Emotion-Aided Semantic Communication System for Reliable Semantic Recovery Under Low SNR

Abstract

The vision for sixth generation wireless communication entails an intelligent and personalized communication scheme that addresses individual needs. To accomplish this, semantic communication with the help of machine learning and natural language processing technologies can play a crucial role. However, existing semantic communication systems fail to model the interaction between semantics and emotion, which is vital for personalized communication. This letter proposes an emotion-aided semantic communication system that utilizes an external knowledge base for emotion extraction at the transmitter and adopts emotion recognition and selection at the receiver for semantic recovery. Simulation results demonstrate the effectiveness of incorporating emotional knowledge into semantic communication, especially in the low signal-to-noise ratio (SNR) region.