深海远程正交频分复用水声通信簇约束的分布式 压缩感知信道估计*

在深海远程正交频分复用(Orthogonal Frequency Division Multiplexing，OFDM)水声通信中，信道时延较长，导致信道频率选择性衰落严重，传统的压缩感知信道估计算法性能大幅下降。为此，本文利用深海远程信道在一定时间内具有相关性并呈较稳定簇状分布的特点，在分布式压缩感知信道估计中引入簇区域信息，进行簇约束的多数据块联合稀疏信道估计，提出一种簇约束的分布式压缩感知信道估计方法，并在深海开展了定点远程水声通信实验进行验证，实验结果表明，与传统的分布式压缩感知信道估计算法相比，本方法的能够降低50%左右的误码率。

Cluster restricted distributed compressed sensing channel estimation for long-range OFDM underwater acoustic communication in deep sea

In long-range Orthogonal Frequency Division Multiplexing (OFDM) communication of deep sea, underwater acoustic channel is characterized by long delay spread, which results in serious selective fading of channel frequency. In this case, traditional compressed sensing algorithm performance deteriorates significantly. To solve this problem, we take advantage of the characteristics of long-range and deep-sea channels that are correlated and distributed in stable clusters within a certain period of time, and introduces cluster region information into diatributed compressed sensing channel estimation, to perform cluster-restricted multi-data block joint sparse channel estimation. So, we propose a distributed compressed sensing channel estimation method with cluster constraint in this paper. To validate the effectiveness of this method, a point to point long-range communication experiment was carried out in deep sea, the experimental results show that with the proposed method, the bit error rate can steadily drop to 50% compared with traditional distributed compressed sensing channel estimation algorithms.