Search

In this article, a regression neural network is proposed instead of the commonly used classification neural network to address the direction of arrival (DOA) estimation problem. Unlike the method treating the DOA estimation problem as a classification task, our proposed method utilizes a regression neural network to directly estimate the signal angles by mapping the input data to the high-dimensional features. It is an end-to-end method where the input data are the signal covariance matrix estimated from the received signals by the uniform linear array (ULA). The network output consists of two branches: one branch provides the estimated angles of signals, and the other branch provides the confidences corresponding to angles, indicating the probabilities that signals are incident at the estimated angles. Owing to the dual-branch output of our proposed method, it is adept at accommodating a variable number of sources without the requirement to alter the network structure. Compared to classification neural networks, our proposed method circumvents the issue of quantization errors and has the capability to handle the imbalance of positive and negative samples. Simulation results validate the superior performance of our proposed method in terms of estimation success rate, angular resolution, and estimation accuracy, especially in scenarios with a variable and unknown number of coherent signals.

For radar target tracking with nonlinear position and range rate measurements, a debiased measurement matrix-based linear sequential filtering method with an adaptive parameter is proposed. A concise linear measurement equation with an adaptive parameter is constructed based on the relationship between the pseudomeasurement and target's state vector. For the involved unknown parameters in the measurement matrix, the filtering results from the best linear unbiased estimator are utilized to estimate in the sequential linear filter, where the measurement matrix estimation error is fully considered and is incorporated with the original converted pseudomeasurement error to form a synthetical measurement error. The statistical characteristics of it are deduced based on which the measurement matrix is debiased. The adaptive parameter in the linear measurement matrix is optimized to maximize the information gain in the sequential filtering. Simulation results demonstrate the effectiveness of the proposed algorithm and its extension to the maneuvering target tracking case. Compared with existing algorithms, the proposed ones can achieve the best tracking performance in different scenarios.

Adaptive monopulse algorithm always faces low angle measurement performance in multi-interference environments with low signal-to-noise ratio. To clarify the impact factors, we analyze the performance of monopulse ratio estimation and further derive the Cramer-Rao bound (CRB) for angle estimation. Based on the analysis, a constraint-relaxed adaptive monopulse (RCAM) optimization problem is proposed, and the closed-form solution of the problem is obtained by alternative-direction-method-of-multipliers (ADMM) with fast convergence. Numerical simulations have been provided to demonstrate the effectiveness of the proposed algorithm in improving beamforming and angle estimation.

Radio frequency-based (RF-based) detection methods are currently the main means of countering drones. However, these prevalent approaches frequently exhibit deficiencies in effectively addressing noise and interference, making them potentially unsuitable for application in realistic urban environments. This letter proposes a generalized RF signal-enhanced framework that explicitly addresses noise and interference. We decompose the RF signal into three components and uniformly integrate them into the proposed framework for decomposition. To accomplish this, three innovative loss functions and two appropriate neural networks are devised. To validate our framework, we create a real-world drone RF dataset sampled from urban surroundings, faithfully representing drone RF signals in real-world scenarios. Experimental results demonstrate that our framework exhibits satisfactory denoising and interference-removal performance, significantly improving the accuracy of multiple detection methods.

This article addresses the problem of subspace detection in the compound Gaussian (CG) sea clutter with lognormal (LN) texture and structured interference. We proposed three novel subspace detectors by two-step maximum a posteriori (MAP) generalized likelihood ratio test (GLRT), the Rao test, and the Wald test. In the first step, we assume that the texture component and the speckle covariance matrix (CM) are known, and we derive the test statistics of the proposed detectors. Then, in the second step, we substitute the estimated texture component and speckle CM to obtain the adaptive detectors. Furthermore, we exploit the persymmetric property of the speckle CM to improve the detection performance of the proposed detectors. Moreover, we prove the constant false alarm rate (CFAR) properties of the novel subspace detectors with respect to the speckle CM and the scale parameter of the texture component of the non-Gaussian sea clutter. Besides, we verify the detection performance of the proposed subspace detectors by numerical experiments in both simulated and measured sea clutter. The simulation results show that the novel subspace detectors perform better than the comparison detectors in the case of limited training data, mismatched signals, and structured interference.

