Interplay of Semantic Communication and Knowledge Learning

TL;DR

This work addresses the challenge of enhancing semantic communication by infusing structured knowledge. It proposes a KG-enhanced SemCom system with a receiver-side Transformer knowledge extractor that uses a knowledge vector to assist decoding while keeping the transmitter unchanged, and it extends this with a KG evolving-based approach that introduces a unified semantic space and contrastive learning to cope with dynamic knowledge bases. Additionally, the study explores LLM-assisted data augmentation to enrich the KG without manual annotation, validating gains on WebNLG data in low-SNR regimes. Across extensive simulations, the proposed methods yield improved BLEU and semantic similarity metrics, demonstrating improved robustness and adaptability in evolving knowledge scenarios and challenging channels, with practical implications for knowledge-aware communication and reasoning-enabled transmission.

Abstract

In the swiftly advancing realm of communication technologies, Semantic Communication (SemCom), which emphasizes knowledge understanding and processing, has emerged as a hot topic. By integrating artificial intelligence technologies, SemCom facilitates a profound understanding, analysis and transmission of communication content. In this chapter, we clarify the means of knowledge learning in SemCom with a particular focus on the utilization of Knowledge Graphs (KGs). Specifically, we first review existing efforts that combine SemCom with knowledge learning. Subsequently, we introduce a KG-enhanced SemCom system, wherein the receiver is carefully calibrated to leverage knowledge from its static knowledge base for ameliorating the decoding performance. Contingent upon this framework, we further explore potential approaches that can empower the system to operate in evolving knowledge base more effectively. Furthermore, we investigate the possibility of integration with Large Language Models (LLMs) for data augmentation, offering additional perspective into the potential implementation means of SemCom. Extensive numerical results demonstrate that the proposed framework yields superior performance on top of the KG-enhanced decoding and manifests its versatility under different scenarios.

Figures (12)

• Figure 1: The framework of the SemCom system.
• Figure 2: The KG-enhanced semantic decoder.
• Figure 3: The BLEU and Sentence-BERT score versus SNR for the KG-enahnced SemCom system based on Transformer.
• Figure 4: The BLEU and Sentence-BERT score versus SNR for the KG-enahnced SemCom system based on UT.
• Figure 5: The precision and recall rate versus SNR for the KG-enhanced SemCom system based on Transformer.