Wireless Semantic Transmission via Revising Modules in Conventional Communications
Peiwen Jiang, Chao-Kai Wen, Shi Jin, Geoffrey Ye Li
Abstract
Semantic communication has become a widely researched area due to its high spectrum efficiency and error correction capabilities. Some studies used deep learning to extract semantic features, typically resulting in end-to-end semantic communication systems that are difficult to adapt to varying wireless environments. Therefore, new semantic-based coding methods and performance metrics were investigated. Designed semantic systems incorporate various modules similar to those for traditional communication but with enhanced functionality. This study discusses recent advancements in state-of-the-art semantic communication, utilizing conventional modules in wireless systems. Moreover, through two examples, this study shows that traditional hybrid automatic repeat request and modulation methods can be re-purposed for novel semantic coding and metrics to further improve the performance of wireless semantic communication. Lastly, this study identifies some open research questions.