Adaptive Semantic-Bit Communication for Extended Reality Interactions
Chaowei Wang, Yehao Li, Feifei Gao, Danhao Deng, Jisong Xu, Yuhan Liu, Weidong Wang
Abstract
Semantic communication is a novel paradigm that conveys intention or goal from the source to the destination. It can greatly improve communication efficiency, especially for the applications that require extremely low latency and high reliability, such as augmented reality (AR), virtual reality (VR) or extended reality (XR). An adaptive semantic-bit communication structure based on resource efficiency enhancement for XR is proposed, in which part of the XR users employ semantic communication, while others employ the conventional way. We utilize adaptive communication and power allocation to maximize the system-level achievable performance indicated by equivalent semantic rate. The formulated problem is addressed by a two-step optimization. First, we propose a signal-to-interference-plus-noise ratio (SINR)-based paradigm selection scheme as the semantic communication outperforms the conventional way in low and moderate SINR regimes. Then we propose a genetic algorithm-based power allocation to solve the non-convex optimization. Simulation results demonstrate that the proposed scheme achieves a higher equivalent semantic rate against the baseline schemes.