Exploiting Computation Dimension for Communication Improvement in Semantic Communication Networks
Guangyuan Zheng, Miaowen Wen, Xueqin Jiang, Zhiguo Ding · 2024
As a new paradigm focusing on transmitting the meaning of information, semantic communications (SCs) have revealed significant potential in alleviating network congestion and saving energy consumption. By extracting a small-size semantic feature from the large-size transmitted messages, the communication loads can be converted into computation costs through SCs. In this paper, we investigate the use of computation dimensions to improve the communication performance in an uplink SC-based system. The SC user first compresses its original data via local computation during other users’ transmission time, and then transmits it to the base station within the assigned time. We formulate an optimization problem to minimize the energy consumption of all users by considering the relationship between computation and communication. To solve this problem, we first propose a general low-complexity algorithm that can be applied to different SC models. To gain more insights, we then derive and analyze the optimal solutions in two special cases: equal-time allocation and two-user transmission. Simulation results show that our proposed SC-based scheme achieves high performance in terms of saving the energy consumption compared to conventional communications.