Compared with monostatic phased array radar, a colocated multiple-input–multiple-output (MIMO) radar network (CMRN) that owns richer controllable resources is able to offer superior tracking performance. In this article, for targets tracking in a cluttered environment with the CMRN, a joint detection threshold optimization and transmit resource allocation (JDTO-TRA) algorithm is proposed to simultaneously maximize the tracking performance and minimize the resource consumption, whose key mechanism is to adaptively control the working parameters at both transmitters and detectors via optimization technique. To be more specific, we first formulate the JDTO-TRA strategy as a biobjective optimization problem, where both the tracking performance and the resource consumption are considered in the objective function. The analytical expression for the posterior Cramér–Rao lower bound incorporated with information reduction factor, which is caused by the measurement origin uncertainty, is given and adopted as the tracking performance metric, and the total transmit energy of the CMRN is employed as the resource consumption metric. Hereafter, to tackle the formulated mixed-integer and nonconvex problem effectively, an iterative and efficient two-step solution technique incorporating the simulated-annealing-based hybrid particle swarm optimization and the cyclic minimization algorithm is proposed, where the radar-target assignment, the subarray number and the transmit energy of each activated colocated MIMO radar, and the false alarm rate of the target can be controlled jointly and adaptively. Numerical simulation results are provided to demonstrate the effectiveness as well as the advantages of the JDTO-TRA algorithm compared with other popular existing algorithms.

To enhance the coexistence performance of multiple-input multiple-output (MIMO) radar system and multiuser MIMO communication system with hybrid beamforming architecture, this work proposes a codesign problem regarding the digital and analog beamformers of communication base station, the receive beamformers of downlink users, and the transmit waveform of radar. An information-theoretic criterion is adopted to improve the radar detection performance, while the signal-to-interference-plus-noise ratio requirement is imposed on each downlink user to ensure communication quality. In addition, similarity restriction is also involved to manage some significant radar waveform characteristics. An iterative algorithm is introduced to tackle the resulted nonconvex and NP-hard problem. In each iteration, several subproblems are developed to decompose the original problem and are solved by the lower bound maximization and the alternating direction method of multiplier algorithm. Numerical simulations show that the proposed algorithm allows the objective function to increase monotonically and finally converge to a finite value. It also shows that the codesign scheme can effectively optimize the radar detection performance while ensuring the achievable sum rate and the bit error rate of the communication system.

Visual object tracking for unmanned aerial vehicles (UAVs) is extensively used in civil and military applications, such as intelligent transportation, disaster rescue, and military reconnaissance. However, when a UAV encounters haze weather, its tracking performance deteriorates greatly because of problems, such as blurring and quality degradation of the collected images. Most trackers are designed to operate in good weather conditions, and this greatly limits the robustness of UAV tracking in haze weather. To solve this problem, this article proposes an adaptive dehazing Siamese network, which consists of an adaptive dehazing module, a dynamic template update branch, and feature extraction and bounding box prediction subnetworks. The dehazing module yields haze-free images and preserves the edge information of the object, so that the tracker can extract the object features with better characterization. Subsequently, to mitigate the risk of tracking drift caused by the changing appearance of the object, the dynamic template update branch dynamically selects more reliable template images during tracking. In addition, to solve the problem of insufficient UAV object tracking testing datasets in haze scenes, a haze testing dataset for UAV video object tracking is obtained through artificial synthesis and data collection in real-world haze scenes. Furthermore, experiments are conducted on haze testing datasets, namely, UAV123-AH, UAV20L-AH, UAVDT-AH, DTB70-AH, and UAVhaze; the results verified that the proposed tracker yields more competitive performance and more robustness in haze scenes compared with the existing algorithms.

