Adaptive Modulation and Retransmission Scheme for Semantic Communication Systems

Traditional adaptive modulation scheme aims to maximize the spectral efficiency by selecting the appropriate modulation scheme under the premise of perfect bit-level data transmission. However, in task-oriented semantic communication systems, imperfect transmission can still lead to a good inference performance of semantic tasks due to the error correction capability of neural networks. In this paper, we propose novel adaptive modulation and retransmission schemes to maximize the spectral efficiency while guaranteeing the performance of semantic tasks. Specifically, we first introduce the robustness verification problem in semantic communication systems to analyze the robustness of neural network inference. We then formulate and solve the modulation scheme selection problem constrained by a robustness probability threshold. Consequently, a novel adaptive modulation scheme is developed to maximize the spectral efficiency while guaranteeing the goal of semantic communication. We also develop a retransmission scheme using existing combining techniques to further increase the data rate under harsh channel conditions. Extensive simulations are performed in unencoded and encoded semantic communication systems to validate the effectiveness of the proposed schemes.