To address security concerns in unmanned aerial vehicle (UAV) swarm wireless communication, a joint optimization algorithm integrating beamforming and trajectory design is proposed to maximize the average secrecy rate in this paper. First, the constraint model of swarm trajectory is developed based on the swarm's central point and flight dynamics. Each UAV in the swarm is equipped with an omnidirectional antenna, forming a large-scale virtual antenna array (VAA). By analyzing the signals from the antenna array, the wireless communication channel model is developed that accounts for the uncertainty in the eavesdropper's position, leading to the derivation of the worst-case average secrecy rate. Building on this, the beamforming weights and the swarm's trajectory are treated as optimization variables, and the optimization problem is formulated to maximize the average secrecy rate. To solve the problem, it is decomposed into two subproblems: beamforming and trajectory design. The beamforming problem is transformed into a linear programming problem, while the trajectory design problem is reformulated as a convex optimization problem using successive convex approximation. The numerical results validate the feasibility and reliability of the proposed algorithm under uncertain eavesdropper's location. A comparative analysis further demonstrates the superiority of the algorithm in enhancing the secrecy rate.

Distributed multiple-input multiple-output (MIMO) radar networks operating at millimeter-wave frequencies enable high-resolution microwave imaging but face fundamental limitations in antenna array synthesis. Sparse virtual apertures reduce hardware complexity yet introduce severe grating lobes and sidelobe artifacts, which degrade image fidelity via aliasing in the electromagnetic (EM) spatial frequency domain. To address these antenna array synthesis challenges, we propose a noisy tensor completion framework exploiting joint low-rank and sparsity constraints inherent in radar scattering. Our method reorganizes radar echoes into high-dimensional tensors with dispersed missing elements, explicitly modeling the sparse array sampling process. A key innovation is an adaptive singular-value reweighting scheme that preserves dominant EM scattering components while suppressing noise-corrupted interference. The resulting optimization is solved via an alternating direction method of multipliers (ADMM) algorithm. Extensive EM simulations and experimental validation using a prototype W-band (77 GHz) distributed MIMO radar system demonstrate superior artifact suppression and target reconstruction over state-of-the-art methods. This establishes a robust imaging solution for sparse-aperture systems by directly addressing antenna array pattern limitations through tensor-based aperture synthesis.

A key challenge in sparse linear arrays (SLAs) is that only partial elements can be observed in a single snapshot. To address this, we employ low-rank Toeplitz matrix completion to estimate missing entries. This method is integrated into a nuclear norm minimization framework tailored for SLA sampling patterns, which directly governs the matrix recovery quality. From this perspective, we establish an empirical performance guarantee linked to the spectral norm of the sampling matrix, providing a quantitative metric for sparse array design. Arrays designed with lower spectral norms demonstrate superior recovery performance. Simulations validate this correlation and suggest using the spectral norm as a preprocessing tool for array screening, substantially reducing design iterations.

Extracting distinct micro-Doppler signatures for human posture recognition using radar sensing remains challenging, especially under small-sample and low signal-to-noise ratios (SNRs) situations. Taking the outstanding measurement resolution of multiple-input multiple-output (MIMO) radar and rapid development of deep learning into consideration, a human posture recognition method based on micro-Doppler effect and Transformer network is proposed. Firstly, micro-Doppler mathematical expressions corresponding to specific human limb are derived to clarify the micro-Doppler modulation mechanism. Secondly, multi-channel signal accumulation and two-dimensional fast Fourier transform (2D FFT) are performed on measured data to form and enhance micro-Doppler images. After that, micro-Doppler sequences are obtained by mapping micro-Doppler images along the frequency axis. Thirdly, a single-channel image Transformer network is designed to learn global micro-Doppler characteristics, and by comparing with other advanced neural networks, the advantages of the designed network in small-sample scenes were demonstrated. Fourthly, another sequence Transformer network is established to extract features from micro-Doppler sequence. Combining the sequence Transformer network with the image Transformer network, an entropy-weighted micro-Doppler feature fusion strategy and a weighted cross-entropy loss function are proposed, forming a dual-channel feature fusion model to improve small-sample recognition performance under SNRs. At last, by designing ablation experiments and conducting interpretable analysis, it is verified that the designed dual-channel feature fusion model can learn more effective micro-Doppler features from measured sensor data of different scenes. Research results in this paper would help contribute to more robust and accurate human posture recognition.

