Channel estimation is very challenging for multiple-input and multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems in high mobility environments with non-stationarity channel characteristics. In order to handle this problem, we propose a deep learning (DL)-based MIMO-OFDM channel estimation algorithm. By performing offline training to the learning network, the channel state information (CSI) generated by the training samples can be effectively utilized to adapt the characteristics of fast time-varying channels in the high mobility scenarios. The simulation results show that the proposed DL-based algorithm is more robust for the scenarios of high mobility in MIMO-OFDM systems, compared to the conventional algorithms.

中文翻译：

---

基于深度学习的快速时变 MIMO-OFDM 系统信道估计算法

对于具有非平稳信道特性的高移动性环境中的多输入多输出正交频分复用 (MIMO-OFDM) 系统，信道估计非常具有挑战性。为了解决这个问题，我们提出了一种基于深度学习 (DL) 的 MIMO-OFDM 信道估计算法。通过对学习网络进行离线训练，可以有效利用训练样本生成的信道状态信息（CSI）来适应高移动性场景下快速时变信道的特性。仿真结果表明，与传统算法相比，所提出的基于 DL 的算法对于 MIMO-OFDM 系统中的高移动性场景具有更强的鲁棒性。