Boon-Chong Seet

Ship detection in synthetic aperture radar (SAR) imagery plays an important role in maritime surveillance and security applications. However, in complex near-shore and port scenes, strong background clutter, significant variations in target scale, and the need for efficient feature representation make it challenging to achieve a good balance between detection accuracy and computational efficiency. To address these challenges, this paper proposes a lightweight oriented bounding box (OBB)-based SAR ship detection network based on feature enhancement and cross-scale fusion. By systematically redesigning key components, including feature enhancement, downsampling modeling, multi-scale fusion, and oriented target representation, the proposed method achieves coordinated optimization of accuracy and efficiency in complex scenarios. Specifically, a Multi-Scale Coordinate-Channel Attention mechanism (MSCCA) is constructed to enhance discriminative ship features while suppressing background interference. An Adaptive Dual-pooling Fusion Downsampling strategy (ADFD) is then introduced to reduce spatial resolution while alleviating fine-grained feature loss. In addition, a Lightweight Adaptive Multi-Scale Feature Enhancement module (LAMFE) is designed to improve multi-scale modeling capability under strict computational constraints. Finally, a Lightweight Oriented Bounding Box detection head (LwOBB) is employed to enable accurate localization of ship targets with arbitrary orientations. The experimental results on the publicly available R-SSDD and RSDD-SAR datasets show that the proposed detection network achieves higher detection accuracy while maintaining a lightweight structure. The model contains only 2.06 M parameters and has a computational complexity of 5.7 GFLOPs. Compared with several mainstream detection methods, the proposed method has achieved significant improvements in detection accuracy and effectively reduces the occurrence of missed detections and false alarms.

Unmanned aerial vehicle (UAV) wireless image transmission has gained widespread application across various fields due to its flexibility, yet it faces critical challenges such as resource constraints and degradation of reconstruction quality caused by harsh channel conditions. To address these issues, we designed a lightweight semantic communication backbone network that substantially reduces the computational and storage overhead of UAVs through codebook assistance and efficient encoder-decoder design. On this basis, to tackle severe image degradation under adverse channel conditions, we introduced a generative artificial intelligence-based (GAI) enhancement module. Specifically, we developed a semantic refinement network (SRN) that employs an innovative signal-to-noise ratio (SNR) adaptive feature-wise linear modulation (FiLM) layer to dynamically adjust its refinement strategy based on real-time channel quality, fundamentally transforming the image reconstruction paradigm from traditional signal recovery to conditional content generation. Extensive experimental results demonstrate that our proposed framework significantly outperforms the current state-of-the-art method under extreme channel conditions, highlighting its great potential for achieving robust UAV image transmission in challenging operational environments.

Millimeter-wave (mmWave) detection is a widespread human activity recognition (HAR) method. However, due to the mmWave characteristics, it is challenging to per-form HAR in non-line-of-sight (NLOS) environments. Before the prevalence of artificial intelligence (AI) technologies, mmWave-based HAR systems mainly relied on tradi-tional signal processing and handcrafted feature extraction (e.g. time–frequency analy-sis, multi-path modeling, micro-Doppler signature analysis), combined with classical classifiers such as Support Vector Machine (SVM) or Random Forest. However, these approaches were highly sensitive to environmental variations and failed to generalize in NLOS conditions. With the advent of AI, techniques such as Convolutional Neural Network (CNN) have played a crucial role in feature extraction from two-dimensional images generated by mmWave radar signals. In this paper, for the first time, a frame-work for Federated Continual Learning (FCL) based on CNN is proposed for improving radar detection accuracy in NLOS environments while preserving personal privacy through local training without uploading radar tensors and personal data to the global server for model aggregation. Additionally, the FCL model learns from both LOS and NLOS environments for improving cross-domain adaptability and recognition.

