一种稀疏信道非零抽头检测的新方法

为了提高稀疏信道估计时的精确度,需对其非零抽头位置进行检测。传统的匹配追踪系列算法在检测时须以信道的稀疏度为先验信息,或给出严格的迭代终止条件。针对这一情况,引入了一种新的方法——分层检测算法实现信道非零抽头位置检测。实验仿真表明,分层检测算法在SNR大于10 dB时检测信道非零抽头位置的性能是与匹配追踪算法相当的,而当信噪比较低时,其性能更优。在算法实现过程中同时分析了影响其性能的原因,并在此基础上对其进行了改进。

A novel method for nonzero taps detection of sparse channel

To improve the accuracy of sparse channel estimation, nonzero taps detection is required. In the matching pursuit algorithm, the sparsity of channel must be taken as prior information, or strict stop criterion of algorithm iteration must be given. To avoid the mentioned problem, the multistage detection algorithm is introduced to obtain the location information. Experiment result shows that the multistage detection algorithm performs almost equivalently to matching pursuit algorithm when it is used to detect the nonzero taps under SNR>10 dB, and its performance will be much better than that of matching pursuit algorithm when the SNR is low. And the reasons causing the unstable performance of the algorithm are also analyzed, then an improved method is provided.