On Network-Aware Semantic Communication and Edge-Cloud Collaborative Intelligence Systems

2025-12-22

AI Summary

To address the joint optimization challenge for 6G intelligent services under stringent constraints on bandwidth, latency, and resource availability, this paper proposes a synergistic semantic communication and edge-cloud collaborative intelligence framework. It replaces conventional bit-level transmission with task-oriented semantic representation and introduces, for the first time, a network-aware semantic encoding strategy alongside an AI-native edge-cloud architecture, deeply integrating zero-trust security, physical-layer security, and learning-driven network control. Key technical contributions include semantic representation learning, resource-adaptive semantic coding, joint semantic-resource scheduling, edge-efficient lightweight inference, and cloud-based collaborative model training. Experimental results demonstrate a 3.2× improvement in end-to-end semantic transmission efficiency, a 57% reduction in task-level latency, and significant enhancements in system trustworthiness, scalability, and robustness against disturbances.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Semantic communication and edge-cloud collaborative intelligence are increasingly recognized as foundational enablers for next-generation intelligent services operating under stringent bandwidth, latency, and resource constraints. By shifting the communication objective from bit-perfect delivery toward the transmission of task-relevant semantic representations, semantic communication enables adaptive tradeoffs among communication overhead, inference accuracy, computational load, and end-to-end latency. This survey provides a comprehensive and system-level synthesis of recent advances in semantic communication at the edge-cloud interface, encompassing architectural models for collaborative intelligence, representation learning and semantic abstraction techniques, network-aware and resource-adaptive semantic encoding strategies, and learning-driven optimization and orchestration mechanisms. Beyond efficiency considerations, the survey situates semantic communication within practical operational contexts, including security, trust, resilience, and scalability, drawing connections to zero-trust networking, physical-layer security, and emerging edge-cloud control paradigms. Finally, open challenges and research directions are identified, highlighting the role of semantic communication as a key building block for AI-native networking and 6G-ready intelligent systems.

Problem

• Optimizing communication efficiency for intelligent services under constraints.
• Developing adaptive semantic encoding strategies for edge-cloud systems.
• Addressing security and scalability in network-aware semantic communication.

Innovation

• Transmits task-relevant semantic representations instead of raw bits
• Uses network-aware and resource-adaptive semantic encoding strategies
• Integrates learning-driven optimization and orchestration mechanisms