reduction and target recognition by merging the model-driven SR method and the data-driven deep learning method. Firstly, a combined estimation model for clutter space-time spectrum and Array Error (AE) parameters is established and solved using the Alternating Direction Method of Multipliers (ADMM) algorithm. Secondly, the ADMM algorithm is unfolded to a deep neural network, named AE-ADMM-Net, to optimize all iteration parameters using a complete training dataset. Finally, the training range cell data is processed by the trained AE-ADMM-Net, jointly estimating the clutter space-time spectrum and the radar AE parameters efficiently and accurately. Simulation results show that the proposed DU-STAP method can achieve higher clutter suppression performance with lower computational cost compared to typical SR-STAP methods.
2022, 11(4): 692-704.
Signal Direct Position Determination (DPD) is a novel passive localization technology, which shows superior performance in terms of low signal noise rate adaptability and no parameter association necessity. To adapt to the complex electromagnetic environment, this study proposes a coprime array-based DPD method with single moving observation. Considering narrowband signals as an example, this study first formulates the intercepted signal model, then derives its equivalent model related to the corresponding difference co-array, and finally builds the DPD cost function via spatial spectrum technology. Simulation results show that the proposed method can greatly improve the degree of freedom compared to the traditional DPD with a minor loss of resolution and accuracy when an identical coprime array is used. Meanwhile, compared to the uniform linear array-based DPD, the proposed method shows superior performance in terms of the degree of freedom, resolution, and accuracy of localization.
Signal Direct Position Determination (DPD) is a novel passive localization technology, which shows superior performance in terms of low signal noise rate adaptability and no parameter association necessity. To adapt to the complex electromagnetic environment, this study proposes a coprime array-based DPD method with single moving observation. Considering narrowband signals as an example, this study first formulates the intercepted signal model, then derives its equivalent model related to the corresponding difference co-array, and finally builds the DPD cost function via spatial spectrum technology. Simulation results show that the proposed method can greatly improve the degree of freedom compared to the traditional DPD with a minor loss of resolution and accuracy when an identical coprime array is used. Meanwhile, compared to the uniform linear array-based DPD, the proposed method shows superior performance in terms of the degree of freedom, resolution, and accuracy of localization.
2022, 11(4): 705-712.
Cross-eye jamming is an electronic countermeasure technology used to cause angular deviation of monopulse radar via multiple coherent sources. Despite a complex electromagnetic environment, using active-passive composite monopulse radar is a development trend for improving the anti-interference ability of modern terminal guidance radar. A mathematical model of cross-eye jamming considering the active-passive composite monopulse radar is established. The influence mechanism of cross-eye jamming on the active-passive composite monopulse radar is revealed by comparing the effect of active and passive monopulse radar systems on angle measurement. Furthermore, the results can provide theoretical guidance and simulation data to reasonably apply Electronic Countermeasures (ECMs) and Electronic Counter-Countermeasures (ECCMs).
Cross-eye jamming is an electronic countermeasure technology used to cause angular deviation of monopulse radar via multiple coherent sources. Despite a complex electromagnetic environment, using active-passive composite monopulse radar is a development trend for improving the anti-interference ability of modern terminal guidance radar. A mathematical model of cross-eye jamming considering the active-passive composite monopulse radar is established. The influence mechanism of cross-eye jamming on the active-passive composite monopulse radar is revealed by comparing the effect of active and passive monopulse radar systems on angle measurement. Furthermore, the results can provide theoretical guidance and simulation data to reasonably apply Electronic Countermeasures (ECMs) and Electronic Counter-Countermeasures (ECCMs).
2022, 11(4): 713-727.
This paper proposes an SEI method based on cost-sensitive learning and semisupervised generative adversarial networks to address the problem of incomplete sample labels and imbalanced data category distribution in Specific Emitter Identification (SEI), which leads to a decline in inaccuracy. Through semisupervised training, the method optimizes the network parameters of the generator and discriminator, adds a multiscale topological block to ResNet to fuse the multi-dimensional resolution features of the time-domain signal, and attributes additional labels to the generated samples to directly use the discriminator to complete the classification. Simultaneously, a cost-sensitive loss is designed to alleviate the imbalance of gradient propagation caused by the dominant samples and improve the recognition performance of the classifier on the class-imbalanced dataset. The experimental results on four types of imbalanced datasets show that in the presence of 40% unlabeled samples, the average recognition accuracy for five emitters is improved by 5.34% and 2.69%, respectively, compared with the cross-entropy loss and focus loss. This provides a new idea for solving the problem of SEI under the conditions of insufficient data labels and an unbalanced distribution of data.
This paper proposes an SEI method based on cost-sensitive learning and semisupervised generative adversarial networks to address the problem of incomplete sample labels and imbalanced data category distribution in Specific Emitter Identification (SEI), which leads to a decline in inaccuracy. Through semisupervised training, the method optimizes the network parameters of the generator and discriminator, adds a multiscale topological block to ResNet to fuse the multi-dimensional resolution features of the time-domain signal, and attributes additional labels to the generated samples to directly use the discriminator to complete the classification. Simultaneously, a cost-sensitive loss is designed to alleviate the imbalance of gradient propagation caused by the dominant samples and improve the recognition performance of the classifier on the class-imbalanced dataset. The experimental results on four types of imbalanced datasets show that in the presence of 40% unlabeled samples, the average recognition accuracy for five emitters is improved by 5.34% and 2.69%, respectively, compared with the cross-entropy loss and focus loss. This provides a new idea for solving the problem of SEI under the conditions of insufficient data labels and an unbalanced distribution of data.
2022, 11(4): 728-735.
Combining Terahertz (THz) and Orbital Angular Momentum (OAM) technologies has great potential in high-speed wireless communication. Theoretically, OAM with different modes has strict orthogonality. The communication capacity of the system will improve significantly if OAM technology is applied to the THz communication system. Thus, the manner to generate a high-quality and dynamically controllable THz-OAM beam has been of significant interest to researchers in related fields. In this study, a double-layer transmissive metasurface that uses 3D printing as the processing method with a low cost and processing difficulty is designed. Note that the height of the unit cell for constructing the metasurface is configurable. As the height changes continuously, the phase of the transmitted wave covers 0～2\begin{document}${\pi }$\end{document} ![]()
![]()
\begin{document}$ l=1, \mathrm{2,3} $\end{document} ![]()
![]()
\begin{document}$ l=1,\mathrm{ }2,\mathrm{ }3 $\end{document} ![]()
![]()
\begin{document}${\pi }$\end{document} ![]()
![]()
\begin{document}$ l=1, \mathrm{2,3} $\end{document} ![]()
![]()
\begin{document}$ l=1,\mathrm{ }2,\mathrm{ }3 $\end{document} ![]()
![]()
微信 | 公众平台 
