基于混合范数约束的非均匀稀疏水声信道估计方法*

水声信道具有明显的簇状稀疏特性,即稀疏的信道冲激响应大部分为零或接近零的小值系数,而非零值系数是以簇的形式非均匀分布于时延域。为此本文提出了一种基于非均匀混合范数约束仿射投影算法的水声信道估计方法。该方法首先根据信道簇状结构对其进行非均匀分组,基于此将范数约束规则加入仿射投影算法中,具体方法为：对簇状部分施加范数约束,有效提高系数间的相关性,而簇状结构与其他零值抽头之间利用范数约束实现了整体的稀疏特性。数值仿真以及深海远程水声通信实验数据处理结果表明了本文所提出的水声信道估计算法相较现有稀疏信道估计方法能够实现更快的收敛速度以及更高精度的信道估计结果。

Hybrid Norm Constraint based Non-Uniform Sparse Estimation for Underwater Acoustic Channels

The underwater acoustic channels(UAC) have cluster-sparse characteristics in nature<sub></sub><sub>1]</sub>, that is, the sparse channel impulse response is mostly zero or near-zero taps, while only a few none-zero ones are unevenly distributed in the time domain in the form of clusters. In this paper, a non-uniform hybrid norm constraint based improved proportionate affine projection algorithm(IPAPA) is proposed for underwater acoustic channel estimation. Firstly, the channel partition are initialized according to the UAC cluster structure. Then the mixed -norm is added on the IPAPA to promote the cluster-sparsity of the UAC: it improves correlation among coefficients inside each cluster via the norm and uses the norm to realize the overall sparsity. Numerical simulation results and the data processing results of long range deep-water acoustic communication experiment show that the proposed UAC estimation algorithm can achieve a better performance in terms of convergence speed and estimation accuracy compared to existing sparse channel estimation methods.