In this paper, a track deception scheme by unmanned aerial vehicle (UAV) swarm against radar network with limited communication is proposed. Firstly, a track deception model for single UAV is established, along with a communication framework for information exchange within the swarm. To address packet dropouts, a track deception scheme based on predictive mechanism is introduced, in which cluster head (CH) UAVs estimate the state information of cluster member (CM) UAVs and generate optimal control decisions over multiple time steps. Based on this mechanism, a trajectory planning optimization model is developed with the objective of minimizing the flight velocity of UAVs, subject to kinematic performance constraints and safe distance requirements. This optimization problem is solved by particle swarm optimization (PSO) algorithm. Simulation results demonstrate that the phantom tracks generated with limited communication can effectively pass the homology test from radar network, validating the robustness and feasibility of the proposed scheme.

To maximize the matched-detection gain between the transmitted radar waveform and uncertainty target characteristics, an uncertainty-aware expected mutual information metric for radar information detection and a broadband radar frequency-domain waveform optimization model are established. The resulting model is solved via the projected gradient-descent algorithm, enabling iterative computation of the waveform's power spectrum. To further enhance the low probability of interception (LPI) performance of radar systems under uncertain target detection scenarios, an LPI information metric is formulated based on the full signal interception and recognition process. A corresponding frequency-domain waveform optimization model is developed, jointly optimizing the bandwidth and center frequency of wideband radar signals. Given the nonconvex and nonlinear nature of the proposed optimization problem, a sequential convex approximation method is employed to convert the problem into a second-order cone programming form at each iteration. Additionally, a heuristic frequency band programming strategy is incorporated to address the waveform design problem under echo uncertainty. Simulation results validate the effectiveness of the proposed uncertainty-aware detection and LPI optimization metrics, demonstrating the superior detection and LPI performance of the designed wideband radar waveforms in maneuvering target detection and tracking scenarios.

How to efficiently schedule the transmit beams and power resources in colocated multiple-input-multiple-output (C-MIMO) radar network to achieve better multitarget tracking (MTT) accuracy is a critical research issue. However, in electronic countermeasure environments, excessive resource allocation may lead to the interception of transmit pulses by hostile receivers. As the number of intercepted pulses increases, the transmit signals of C-MIMO radar nodes would be identified and subject to precise suppression jamming, resulting in a sharp degradation in MTT performance. To overcome this shortcoming, we propose a low probability of identification (LPID)-based composed resource allocation (CRA) scheme for MTT in C-MIMO radar network. Closed-form expressions for the predicted conditional Cramer-Rao lower bound (PC-CRLB) and the radar signal identification probability under suppression jamming are derived. Based on the tactical significance map function, we compute the comprehensive threat level for each target in real time. We then employ the sum of products between PC-CRLB values and threat levels, and the worst-case identification probability as the quantitative metrics for MTT performance and LPID performance, respectively. The LPID-CRA is developed to jointly balance these two metrics by dynamically allocating beam-pairing and power resources within each tracking frame. To address the resulting nonconvex nonlinear problem, we perform an equivalent transformation of the optimization model and develop an efficient five-stage solution method that incorporates the Zoutendijk methods of feasible directions and the min-cost flow algorithm. Simulation results demonstrate that the proposed LPID-CRA strategy significantly enhances overall system performance.

