Joint block estimation and attention-based long short-term memory network for doppler shift mitigation in UWA communication systems

Aiming to address the performance degradation of OFDM systems caused by Doppler shift in underwater acoustic (UWA) communications, this paper proposes a two-step algorithm that combines the traditional block estimation method with an Attention-based Long Short-Term Memory Network (ALSTM) to mitigate the Doppler effect. In the first step, the Doppler factor is coarsely estimated using a conventional block estimation technique based on linear frequency modulation (LFM) signals. The estimated value is then used to resample the received data, which is subsequently fed into a neural network model to predict the carrier frequency offset (CFO) of the resampled signal. Simulation results demonstrate that the proposed algorithm outperforms both traditional methods and basic neural network approaches in terms of mean square error (MSE) and system bit error rate (BER), and achieves superior accuracy and robustness in Doppler shift estimation, offering a reliable solution for communication in complex UWA environments.