In this letter, we study the Hybrid Beamforming (HBF) design for a Dual-Function Radar-Communication (DFRC) system, which serves Multiple Users (MUs) and detects a target in the presence of signal-dependent clutters, simultaneously. Unlike conventional beamforming strategies, we propose a novel one on the symbol level, which exploits Constructive Interference (CI) to achieve a trade-off between radar and communication using one platform. To implement this novel strategy, we jointly design the DFRC transmit HBF and radar receive beamforming by maximizing the radar Signal to Interference plus Noise Ratio (SINR) while ensuring the Quality of Service (QoS) of downlink communication. To tackle the formulated non-convex problem, we propose an iterative algorithm, which combines the Majorization-Minimization (MM) and Alternating Direction Method of Multipliers (ADMM) judiciously. The numerical experiments indicate that our algorithm yields the CI properly for robust communications and achieves better performance than the conventional HBF benchmarks in both communication bit error rate and radar SINR.

Compared with the traditional frost space-time beamformer (STB) with a pre-steering delay processing (PSDP), the method without the PSDP is able to provide a more intuitive observation of the space-frequency response (SFR) for both the desired signal and interference in the beam pattern. In this paper, a new SFR constraint is proposed to guarantee the same time delay for all different frequency components in the desired signal by analyzing the desired signal process of the digital STB. Specifically, in order to make all frequencies in the desired bandwidth meet the SFR constraint, we propose a method to obtain the number of the sample frequency points of the new constraint. After that, the equivalent beamformer with lower complexity is acquired via transforming the SFR constraints into a part of the optimization objectives. On this basis, a robust wideband beamformer against the direction of arrival (DOA) mismatch of the desired signal is established. To be more concretely, in the proposed robust beamformer, the covariance matrix of the interference plus noise and the covariance matrix of the desired signal are reconstructed with the observed data containing the desired signal. Additionally, the reconstructed two matrices are used to estimate the actual desired signal steering vector by considering a maximization of the desired signal output power. Numerical results verify the effectiveness of those proposed algorithms. [ABSTRACT FROM AUTHOR]

The geostationary weather radar reflectivity data of low horizontal resolution would limit its application in the observation of small-scale weather system and increase the errors of rainfall retrieval, therefore, the capability to enhance the horizontal resolution of reflectivity data is of special interest. In this study, a new super-resolution model based on the idea of sub-division in one resolution volume for geostationary weather radar and an oversampling technique along the radar’s spiral scan track are presented. The reconstruction is physically based on the occurrence of multiple partially correlated measurements. Mathematically, the process is equivalent to a linear inversion of ill-posed problem, and its solution is pursued by means of the Tikhonov and truncated singular value decomposition (TSVD) inversion method. Experiment results show that the proposed model and reconstruction process are efficient for the horizontal resolution improvement of reflectivity data, and more fine detail could be present through reconstruction.

A Velocity Measurement Method Based on Scaling Parameter Estimation of a Chaotic SystemIn this paper, we propose a new method of measuring the target velocity by estimating the scaling parameter of a chaos-generating system. First, we derive the relation between the target velocity and the scaling parameter of the chaos-generating system. Then a new method for scaling parameter estimation of the chaotic system is proposed by exploiting the chaotic synchronization property. Finally, numerical simulations show the effectiveness of the proposed method in target velocity measurement.

Abstract: A multiple-input multiple-output (MIMO) radar uses multiple antennas to simultaneously transmit multiple independent probing signals, and uses multiple antennas to receive the backscattered signals. The modeling of MIMO radar with transmit spatial diversity and coherent reception is addressed herein. The maximum likelihood (ML) method for parameter estimation using MIMO radars is considered, and two approximate ML algorithms are proposed. In the uniform noise scenario, one of the proposed algorithms performs similarly to the delay-and-sum beamformer which is optimal in the ML sense in single target case, while it outperforms the other proposed approximate ML algorithm at the cost of more computational load. In the non-uniform noise scenario, the proposed approximate ML algorithms both outperform the delay-and-sum beamformer. The efficiency of the proposed methods is validated by the simulation results.

