Visual Object Tracking (VOT) remains challenging under occlusion scenarios, where traditional trackers often suffer from feature degradation and target loss. To address this issue, we propose OASAMT, an occlusion-aware tracking framework that equips SAM2 with explicit temporal occlusion reasoning via two Temporal Convolutional Networks (TCNs) and a Dual-Memory Bank (DMB). Specifically, two TCN-based modules are designed to model temporal occlusion dynamics: the Temporal Occlusion Classifier (TOC) for inferring target occlusion states using confidence scores, mask IoU, and area ratio; and the Temporal Occlusion Predictor (TOP) for forecasting target bounding boxes during occlusion. The proposed DMB consists of a Non-Occlusion Memory Bank (N-OMB) and an Occlusion Memory Bank (OMB), explicitly decoupling reliable and occluded representations to prevent memory contamination and improve re-detection after occlusion. Additionally, to facilitate systematic evaluation under occlusion scenarios, we construct OccTrack, a dedicated occlusion-oriented dataset derived from four UAV-view benchmarks. Extensive experiments were conducted on the OccTrack, LaSOT, LaSOT ext , and GOT-10k datasets. The results demonstrate that OASAMT consistently outperforms SAM2.1 and other advanced trackers in both occlusion-specific and general tracking scenarios. The code and the dataset are available at https://github.com/ChaseFalcon99/OASAMT .

Movable antenna (MA) enhances wireless communication performance by enabling the flexibility in antenna movement. Prior works on MA usually ignore the time overhead for antenna movement in channel estimation and/or performance improvement. However, due to the mechanically constrained MA movement speed and limited channel coherence time, the antenna movement time can significantly affect the effective communication rate of MA systems in practice. To address this issue, we propose to jointly design the MA’s trajectories for channel measurements and rate-optimal repositioning to maximize the average effective communication rate of an MA-aided receiver subject to the given transmission block duration. Specifically, we propose a two-timescale optimization approach, in which the MA’s trajectory for channel measurements is optimized in the long term based on the known channel distribution, while the MA’s trajectory for moving to the rate-optimal position is adaptively designed in the short term to cater to the instantaneous channel realizations, thus simplifying the design complexity and yet providing high adaptability to channel variations. In particular, we propose a kernel-based regression (KBR) method to efficiently reconstruct the channel map over the whole antenna moving region based on the limited channel measurements, utilizing an offline-learned kernel for which closed-form expressions are theoretically derived under different multipath channel distributions. Numerical results demonstrate that the proposed scheme outperforms the fixed-position antenna (FPA) system and other benchmark MA designs that neglect the antenna movement time, and even achieves performance comparable to that of the single-input multiple-output (SIMO) beamforming system. Furthermore, the implementation cost of the proposed MA scheme is analyzed to assess its practical feasibility. Although mechanical antenna movement incurs additional energy consumption, the proposed MA scheme achieves superior energy efficiency to both the FPA and SIMO systems.

In this paper, we investigate the channel estimation problem for extremely large-scale multiple-input multiple-output (XL-MIMO) systems with a hybrid analog-digital architecture, implemented within a decentralized baseband processing (DBP) framework with a star topology. Existing centralized and fully decentralized channel estimation methods face limitations due to excessive computational complexity or degraded performance. To overcome these challenges, we propose a novel two-stage channel estimation scheme that integrates local sparse reconstruction with global fusion and refinement. Specifically, in the first stage, by exploiting the sparsity of channels in the angular-delay domain, the local reconstruction task is formulated as a sparse signal recovery problem. To solve it, we develop a graph neural networks-enhanced sparse Bayesian learning (SBL-GNNs) algorithm, which effectively captures dependencies among channel coefficients, significantly improving estimation accuracy. In the second stage, the local estimates from the local processing units (LPUs) are aligned into a global angular domain for fusion at the central processing unit (CPU). Based on the aggregated observations, the channel refinement is modeled as a Bayesian denoising problem. To efficiently solve it, we devise a variational message passing algorithm that incorporates a Markov chain-based hierarchical sparse prior, effectively leveraging both the sparsity and the correlations of the channels in the global angular-delay domain. Simulation results show the effectiveness and superiority of the proposed SBL-GNNs algorithm over existing methods, demonstrating improved estimation performance and reduced computational complexity.

