Semantic Communication Systems with a Shared Knowledge Base
Peng Yi, Yang Cao, Xin Cai Kang, Ying‐Chang Liang · 2023
With the ability to improve transmission efficiency, deep learning-enabled semantic communication is regarded as a promising candidate for future 6G networks. Most existing semantic communication systems are based on the end-to-end architecture that is considered as a black box, leading to the lack of explainability. To tackle this issue, in this paper, a novel semantic communication system with a shared knowledge base is proposed for text transmissions. To be specific, a textual knowledge base for semantic communication is constructed based on a similarity-comparison method, in which a pre-configured threshold can be leveraged to control the size of the knowledge base. The proposed system integrates the message and corresponding knowledge from the shared knowledge base to obtain the residual information, which enables the system to transmit fewer symbols without semantic performance degradation. The simulation results demonstrate that our proposed approach can outperform existing baseline methods in terms of transmitted data size and the sentence similarity.