To address the survivability challenge of multi-UAVs in hostile environments, a three-dimensional (3D) stealth trajectory planning scheme based on an improved A-star algorithm is proposed for threat avoidance. First, a 3D environment model incorporating radar detection and no-fly zones is established. Subsequently, four evaluation metrics are constructed to assess UAV mission fault tolerance, flight stability, trajectory smoothness, and stealth performance. The weighted sum of these evaluation metrics is defined as the multi-UAV 3D trajectory planning efficiency function. Building upon this framework, an optimization model for multi-UAV stealth trajectory planning in 3D dynamic environments is formulated, with the objective of maximizing the efficiency function while incorporating constraints such as no-fly zones and UAV maneuverability parameters. Finally, a two-step solution method based on the improved 3D A-star algorithm is employed to solve the optimization model. Simulation results demonstrate that, compared to the benchmark, the proposed scheme generates feasible 3D stealth paths for multiple UAVs that satisfy maneuverability constraints and exhibit the low probability of being detected by the radar network.

In this paper, we propose a low probability of identification (LPID)-based composed resource allocation (CRA) scheme for multi-target tracking (MTT) in colocated multiple-input-multiple-output (C-MIMO) radar network under electronic countermeasure environments. Closed-form expressions for the predicted conditional Cramér-Rao lower bound (PC-CRLB) and the radar signal identification probability under suppression jamming are derived. We then employ the sum of PC-CRLB values and the worst-case identification probability as the quantitative metrics for MTT performance and LPID performance, respectively. The LPID-CRA is developed to jointly balance these two metrics by dynamically allocating beam-pairing and power resources within each tracking frame. To address the resulting non-convex non-linear problem, we perform an equivalent transformation of the optimization model and develop an efficient five-stage solution method that incorporates the Zoutendijk methods of feasible directions (ZMFD) and the min-cost flow algorithm. Simulation results demonstrate that the proposed LPID-CRA strategy significantly enhances overall system performance.

Adaptive Sidelobe Cancellation (ASLC) is a core technology for modern radar systems to suppress active sidelobe jamming. From the perspective of disrupting the ASLC system’s ability to stably track the jamming direction, this paper proposes a distributed jamming method based on random phase perturbation. The method employs two spatially separated jamming sources that simultaneously transmit coherent signals. By actively applying controllable random jumps to the relative phase between the two sources, the equivalent wavefront direction of the synthesized signal at the radar receiver changes rapidly, forming a non-stationary jamming that destroys the null-tracking capability of ASLC. An analytical model of the ASLC cancellation ratio (CR) under random phase perturbation is established, with a focus on analyzing the effects of time synchronization accuracy and phase synchronization accuracy on jamming performance. Monte Carlo simulation results show that the proposed method can reduce the average ASLC CR from 26.80 dB to 20.29 dB (a decrease of 6.51 dB). Under identical conditions, this performance is comparable to asynchronous blinking jamming while requiring no precise timing matching, and outperforms multi-source saturation jamming in resource efficiency (two vs. four jammers). This study provides promising simulation-level evidence for the effectiveness of the proposed jamming method. The quantitative results and sensitivity analyses offer a simulation-level theoretical reference for parameter design of distributed cooperative jamming. Further validation in semi-physical simulations or field trials is necessary before claiming engineering readiness.

In this paper, the joint design of transmit and receive beamformers for transmit subaperturing multiple-input-multiple-output (TS-MIMO) radar is investigated, aiming to enhance its low probability of intercept (LPI) capability. The main objective is to simultaneously minimize the transmission power, suppress the transmit sidelobe levels, and minimize the probability of intercept, thus bolstering the LPI performance of the radar system while maintaining the desired target detection performance. An alternative optimization method is proposed to jointly optimize the transmit and receive beamformers, yielding an unified LPI optimization framework. Particularly, the proposed iterative algorithm based on the Lagrange duality theory for transmit beamforming is more efficient than the conventional convex optimization method. Numerical experiments highlight the effectiveness of the proposed approach in sidelobe suppression and computational efficiency.

