From Semantic Communication to Semantic-Aware Networking: Model, Architecture, and Open Problems

TLDR

Current communication systems, rooted in Shannon theory, ignore semantics, while 5G and beyond aim to support content‑driven services, sparking interest in semantic communication that focuses on message meaning. The article reviews existing semantic communication frameworks, identifies key challenges, and proposes a federated edge‑intelligence architecture to enable resource‑efficient semantic‑aware networking. The proposed architecture offloads semantic encoding and decoding to edge servers, using intermediate results to protect proprietary models while enabling federated edge intelligence. We find that semantic processes can be resource‑intensive, but simulations demonstrate that the federated edge architecture reduces resource consumption and markedly improves communication efficiency.

Abstract

Existing communication systems are mainly built based on Shannon's information theory, which deliberately ignores the semantic aspects of communication. The recent iteration of wireless technology, 5G and beyond, promises to support a plethora of services enabled by carefully tailored network capabilities based on contents, requirements, as well as semantics. This has sparked significant interest in semantic communication, a novel paradigm that involves the meaning of messages in communication. In this article, we first review classic semantic communication frameworks and then summarize key challenges that hinder its popularity. We observe that some semantic communication processes such as semantic detection, knowledge modeling, and coordination can be resource-consuming and inefficient, especially for communication between a single source and a destination. We therefore propose a novel architecture based on federated edge intelligence for supporting resource-efficient semantic-aware networking. Our architecture allows each user to offload computationally intensive semantic encoding and decoding tasks to edge servers and protect its proprietary model-re-lated information by coordinating via intermediate results. Our simulation result shows that the proposed architecture can reduce resource consumption and significantly improve communication efficiency.