The terahertz (THz) band offers ultra-wide bandwidth for next-generation high-speed wireless communication systems. However, achieving compact, high-gain, and beam-symmetric THz antennas remains challenging due to fabrication and propagation constraints. This paper presents a simulation-based design and optimization of a stacked substrate-integrated waveguide (SIW) pyramidal horn antenna achieving equal half-power beamwidths (HPBWs) in both E- and H-planes. The design employs vertically stacked SIW layers coupled through optimized slot apertures to ensure dominant TE₁₀ mode propagation with minimal reflection. Using full-wave electromagnetic simulations, the effects of layer number, dielectric loading, amplitude tapering, and phase distribution are systematically analyzed. The optimized five-layer configuration exhibits 10 dBi gain, 41° HPBW, and sidelobe levels around −3.2 dB at 210 GHz. This framework aims to develop high-performance, beam-symmetric THz SIW antennas compatible with standard LTCC/PCB technologies.

Orthogonal Time Frequency Space (OTFS) modulation presents a promising approach for underwater acoustic (UWA) communication. This letter proposes an iterative decision feedback equalizer based on optimal combining (OC) for UWA-OTFS systems, addressing challenges posed by the full-matrix channel characteristics in the delay-Doppler domain. The proposed detector maximizes the signal-to-interference-plus-noise ratio of the combined signal. Furthermore, by introducing inter-Doppler interference depth, the OC detector dynamically constrains the number of combining terms. Simulation results demonstrate that the proposed detector effectively balances bit error rate performance with computational complexity.

Vessel trajectory prediction tasks currently face several challenges, including the diversity of behavioral patterns, interference from anomalous data, and limited generalization capabilities of existing models. These challenges are particularly pronounced when dealing with multi-pattern trajectory characteristics, where achieving both high prediction accuracy and robust cross-scenario adaptability remains difficult. To address these issues, this paper proposes a high-precision vessel trajectory prediction framework. Firstly, the local outlier factor algorithm is applied to eliminate anomalies from raw automatic identification system data, thereby improving data quality. The Douglas–Peucker simplification and the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) trajectory clustering algorithm are then used to construct datasets characterized by different turning behaviors. Subsequently, a prediction model called local-causal temporal attention network (LCTA-Net) is developed, based on a pruned Transformer encoder. LCTA-Net incorporates a temporal residual encoding module and a causal feature-augmented attention mechanism. These components jointly improve the model’s capacity to model temporal dependencies within trajectory sequences. In addition, a fixed-window local neighborhood search strategy is introduced to improve the spatial continuity and physical feasibility of the predicted trajectories. Finally, experimental results on two constructed trajectory datasets demonstrate that the proposed method significantly outperforms several state-of-the-art models across multiple error metrics, confirming its superior prediction accuracy and strong cross-scenario generalization capability.

Underwater wireless communication (UWC) is an emerging technology crucial for automating marine industries, such as offshore aquaculture and energy production, and military applications. It is a key part of the 6G vision of creating a hyperconnected world for extending connectivity to the underwater environment. Of the three main practicable UWC technologies (acoustic, optical, and radiofrequency), acoustic methods are best for far-reaching links, while optical is best for high-bandwidth communication. Recently, utilizing reconfigurable intelligent surfaces (RISs) has become a hot topic in terrestrial applications, underscoring significant benefits for extending coverage, providing connectivity to blind spots, wireless power transmission, and more. However, the potential for further research works in underwater RIS is vast. Here, for the first time, we conduct an extensive survey of state-of-the-art of RIS and metasurfaces with a focus on underwater applications. Within a holistic perspective, this survey systematically evaluates acoustic, optical, and hybrid RIS, showing that environment-aware channel switching and joint communication architectures could deliver holistic gains over single-domain RIS in the distance–bandwidth trade-off, congestion mitigation, security, and energy efficiency. Additional focus is placed on the current challenges from research and realization perspectives. We discuss recent advances and suggest design considerations for coupling hybrid RIS with optical energy and piezoelectric acoustic energy harvesting, which along with distributed relaying, could realize self-sustainable underwater networks that are highly reliable, long-range, and high throughput. The most impactful future directions seem to be in applying RIS for enhancing underwater links in inhomogeneous environments and overcoming time-varying effects, realizing RIS hardware suitable for the underwater conditions, and achieving simultaneous transmission and reflection (STAR-RIS), and, particularly, in optical links—integrating the latest developments in metasurfaces.