This paper design waveform for high frequency radar to optimize its performances in co-channel interference/colored noise environment. Considering that radar applies the optimal filter for range processing, a novel objective function accounting for both Signal to Interference plus Noise Ratio (SINR) and Integrated Sidelobes Level (ISL) is introduced and the constant modulus constraint on the probing waveform is forced. To resolve the resulting non-convex quartic optimization problem, an efficient iterative algorithm based on Majorization-Minimization (MM) and Pattern Search (PS) is proposed. In each iteration, the proposed algorithm produces a multi-dimension quadratic majorized problem and then splits it into multiple one-dimension problems with closed-form solutions. Simulation results show that the SINR and ISL performances of designed waveform can be flexibly adjusted. Besides, the convergence speed and stability of the proposed algorithm are more satisfactory than the other methods. [ABSTRACT FROM AUTHOR]

Adaptive dwell scheduling is a key for simultaneous multi-beam radar system to give full play to its tactical performance. The dwell scheduling for such radar system based on the selection of array elements with different polarization characteristics is first investigated in this manuscript. The optimal dwell scheduling model whose goal is to achieve the maximal scheduling gain for the system is established under the time, array element, frequency and electromagnetic compatibility constraints of the system. The solution of this optimization problem results in our proposed dwell scheduling algorithm, where the dwell tasks are selected to be analyzed sequentially according to synthetic priority at each scheduling analysis time point, and the element sets are chosen for dwell tasks under all kinds of constraints to realize the simultaneous execution of multiple tasks. Simulation results demonstrate that the proposed algorithm can realize real-time dwell scheduling and achieve better scheduling performance for the multi-beam radar system compared with existing algorithm. [ABSTRACT FROM AUTHOR]

The interrupted sampling repeater jamming (ISRJ) is a coherent interference, and it will result in a large amount of intra-pulse false targets in pulse compression (PC) radar. If the intercepted slices of the ISRJ, named as the screening waveform (SW), are orthogonal to the un-intercepted transmit waveform, named as the detection waveform (DW), the ISRJ can be suppressed by filtering with DW. In this paper, we investigate the problem of ISRJ suppression by jointly designing the DW and SW. Based on the inaccuracy of ISRJ parameter estimation, the joint design problem is formulated by minimizing the integrated sidelobe level (ISL) of the cross-correlation between the ISRJ and DW. The resulting design problem, which involves constraints of constant modulus and orthogonality, is NP-hard, and an alternating optimization like (AltOpt-like) algorithm is developed. Concretely, the SW is first obtained with the aid of the block coordinate descent (BCD) method, then the DW is optimized based on the alternating direction method of multipliers (ADMM) framework. Moreover, in the ADMM framework, low complexity algorithms are devised to solve the subproblems of the primal variables. Numerical simulations are provided to demonstrate the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

In a synthetic aperture radar system, long-time coherent integration for ground moving targets suffers from the complex range migrations (RMs) and Doppler frequency migration (DFM), which lead to a serious degradation of the target detection. To address these problems and obtain the motion parameters of the moving target, a new algorithm based on modified location rotation transform (MLRT) and improved discrete chirp Fourier transform (IDCFT), i.e., MLRT-IDCFT, is proposed in this paper. More concretely, the matched filtering function is first utilized to remove the range curvature (RC) for improving the computational efficiency. After that, the MLRT-IDCFT corrects the residual linear range migration (LRM) through rotating coordinate locations of the echo trajectory and achieves the coherent integration as well as the motion parameter estimation via the IDCFT. The IDCFT exploits the prior information to adaptively adjust the parameter estimation range, which not only improves the parameter resolution but also guarantees the validity of the parameter estimation. Due to the linear transform property, the proposed method can work well under the low signal-to-noise ratio (SNR) environment. Additionally, it has a relatively low computational complexity since the exhaustive searching for acceleration is avoided. Several simulations are provided to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

In this paper, two low computational complexity accurate array response control algorithms for large-scale antenna array are presented. The first proposed algorithm is an Improved Weight vector ORthogonal Decomposition (I-WORD) approach. We extend the WORD method by selecting the non-negative coefficient as the ultimate solution, and thus the ultimate weight vector will be in an exact form. Therefore, the proposed I-WORD algorithm has no longer a selection procedure and has low computational complexity compared with the WORD algorithm. Moreover, to achieve multi-point accurate array response control, we further develop a Multi-point control based on I-WORD (MI-WORD) algorithm. The MI-WORD algorithm is able to control multi-point response simultaneously by finding the weight vector from the intersection of weight vector sets with a new matrix constructing manner, which is different from the multi-point accurate array response control (MA2RC) method. Both the proposed algorithms can be applied to adjust the response accurately to synthesize the beampattern. Furthermore, the proposed MI-WORD method has the advantage of low computational complexity, especially when the number of array antennas is large. Simulation results show the effectiveness of the two algorithms and the property of low complexity of the MI-WORD algorithm for beampattern synthesis. [ABSTRACT FROM AUTHOR]

