一种改进的稀疏度自适应弱选择匹配追踪算法

An modified sparsity adaptive weak selection matching pursuit algorithm

摘要:

针对实际水声信道无法获知先验稀疏度和导频资源问题，提出一种改进的稀疏度自适应弱选择匹配追踪算法(MSASWOMP)。将稀疏度初始估计作为初始支撑集的大小，对原子进行阈值弱选择得到的原子支撑集作为回溯筛选的候选集；再以初始支撑集大小为回溯开始初始条件值进行二次筛选；最后利用变阶段步长方法进行稀疏度逐步精确估计，自适应更新回溯开始条件值。仿真实验分析了阈值参数、稀疏度估计步长和导频数目对于MSASWOMP算法的影响，结果表明，该算法能以更少的导频数目获得更精确的信道估计值，节省导频资源的同时，其均方误差(MSE)优于传统算法。

Abstract:

A Modified Sparsity Adaptive Stagewise Weak Orthogonal Matching Pursuit (MSASWOMP) algorithm is proposed to solve the problem of a priori sparsity and pilot resources for the actual underwater acoustic channel. Firstly, the initial sparsity estimation is used as the size of the initial support set, and the atom support set is obtained as the candidate set of backtracking screening by threshold weak selection of atoms. Then, the initial support set size is used as the initial condition value of backtracking to conduct secondary screening. Finally, the variable stage step size method is utilized to estimate the sparsity gradually and update the backtracking start condition value adaptively. The simulation results show that the proposed algorithm can obtain more accurate channel estimation with less pilot number, which can save pilot resources, and its Mean Square Error(MSE) is smaller than that of the traditional algorithm.