To cope with the unendurable delay of maritime wireless networks (MWNs), this paper proposes a collaborative transmission framework utilizing a multi-antenna uncrewed surface vessel (USV) and multiple cache-aided buoys to satisfy the on-demand file requirements for remote users (RUs). Specifically, a direct transmission scheme is adopted for hit-requested files and a multi-hop transmission scheme is devised to handle cache misses. To fully exploit the local cache and signal processing capabilities, we integrate two schemes into a collaborative transmission framework, where the USV dynamically supports buoys in uncached file fetching, and buoys collaborate to forward both cached and fetched files to RUs through a cooperative beamforming policy. We aim to minimize the overall transmission completion time by jointly optimizing the USV trajectory, cooperative beamforming, and transmission duration under the constraints of USV kinetic, transmit power, and file requirements. By leveraging the completion condition analysis, the original problem is transformed into a sequence of one-slot problems and a finite-horizon problem, where the closed-form solution for the local caching beamforming at each buoy is derived. Due to the complexity of the multivariable coupling, we propose an equivalent rate transformation method for transmission strategy design. Numerical results validate the effectiveness of the proposed scheme and algorithm.

Due to long-term reliability, wireless control systems (WCSs) have attracted significant interest recently. However, mission-critical control requires stringent ultra-reliability and low-latency communication (URLLC) with massive data delivery, which are major challenges for conventional wireless networks. This paper investigates downlink URLLC in WCS, where the semantic communication is adopted at the control center to extract task-oriented semantic information from original large-sized data. To efficiency, the control center utilizes the rate splitting policy to deliver semantic information through private messages, while the semantic knowledge is transmitted through one common message. We aim to maximize the weighted sum semantic information transmission rate by jointly optimizing the semantic information extraction, delivery duration, rate splitting, and transmit beamforming, subject to several practical constraints, including recovery accuracy, quality of service requirements, communication latency and computation delay. By the problem decomposition, two sub-problems are obtained, where the closed-form solution for the semantic information extraction is derived at each step. Due to the complexity of the multivariable coupling in the channel dispersion, we propose fractional transformation methods for rate splitting design. Numerical results confirm that the RSMA and semantic communication design can complement each other for multiplexing gains enhancement and latency reduction to achieve overloaded connections.

Driven by the ubiquitous sensing demands, integrated sensing and communication (ISAC) is viewed as an essential technology in future networks. In this paper, we investigate a multiple ISAC-enabled user terminal (UT) system that multi-antenna UTs perform radar sensing and communicate with the BS at the same time. Then, we formulate a multi-UT sum rate maximization problem by jointly considering UT's maximum transmit power and minimum radar signal-to-clutter plus interference and noise ratio (SCINR) requirements. To solve the transmit precoding optimization problem, we first handle the non-convex rate function with weighted minimum mean-squared error method. Then, we use first-order Taylor expansion to deal with the minimum radar SCINR constraints. At last, we propose an iterative optimization algorithm to solve the problem. Simulation results verify the effectiveness of our proposed design.

Due to the flexible deployment, unmanned surface vessels (USVs) have attracted much interest recently. To solve the resource scarcity problem at sea, USV needs to leverage existing terrestrial and satellite systems for efficient backhaul and spectrum sharing. In this case, the multiple input multiple output (MIMO) technology can be applied for diversity gain improvement and interference coordination. However, how to adopt MIMO technology into maritime networks with a sparse scattering environment is still an open issue. In this paper, we employ a multi-antenna USV to support on-demand communications. Utilizing the two-ray channel, we aim to maximize the sum throughput over all USV intended users, by jointly optimizing the cooperative beamforming and trajectory, subject to several practical constraints, including the USV kinetics, quality of service requirement and backhaul capacity. Different from existing whole period designs, we decompose the problem into sequential one-slot problems. Within each slot, the non-convex problem is solved iteratively by using problem decomposition and successive convex optimization methods. Then, channel estimation errors are considered to investigate a robust beamforming scheme. Numerical simulations validate that the USV coexists well with the satellite network and show that the beamforming scheme and trajectory design complement each other for performance improvement.