The low probability of intercept (LPI) performance is pivotal for the self-safeguarding of dual-function radar-communication (DFRC) systems. This paper proposes an LPI-based resource optimization algorithm for the multi-carrier DFRC system. Different subcarriers are allocated reasonably to realize radar detection or communication transmission functions, which establishes the DFRC system model. The radar conditional mutual information and data transmission rate are derived to quantify the performance of radar and communication systems, respectively. These indicators are imposed on the constraints, aiming to ensure the performance of multi-target detection and multi-user communication. The optimization objective is to minimize the total transmit power to improve the LPI performance. To address the non-convexity of the optimization problem, a convex relaxation-based alternating algorithm is proposed. The variables are placed within an alternating framework, and the nonlinear inequality constraints are relaxed to apply Karush-Kuhn-Tucker (KKT) conditions for each iteration. Simulations validate that the proposed algorithm can provide lower transmit power for multi-target detection and multi-user communication, thereby facilitating the LPI performance of the DFRC system.

The rapid proliferation of low-altitude drones has led to increasingly congested and heterogeneous electromagnetic environments, posing significant challenges to fine-grained spectrum awareness and reliable drone management. Specific emitter identification (SEI), which exploits inherent hardware-dependent radio frequency fingerprints, provides an effective physical-layer solution for emitter-level discrimination. However, practical SEI systems often suffer from two critical issues: extremely limited labeled samples for newly emerging emitters and heterogeneous data distributions collected by geographically distributed receivers with mismatched label spaces. To address these challenges, this paper proposes a heterogeneous federated learning (HFL)-based framework for few-shot specific emitter identification (FS-SEI). The proposed framework decouples feature embedding learning from task-specific classification and enables collaborative representation learning across distributed receivers without sharing raw signal data. A metric learning-based training strategy is adopted, where only the feature embedding models are aggregated in the federated process, effectively alleviating the impact of label space mismatch by utilizing center loss and an improved triplet loss. Moreover, two federated optimization schemes, namely gradient averaging (GA) and model averaging (MA), are systematically investigated to analyze their effectiveness under fully heterogeneous settings. Extensive experiments conducted on a real-world dataset demonstrate that the proposed HFL framework significantly outperforms isolated local training. In particular, the GA-based scheme achieves a few-shot identification performance that closely approaches centralized learning while preserving data privacy and robustness against data heterogeneity. The results validate the effectiveness of the proposed approach for practical FS-SEI in low-altitude drone management scenarios.

Regularly undersampled planar near-field scattering measurements can substantially reduce scan time and facility requirements for electromagnetic characterization of electrically large objects; however, regular undersampling often causes severe spatial aliasing and degrades subsequent sector-limited far-field radar cross section (RCS) evaluation. This paper presents a physics aware measurement-to-prediction workflow for 3-D polarimetric reconstruction and broadside-sector far-field RCS prediction from regularly undersampled planar near field HH/HV/VV data. The workflow contains three steps: (i) robust wideband time-domain back-projection (TDBP) for artifact-suppressed pre-imaging and support localization using frequency diversity, (ii) 3-D CLEAN-based support extraction to reduce the candidate set of scattering locations, and (iii) physics-aware orthogonal matching pursuit (PA OMP) to estimate symmetric scattering-tensor coefficients on the extracted support at a selected frequency, with a sector-guided far-field ranking term. Wideband data are used primarily for robust support localization, whereas coefficient inversion and the reported far-field prediction are carried out at a selected single frequency. Experiments on a calibrated corner-reflector array and an amphibious UAV validate the proposed workflow: with a scan density equal to about 22% of the conventional full-angle planar λ/2 benchmark at 8 GHz, the method localizes dominant scattering contributions and synthesizes broadside-sector far-field RCS whose overall trends are consistent with independent compact-antenna-test-range (CATR) measurements. The method is intended for broadside-sector prediction from planar scans rather than full-angular characterization.

