基于改进SAMP算法的UWB多径信道估计

稀疏度自适应匹配追踪（SAMP）算法重构过程中存在其迭代终止条件设置不够合理的情况,需要对SAMP算法进行改进.在信道稀疏度未知时,改进SAMP算法依据残差之差的相对能量小于设定的停止门限来终止迭代过程,通过自适应调整可变步长逐步逼近信道的稀疏度,从而实现了重构UWB信道.仿真结果表明,改进SAMP算法低信噪比时重构精度高于SAMP算法,具有更好的重构性能和广泛的实用性.

Multipath Channel Estimation of UWB Based on Improved SAMP Algorithm

For the signal sparsity required as the prior information and the iteration termination condi- tion less than a predetermine threshold in the Sparse Adaptive Matching Pursuit （SAMP） algorithm, the SAMP algorithm is improved. When the sparsity of channel is unknown, the iteration termination condition of the improved SAMP algorithm based on the relative energy of residual subtraction is less than a stop threshold, the sparse decomposition is adaptive terminated, and the sparsity of channel is adaptive adjusted by a variable step to approximate step by step, thus the improved SAMP algorithm accurately reconstructs the UWB channel. Simulation results show that the improved SAMP algorithm has a higher reconstruction accuracy than the SAMP algorithm in low SNR, and better estimation per- formance and could be used to reconstruct sparse signal effectively.