In response to the rapid growth of global video traffic and the limitations of traditional wireless transmission systems, we propose a novel dual-stage vector quantization system, VQ-DeepVSC, tailored to enhance video transmission over wireless channels. In the first stage, we design the adaptive key-frame extractor and interpolator, deployed respectively at the transmitter and receiver, which intelligently select key frames to minimize inter-frame redundancy and mitigate the cliff-effect under challenging channel conditions. In the second stage, we propose the semantic vector quantization encoder and decoder, placed respectively at the transmitter and receiver, which efficiently compress key frames using advanced indexing and spatial normalization modules to reduce redundancy. Additionally, we propose adjustable index selection and recovery modules, enhancing compression efficiency and enabling flexible compression ratio adjustment. Compared to the existing methods based on joint source-channel coding (JSCC) framework, the proposed method exhibits superior compatibility with current digital communication systems. Experimental results demonstrate that VQ-DeepVSC achieves substantial improvements in both Multi-Scale Structural Similarity (MS-SSIM) and Learned Perceptual Image Patch Similarity (LPIPS) metrics than the H.265 standard, particularly under low channel signal-to-noise ratio (SNR) or multi-path channels, highlighting the significantly enhanced transmission capabilities of our approach.
Semantic communication has garnered increasing attention due to its ability to shift the focus from pixel-level transmission to the transfer of fundamental semantic information representing the core content of images. This paper proposes a novel reinforcement learning-based layered semantic coding (RL-SC) method aimed at optimizing image communication systems under bandwidth constraints. By leveraging deep reinforcement learning (DRL), the proposed method efficiently allocates semantic bits to maximize semantic fidelity while minimizing bitrates. By incorporating semantic segmentation and image reconstruction networks, the framework utilizes both semantic maps and residual information to enhance the image coding process. Experiments demonstrate that the proposed method outperforms traditional image compression techniques and layered image encoding methods without reinforcement learning in preserving semantic content and achieving high-quality image reconstruction. In particular, the proposed method excels in low-bitrate scenarios, effectively maintaining both semantic accuracy and perceptual quality.
10 April 2025 in IEEE Wireless Communications Letters
The large volume of data associated with virtual reality (VR) video presents significant challenges for wireless transmission. This letter proposes an efficient transmission system for VR video, based on semantic vector quantization (VQ) and moving object tracking, termed SeVQ-VR. This system not only substantially reduces the transmission data volume but also ensures full compatibility with existing communication systems. Specifically, it prioritizes the transmission of semantically significant moving objects, optimizing bandwidth utilization. Additionally, it utilizes semantic vector quantization to further compress video feature data by transmitting only the feature indices, while maintaining accurate reconstruction of the content. Simulation results indicate that SeVQ-VR outperforms the H.265 standard in terms of both Multi-Scale Structural Similarity (MS-SSIM) and Learned Perceptual Image Patch Similarity (LPIPS) metrics, especially under low channel signal-to-noise ratio (SNR) in multi-path channels.
1 January 2026 in IEEE Transactions on Cognitive Communications and Networking
Fluid antenna systems (FASs) can enhance communication system performance by flexibly switching antenna positions to ports with favorable channel conditions. To fully realize this benefit, acquiring accurate channel state information is essential. Existing Transformer-based FAS channel estimation methods mainly rely on conventional self-attention (SA), whose quadratic complexity poses challenges to network efficiency, especially when the number of switchable ports is large. To address this issue, we propose a novel transposed angular attention (TAA) that reduces complexity to linear levels while retaining global modeling capabilities. Unlike conventional SA directly modeling relationships between ports, TAA first disentangles the combined contributions of multi-path components (MPCs) at each port through an angular-domain transformation, and then, via a transposed operation, shifts the attention focus to model the relationships between angular features underlying all disentangled MPCs. We further extend TAA into transposed angular self- and cross-attention mechanisms. Building on these, we develop an efficient TAAformer network for FAS channel estimation, a UNet-like encoder-decoder architecture with multi-scale feature capabilities. Simulation results show that the proposed TAAformer can achieve higher channel estimation accuracy compared to existing methods.
1 January 2026 in IEEE Wireless Communications Letters