Long video delivery over wireless links whose SNR (Signal-to-Noise Ratio) falls below 5 dB often hits a cliff: conventional codecs combined with FEC remain decodable, but the decoded frames collapse into blocky mosaics that are semantically unusable. This paper presents AdaGenCom (Adaptive Generative Communication), a multi-modal semantic communication framework that preserves meaning when pixel recovery becomes infeasible. AdaGenCom decomposes a source video into (i) structural semantics as sparse visual anchors (motion-aware keyframes and optional flow cues), (ii) textual semantics as compact scene descriptions, and (iii) conditional semantics as lightweight channel-state metadata. Each stream is transmitted with learned joint source-channel coding to enable graceful degradation. At the receiver, a pre-trained video diffusion model (LTX-Video) reconstructs the full sequence while a Channel-Aware Parameter Mapping (CAPM) mechanism adapts diffusion sampling-denoising steps, guidance scale, and sampling stochasticity-to the estimated channel quality without retraining. On 400+ frame sequences from UVG, AdaGenCom maintains CLIP similarity above 0.96 at SNRs where H.264/H.265 outputs become unrecognizable (CLIP below 0.80). These results indicate that generative priors with channel-aware inference provide a practical path for long-video delivery under extreme wireless conditions.
Bitrate adaptation (or ABR) plays a crucial role in shaping the user QoE of low-latency live streaming (LLLS) applications. However, the unique characteristics of modern WebRTC-based LLLS systems render traditional ABR paradigms inefficient or even ineffective in real-world scenarios. This motivates us to revisit the bitrate adaptation problem within this complex yet realistic context and propose Salmon, an innovative bitrate adaptation framework designed for WebRTC-based LLLS applications. Salmon pragmatically addresses challenges arising from new QoE objectives, two-stream handovers, user interaction behaviors, and application-specific signal semantics. Deployed on a leading e-commerce LLLS platform, Salmon demonstrates significant performance gains over state-of-the-art algorithms. Notably, in low-bandwidth conditions, Salmon reduces startup delay by 17.7%, stall by 11%, frame jumps by 41.2%, and switching rate by 25.9x.
Immersive technologies promise to revolutionize communication through enhanced sense of presence and interactivity. To enable interaction, reliable low-latency transport mechanisms are needed to handle the large volumes of data created by complex 3D objects. In this paper, we propose an open-source, codec-independent, selective forwarding unit (SFU) for real-time volumetric video streaming using WebRTC. For evaluation purposes, we provide a reference client implementation by extending VR2Gather, a TCP-based system for immersive communication. We conduct extensive evaluations using both new and existing datasets to compare the performance of WebRTC against TCP-based protocols in an emulated testbed environment. The evaluations demonstrate that WebRTC outperforms other protocols in high-latency scenarios and adapts video quality to user movement 13% and 36% faster than its TCP-based counterparts in networks with 5 ms and 10 ms of network latency, respectively.
Cloud gaming enables high-fidelity gaming on diverse devices by streaming rendered frames from powerful servers. However, ensuring a consistent and responsive visual experience remains a critical challenge due to the mismatch between variable frame arrival rates and fixed display refresh rates. Traditional Vertical Synchronization (VSync) eliminates screen tearing but introduces significant buffering latency, which is further exacerbated by network jitter in cloud environments. Consequently, gamers are often forced to choose between visual artifacts (tearing) and poor responsiveness (latency). To reconcile these conflicting objectives, we propose Perception-Aware Synchronization (PASync), a novel frame scheduling algorithm that minimizes buffering latency while strictly enforcing zero screen tearing. Our key insight, derived from user studies, is the existence of a perceptually-maximum Frame Rate (pFR) beyond which additional frames do not perceptibly improve smoothness. PASync leverages this by effectively discarding perceptually redundant frames to actively drain the frame buffer. Evaluations using real network traces demonstrate that PASync reduces average buffering latency by 43.5% and tail latency by 33.9% compared to standard VSync.
Liquid leakage detection is critical for product quality and user safety, yet existing methods require line-of-sight (LoS) or direct contact with the liquid, or could even introduce additional health or safety risks. In this paper, we propose SARLiquid, a novel mmWave-based through-package liquid leakage detection method. SARLiquid leverages mmWave signals to see through packaging and identify leakage by reconstructing mmWave images. We employ synthetic aperture radar (SAR) technique to enhance imaging resolution and tame multipath effects. We also develop a dedicated algorithm to calibrate the discontinuous phase in SAR imaging results, and propose a deep learning model for liquid leakage detection and liquid identification. We implement SARLiquid and evaluate it across a wide range of scenarios. Results show that SARLiquid achieves average accuracies above 93% for both liquid leakage detection and liquid identification, 14.70% and 9.98% higher than the baselines, respectively.