Unmanned surface vehicles (USVs) have recently found increasing applications in marine scenarios. In this paper, we investigate the cooperative communication of the hybrid terrestrial-maritime wireless system coexisting with a satellite network, where a multi-antenna USV is used as the relay to assist the communication between the terrestrial base station (TBS) and marine users (MUs). Considering the shortage of communication resources, the USV shares the same frequency spectrum with the satellite network. Using the composite maritime two-ray channel, we aim to maximize the throughput over all MUs by optimizing the cooperative beamforming scheme and association jointly with the USV trąjectory, subject to the constraints of USV kinematics, power consumption, quality-of-service requirements, and information-causality. Since the formulated optimization problem is non-convex, we propose an efficient iterative algorithm by applying the block coordinate descent and successive convex optimization methods. Simulation results confirm the significant performance gains of the proposed design as compared to other benchmark methods.

The fifth-generation mobile communication technology (5G) requires the ability to support a variety of different types of services in parallel. This paper considers the problem of enhanced mobile broadband (eMBB) and ultra-reliable and low-latency communication (URLLC) services being jointly precoded under a multiple-input single-output (MISO) system, where the optimization objective is to minimize the precoding power at the base station (BS). However, the problem is difficult to be solved directly due to the complexity caused by interference between URRLC and eMBB users. Thus, we first transform it into a quadratic constraint quadratic programming (QCQP) problem, and then relax it into a convex problem through semidefinite relaxation (SDR). After that, an SDR-Precoding Power Minimization (SDR-PPM) algorithm is proposed to obtain the optimal solution iteratively. Meanwhile, a low-complexity (LC) comparison algorithm is also proposed. Simulation results verify the effectiveness of the proposed algorithms and show the performance of the algorithms as well as the impact of the parameters on precoding power.

Employing unmanned surface vehicles (USVs) as marine data collectors is promising for large-scale environment sensing in remote ocean monitoring network. In this article, we consider a USV-aided marine data collection network, where a USV collects data from multiple monitoring terminals while avoiding collisions with monitoring terminals and obstacles. Aiming at minimizing energy consumption and data loss, we formulate a trajectory optimization problem with practical constraints, including collision avoidance, steering angle, and velocity limitation. The problem is intractable due to the stochastic arrived data and the random emergence and movement of dynamic obstacles. To efficiently solve it, we transform it as a constrained Markov decision process (MDP) problem and address it using a target-oriented double deep${Q}$-learning network (D2QN)-based collision avoidance and trajectory planning algorithm. In the proposed algorithm, the USV acts as an agent to explore and learn its trajectory planning policy by utilizing the causal knowledge. Numerical results demonstrate that the performance of the proposed algorithm is superior in terms of successful probability, energy consumption, and data loss.

In maritime wireless communications, unmanned surface vehicles (USVs) can improve coverage and transmission performance due to their agile maneuverability and flexible deployment. This paper considers a USV-enabled maritime wireless network, where a USV is employed to assist the communication between the terrestrial base station and ships. Considering the maritime environment characteristics and earth curvature, we establish the systematic USV kinetics and information transmission models. To guarantee fairness, we aim to maximize the minimum expected throughput overall ships by jointly optimizing the trajectory and communication resource allocation, subject to the constraints of the USV kinetics, safe sailing, breakpoint distances, line-of-sight links, resource allocation, and information-causality. Due to the complexity of the maritime two-ray signal propagation model, we propose a channel approximation method to find an upper bound of the throughput for the original problem. By the problem decomposition, two sub-problems are derived and solved iteratively using successive convex approximation and interior-point methods. Simulation results confirm the effectiveness of the proposed method and show that USV can significantly improve transmission performance in maritime wireless networks.