Kefei Liao

The growing frequency of unauthorized UAV activities has increased the demand for real-time perception and rapid response on resource-constrained edge devices. This study proposes an edge-deployable UAV detection and net-capture system based on Net-Capture Detection YOLO (NCDet-YOLO). Developed from YOLOv8n, NCDet-YOLO incorporates C2f_Faster, SPD_Conv, EMA, and a lightweight three-scale detection head, with CrossKD used to compensate for accuracy loss caused by structural compression. The dataset contains 6615 images and was divided into 5292 training and 1323 validation images. The self-collected data include DJI Phantom 4 and DJI Inspire 2 UAVs observed at approximately 4–30 m under different daytime backgrounds. NCDet-YOLO achieves an mAP50–95 of 0.6504 with 1.55 M parameters and 4.1 GFLOPs. On a Jetson Orin NX Super under the 15 W power mode, it achieves 31.53 FPS, representing a 31.67% increase over YOLOv8n. The detector is further integrated with target alignment, distance determination, trigger control, and net-capture execution. In 10 real-platform trials, 8 captures were successful, corresponding to an 80.0% success rate, with one false-trigger event and an end-to-end latency from target detection to net-capture firing of approximately 400 ms.

Reconfigurable intelligent surfaces (RIS) and integrated sensing and communication (ISAC) are key enablers for 6G networks, with their integration enhancing spectrum efficiency and multifunctional operations, as explored in recent RIS-aided ISAC frameworks. However, prior works overlook two issues: distinct channel models for radar typically LoS-dominant versus multipath-heavy communication links, and the need for advanced radar metrics beyond traditional bounds. In this paper, we propose a novel RIS-assisted ISAC system tailored for urban environments and adopt a hybrid fading channel model, where LoS-dominant radar direct links are characterized by Rician fading, while RIS-assisted reflected links are modeled using Nakagami-m fading.We introduce a novel dual-path radar model where the base station receives target echoes from both direct and RIS-reflected illuminations. To rigorously assess radar performance, we adopt the radar estimation information rate (REIR), an advanced metric derived from mutual information principles, which quantifies the information-theoretic potential for extracting target-related information from radar echoes under fading and noise. For communication, closed-form expressions for outage probability and ergodic rate are derived under the hybrid channel conditions. Simulations validate the analytical results and demonstrate superior REIR and rate gains over non-RIS baselines in urban environments.

Frequency diverse array multiple-input multiple-output (FDA-MIMO) radar offers significant advantages in mainlobe deceptive interference suppression, as its transmit steering vector contains both angle and range information, providing additional degrees of freedom beyond the angular dimension. However, conventional FDA-MIMO radar suffers from insufficient angle-range resolution, which limits its ability to suppress interferences located close to the target. Moreover, it lacks robustness under limited snapshots and parameter mismatch conditions. To address these issues, this paper proposes a robust beamforming method based on the FDA-MIMO radar model. A collocated sparse array with a sinusoidal element spacing offset and a logarithmic frequency offset is adopted to enhance beam resolution and resolve the periodic angle-range ambiguity problem. Based on this model, the interference-plus-noise covariance matrix is reconstructed using two-dimensional Capon spatial spectrum, and the steering vector is corrected via a joint objective function that combines MUSIC orthogonality and the flatness of the covariance residual spectrum. Simulation results demonstrate that, under conditions of near-target interferences, random range-angle errors, and frequency offset errors, the proposed method achieves a signal-to-interference-plus-noise ratio (SINR) close to the ideal value, exhibiting excellent mainlobe interference suppression performance and robustness.

This paper presents the design and implementation of a real-time visual tracking system for unmanned aerial vehicles (UAVs), based on the DJIPayload Software Development Kit (PSDK), addressing the challenge of balancing high precision with low latency on resource-constrained edge platforms. By utilizing DJI PSDK to abandon the Robot Operating System (ROS) layer and its associated serialization overhead, the proposed Middleware-Free Architecture reduces end-to-end latency by over 60% to approximately 30 ms. To address computational constraints, a Lightweight Asymmetric De-coupled Visual Servoing (ADVS) strategy is proposed. It adopts orthogonal kinematic de-coupling to bypass Jacobian matrix inversion and integrates a non-linear dead-zone mechanism with dynamics-aware gain scheduling to compensate for sensing anisotropy and gravitational nonlinearity. Simultaneously, a Geometry-Aware Fusion strategy is employed to reject visual outliers, while a Finite State Machine (FSM) strictly enforces temporal consistency. Field experiments in various scenarios verify the system’s stability and tracking capability. Specifically, the platform maintains a robust lock on targets at speeds up to 23 m/s across dynamic maneuvers. The successful implementation of this system confirms that high-performance edge tracking does not rely solely on the scaling of visual model complexity but can also be effectively achieved through the architectural minimization of latency combined with the optimization of theoretically grounded robust control strategies.

In this paper, a novel Frequency Diverse Array–Multiple Input Multiple Output (FDA-MIMO) radar parameter estimation algorithm based on Graph Signal Processing (GSP) is proposed for joint range–angle–velocity estimation. By modeling the FDA-MIMO radar echoes as graph signals and constructing an adjacency matrix that captures the spatial correlations among array elements, the proposed method performs Graph Fourier Transform (GFT) analysis to extract the spectral characteristics of the target signal. The target parameters are then obtained by searching for the spectral peak responses corresponding to the unit eigenvalue of the graph adjacency matrix. The proposed GSP-based FDA-MIMO radar framework provides an efficient and high-precision solution for multi-target parameter estimation, with strong potential for real-time and complex-environment applications.

Frequency diverse array (FDA) has a distance‐angle–dependent beam direction map, which has a greater application advantage than phased array. However, due to the distance‐angle coupling and periodicity of its beam direction map, it makes the parameter information of the target obtained through FDA have ambiguous phenomena. Currently, researchers mostly use different types of frequency offsets, array element arrangement, and optimized parameters to eliminate the coupling and periodicity, but the effect is poor and does not consider the balance between the main flap width, side flaps, and gate flaps. We introduce an improved distributed array with two ends transmitting and middle receiving into the FDA‐MIMO radar and propose a new distance‐angle decoupled beam formation method to eliminate the coupling and periodicity of the FDA beam. The method utilizes the NSGA‐II algorithm to jointly optimize the frequency offset increment and array element layout to achieve complete decoupling of the distance‐angle dimension and spot beam focusing and significantly suppresses the sidelobes and eliminates the grating flaps, providing flexible and adjustable beam characteristics. Simulation results demonstrate the effectiveness of this scheme.