A Deep Learning-Based Method for Underwater Acoustic Channel Estimation
In the field of underwater communications,the propagation of electromagnetic waves is greatly restricted,making underwater acoustic communication the primary means of commu-nication.Utilizing the advantages of Orthogonal Frequency Division Multiplexing(OFDM)modulation technique,communication in complex and variable underwater acoustic channels has become one of the commonly used methods.To overcome the adverse conditions of underwater acoustic channels,channel estimation is required to obtain the channel state information for channel equalization and improved communication performance.The paper proposes a deep learning-based method for underwater acoustic channel estimation.An improved multi-layer perceptron model is designed for underwater channel estimation.Compared to traditional meth-ods,the proposed method achieves improvements in estimation performance and robustness,while also demonstrating the beneficial role of residual learning for enhancing the performance of deep learning-based channel estimation.
NETL
NSTL
万方数据
