当前位置：
X-MOL 学术
›
Signal Process.
›
论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Channel Prediction based Temporal Multiple Sparse Bayesian Learning for Channel Estimation in Fast Time-varying Underwater Acoustic OFDM Communications
Signal Processing ( IF 4.4 ) Pub Date : 2020-10-01 , DOI: 10.1016/j.sigpro.2020.107668 Gang Qiao , Qingjun Song , Lu Ma , Zongxin Sun , Jiarong Zhang
Signal Processing ( IF 4.4 ) Pub Date : 2020-10-01 , DOI: 10.1016/j.sigpro.2020.107668 Gang Qiao , Qingjun Song , Lu Ma , Zongxin Sun , Jiarong Zhang
Abstract In recent years, the multi-blocks joint channel estimation methods for underwater acoustic (UWA) OFDM systems have been proposed to improve the performance by using time correlation across consecutive OFDM blocks. However, the assumption of common sparse support (multi-path delays are invariant across consecutive blocks) in these methods is difficult to be met in fast time-varying channels. Therefore, we propose a channel prediction based joint channel estimation method for UWA fast time-varying channels, where multi-path delays and gains change significantly across consecutive OFDM blocks. Firstly we define a channel offset parameters model by the clustering property of UWA channels, and adopt Orthogonal Matching Pursuit (OMP) algorithm to estimate the channel offset parameters. Then we reconstruct a virtual current received signal based on the prediction method. Finally we combine the virtual received signal and the actual one into a joint estimation model, and utilize temporal multiple sparse Bayesian learning (TMSBL) method to jointly estimate the channel. Results of simulations and sea trial demonstrate the effectiveness of the proposed method in fast time-varying UWA channels, which achieves better performance than both the TMSBL method which is based on the joint processing of multi-blocks, and the OMP method which is based on block-by-block processing.
更新日期：2020-10-01
京公网安备 11010802027423号