中文摘要：

多通道被动时间反转可通过对信道多径进行时间、空间聚焦实现信道匹配，但低信噪比、时变特性造成的信道特性失配对时间反转处理的性能造成严重影响。在垂直阵接收的基础上，本文采用码片级信道估计获取水声信道特性并进行周期性更新，并采用己判决码元产生的扩频码片作为信道估计训练序列，结合稀疏信道估计算法抑制零值抽头上的估计噪声，从而可有效改善时变、低信噪比条件下的被动时间反转处理的时、空多径聚焦效果，提高扩频通信性能。通过湖试实验比较了采用稀疏信道估计、传统信道估计算法的时反扩频接收机，以及经典直扩接收机的通信性能，实验结果表明：本文方案可在低信噪比获得较好的性能，并有效抑制时变信道对时反扩频通信性能的影响。

英文摘要：

In this paper, passive time reversal （TR） direct sequence spread spectrum underwater communica- tion is investigated to achieve further performance enhancement via the multipath temporal-spatial focusing. However, the effects of TR will be deteriorated significantly by mismatch of the channel caused by time vari- ations and low SNR. Aiming at the problem, in this paper we adopted the chip-rate channel estimation to obtain and periodically update the channel response to facilitate the TR processing and DSSS demodulation. With the DS chips produced by previously decided bits used as the training sequence of channel estimation, the sparse channel estimation algorithm is employed to alleviate the estimation noise at zero taps, thus impacts of time-varying channel and low SNR are mitigated effectively. Lake trials with different types of receivers, including the proposed scheme as well as the classic DSSS receiver, are carried out to verify the effectiveness of the proposed scheme.