Insights Into Modulation and Coding Scheme in Mobile Communication

With the growing demand for mobile communications and the advancement of technology, reducing energy consumption has become a critical challenge. Overservice, which leads to unnecessary energy waste, is attracting significant attention from both academia and industry. An optimized modulation and coding scheme (MCS) can reduce the energy consumption per unit of information, thereby preventing overservice in the air interface. This scheme aims to achieve quality of service (QoS) by utilizing transmission power as low as possible. This paper reviews the effects of modulation and coding, the efficiency of adaptive modulation and coding (AMC) schemes, and the application of machine learning (ML) to MCS selection. Reinforcement learning (RL) selects MCS through interaction with the environment without requiring large training datasets. We compare the performance of deep Q-network (DQN) and deep convolutional neural network (DCNN) in MCS selection. The analysis shows that the throughput performance achieved by the DQN algorithm is closer to the optimal than that of DCNN. Therefore, RL is recommended both theoretically and practically as the core algorithm for MCS selection.