Adaptive Modulation and Coding for Underwater Acoustic Communication Based on Neural Networks

Due to the complexity and variability of the marine environment, the modulation and coding schemes used in underwater acoustic communication do not match the actual channel, which can lead to a decrease in communication accuracy or an increase in power consumption. Therefore, adaptive modulation and coding in underwater acoustic communication is very necessary to improve communication throughput. In this paper, we propose a neural network-based adaptive modulation and coding method for underwater acoustic communication. Firstly, we conduct channel modeling to generate diverse channels. Next, we simulate the error rate of different modulation and coding schemes under different channel conditions, and obtain a dataset with each channel corresponding to an optimal modulation and coding scheme. Finally, we use the obtained data set for training, and compare the improved neural network model proposed in this paper with four mainstream models for analysis. The simulation results show the improved neural network has optimized performance on the test set and can find better modulation and coding schemes based on channel characteristics.