Low probability of interception (LPI) technology is critical for the self-protection of dual-function radar-communication (DFRC) systems in hostile environments. However, the range-dimensional uncontrollability of the beampattern imposes a bottleneck limitation on LPI development. This article investigates a transceive beamforming for frequency diverse array (FDA)-multiple-input-multiple-output (MIMO)-based DFRC systems. Initially, we innovatively construct the signal model for the FDA-MIMO DFRC system, which surmounts the inseparability between frequency offsets and transmitting antennas. Subsequently, the equivalent transmitted beampattern and communication channel are derived into 2-D range-angle expressions, aiming to implement joint range-angle beamforming. The communication symbols are strategically placed in predetermined range-angle positions of each user through the beampattern modulation technique. The LPI optimization problem is formulated as the minimization of transmitting beamforming power, subject to the constraints of radar signal-to-interference-plus-noise ratio (SINR) and communication sum rate. To tackle the nonconvex problem with high-dimensional nonlinear constraints, a distributed alternating algorithm is developed for fast solving, involving sequence convex approximation, weighted mean-square error minimization, and alternating direction method of multipliers. Simulation results demonstrate that our method achieves punctate multitarget detection and multiuser communication in 2-D range-angle planes, enormously facilitating the LPI capability.

The integration of multiple-input multiple-output (MIMO) technology with the dual-functional radar and communication (DFRC) system is pivotal for enhancing the capabilities of the Internet of Vehicles (IoV). However, conventional MIMO-DFRC designs lack support for simultaneous multi-user communication in the same direction, limiting their utility in IoV applications. To overcome this limitation, we propose a novel robust transmit beamforming approach for the MIMO-DFRC system, facilitating multi-user communication through beampattern modulation. Initially, the orthogonal Oppermann sequences are allocated to each user. Subsequently, these sequences are embedded into sub-beampatterns, which encode communication symbols for multi-user information transmission. Furthermore, the encoded information is strategically placed within the sidelobes of the transmit beampattern, aiming to minimize the integrated sidelobe levels. This goal serves as the objective function while maintaining mainlobe detection performance, which imposes a constraint, forming an optimization model for transmit beamforming. Due to the non-convex nature of the optimization model, it is decomposed using a coordinate rotation strategy and addressed iteratively via a convex optimization algorithm. Simulations validate that our method facilitates reliable unidirectional communication among multiple users and preserves the integrity of radar detection performance, contributing to the IoV application of the DFRC technology.

Achieving high angular resolution estimation is essential for enhancing environmental perception and accurately detecting extended targets. However, implementing a large aperture radar system as a single sensor faces significant technical and economic challenges. Radar networks, comprising multiple individual radar sensors, offer a promising solution to these obstacles. In this article, we propose a method for high resolution direction-of-arrival (DOA) estimation tailored for coherent distributed millimeter-wave radar systems. Our approach leverages multiple-input multiple-output technology and distributed arrays to construct a virtual sparse array capable of synthesizing a full uniform array covariance matrix through specific correlation transformation. Furthermore, we introduce a joint optimization method for directly angle finding that efficiently addresses large-scale optimization problems by exploiting both sparsity and low-rank properties, thereby circumventing the traditional sequential steps of matrix reconstruction and DOA estimation. We employ the alternating direction method of multipliers to ensure reliable convergence. Our proposed model and method have been extensively validated through comprehensive simulations, highlighting their advantages within the realm of distributed radar systems.

In this study, a joint detection threshold optimization and multidimensional resource allocation (JDTO-MRA) scheme based on low probability of intercept is put forward for multitarget tracking in phased-array radar networks. The foundation of the proposed JDTO-MRA scheme is to adopt the optimization methodology to adaptively coordinate the detection threshold, radar node selection, transmit power, and signal bandwidth of each radar node to minimize the total power consumption of the underlying system, subject to given target detection probability and tracking accuracy constraints and several system resource budgets. The analytical expressions of the average target detection probability and Bayesian Cram & eacute;r-Rao lower bound are derived and utilized as the metrics to depict the detection and tracking performance of multiple targets. By incorporating the simultaneous detection and tracking concept, resource-aware design, and improved probabilistic data association algorithm into a coherent framework, the JDTO-MRA model is established in phased-array radar networks. Due to the optimization parameters are all coupled in both the constraints and criterion function, the resulting JDTO-MRA model demonstrates a nonlinear and nonconvex problem. Combined with the semidefinite programming method and the sequential quadratic programming method, an appropriate five-step solution technique is proposed to solve the original problem. Several numerical results are developed to verify the superiority and effectiveness of the JDTO-MRA scheme.