It is a common requirement for over-the-horizon (OTH) radar to detect the sea and airborne targets in the same direction simultaneously. The long coherent integration time (CIT) is needed to detect the sea target and the weak moving target such as aircraft. However, the range walk effect (RWE) will occur within the long integration time, leading to a serious degradation of the airborne target detection and tracking. To settle this problem, a new approach based on the multi-channel track-before-detect (MC-TBD) is proposed in this paper. More specifically, to improve the detection performance of the airborne target with unknown speed, multiple processing channels corresponding to different speed ranges are set under the MC-TBD framework. The received data are processed in each channel in parallel. In each channel, the constraints of the speed range are utilized to determine the number of pulses in the CIT of the airborne target. Then, coherent and noncoherent integration processing are performed sequentially based on the moving target detection (MTD) and track-before-detect (TBD). Finally, the detection results and trajectories of all channels are fused to achieve the final detection and tracking of the airborne target. Simulation results show the superiority of the proposed MC-TBD approach over the existing methods in terms of the performance of detection and tracking for airborne target with OTH radar under the background of simultaneous detection of airborne and sea targets in the same direction. [ABSTRACT FROM AUTHOR]

This paper investigates the classification and localization of the mixed far-field sources (FFSs) and near-field sources (NFSs) on a symmetrical uniform linear array and proposes a new mixed source localization algorithm. Firstly, we analyze the persymmetric structure property of the far-field (FF) steering vector in matrix differencing. Based on that, we implement the matrix auto-differencing operation on the fourth-order cumulant (FOC) matrix, which eliminates the contribution of the FF signals from the mixed signals. Based on the reserved near-field (NF) parts, an effective subspace-based is performed to estimate NF parameters. Secondly, to reconstruct the NF FOC matrix, we propose a new method to estimate the NF kurtosis from the non-Hermitian statistics matrix. Then, the related NF components can be removed from the mixed signals to avoid interfering with FF direction-of-arrival (DOA) estimations. Based on the subspace technique, the FF DOAs can be determined by the one-dimensional spectrum search. The proposed algorithm does not require additional classification operations, even when NFSs have the same DOAs as FFSs. Finally, several simulations demonstrate the better estimation performance of the proposed algorithm than existing localization algorithms. [ABSTRACT FROM AUTHOR]

This paper presents a pattern synthesis algorithm achieving minimum mainlobe width via sparse optimization. The problem of synthesizing a pattern with minimum mainlobe width is formulated as a sparse optimization problem with l 0 norm by introducing a slack variable. To solve the sparse optimization problem, three existing relaxations for l 0 norm are presented. Moreover, a novel log-sum-exp penalty function is proposed to replace l 0 norm to be minimized in this paper, leading to a convex problem which can be solved directly and do not need to solve a sequence of sparse optimization problems compared with the iterative reweighted l 1 norm. Both the focused beam pattern and shaped beam pattern for arbitrary arrays can be synthesized utilizing the proposed algorithm. Besides, the mainlobe width of pattern synthesized by the proposed algorithm is minimum. Simultaneously, it brings an extra advantage that it is no longer necessary to accurately determine the mainlobe region without synthesis performance degradation. Numerical examples are presented to verify the effectiveness and superiority of the proposed algorithm. • Achieving minimum mainlobe width is formulated as sparse optimization problem with l 0 norm. • A log-sum-exp penalty function is proposed to replace l 0 norm for the sparse optimization problem. • The proposed method is suitable for not only focused beam pattern but also shaped beam pattern. • The proposed method is suitable for arbitrary arrays. • There is no need to accurately determine the mainlobe region. [ABSTRACT FROM AUTHOR]