一种改进的变步长SAMP压缩感知浅海水声信道估计算法

An Improved Variable Step Size SAMP Compressed Sensing Shallow Water Acoustic Channel Estimation Algorithm

针对稀疏度自适应匹配追踪(Sparsity Adaptive Matching Pursuit,SAMP)算法估计性能过度依赖初始步长的问题,提出了改进的变步长稀疏度自适应匹配追踪(Improved Variable Step-size Sparsity Adaptive Matching Pursuit,IVSSAMP)算法,并将IVSSAMP算法应用于水声正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统的压缩感知信道估计框架中。IVSSAMP算法通过残差条件对步长选取进行分类,在不同条件下分别引入微调因子设置残差与测量向量、残差与噪声之间的阈值来调整步长,实现变步长和稀疏度的自适应,同时规避了过度估计情况的发生。实验结果表明,IVSSAMP算法利用步长的灵活选取实现了估计精度与运行速度之间的平衡,降低了初始步长对算法性能的影响,具有较好的稳健性和应用性。

To address the problem of excessive dependence on the initial step size in the estimation performance of the SAMP(Sparsity Adaptive Matching Pursuit)algorithm,an IVSSAMP(Improved Variable Step-size Sparsity Adaptive Matching Pursuit)algorithm is proposed,which is applied to the compressed sensing channel estimation framework of the OFDM(Orthogonal Frequency Division Multiplexing)system in underwater acoustics.The IVSSAMP algorithm classifies the step size selection through residual conditions,and introduces fine-tuning factors to adjust the thresholds between residuals and measurement vectors,and the thresholds between residuals and noise,realizing self-adaption of variable step size and sparsity,while avoiding the occurrence of overestimation.Experimental results indicate that the IVSSAMP algorithm balances estimation accuracy and operating speed by flexibly selecting the step size and reduces the impact of the initial step size on the algorithm performance,thus exhibits good robustness and applicability.