采用确定性测量矩阵的宽带压缩采样的研究

Broadband Compressive Sampling with Deterministic Measurement Matrix

在对宽带信号进行处理的过程中,常运用压缩感知的理论来获得有效的信息。而在实践压缩感知理论的压缩采样的结构中,调制宽带转换器的采样结构更加适合用于处理宽频带信号。文中研究了调制宽带转换器的压缩采样原理,也介绍了随机测量矩阵和确定性测量矩阵。分别将随机矩阵和确定性矩阵作为该调制宽带转换器的测量矩阵,对比分析了该采样结构的重构性能。研究了在确定性测量矩阵的基础上,该采样结构在折叠和非折叠条件下的信号重构性能,同时,也对系统的通道数目对性能重构和信噪比的影响进行了补充分析。

In the processing of broad signal, the theory of compressed sensing is usually used to acquire effective information. However, of among the compressive sampling structures in practicing this theory, the sampling structure for modulated wideband converter is more suitable to processing broadband signal. The compressive sampling principle of modulated wideband converter is described in this paper, random measurement matrix and determinate measurement matrix also discussed in this paper. With random matrix and deterministic matrix respectively as the measurement matrix of this structure, the reconstruction performances of these two matrixes are compared and analyzed, and based on deterministic measurement matrix, the reconstruction performances in folded and unfolded conditions technically explored. Meanwhile, the influence of channel numbers on the performance reconstruction and signal-to-noise ratio is also analyzed.