程云霄,,
覃亚丽
浙江工业大学信息工程学院 ??杭州 ??310023
基金项目:国家自然科学基金(61675184)
详细信息
作者简介:李如春：女，1968年生，副教授，研究方向为多媒体信号处理
程云霄：男，1994年生，硕士生，研究方向为压缩感知信号处理
覃亚丽：女，1963年生，教授，研究方向为光学信号处理
通讯作者:程云霄 com_xd@163.com
中图分类号:TN919.4计量
文章访问数:1152
HTML全文浏览量:452
PDF下载量:54
被引次数:0
出版历程
收稿日期:2018-05-25
修回日期:2018-11-13
网络出版日期:2018-11-22
刊出日期:2019-04-01
Sensing Matrix Optimization for Sparse Signal under Structured Noise Interference
Ruchun LI,Yunxiao CHENG,,
Yali QIN
College of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, China
Funds:The National Natural Science Foundation of China (61675184)
摘要
摘要:针对具有结构性噪声干扰的稀疏信号处理问题，该文提出一种基于贝叶斯理论的感知矩阵优化设计方法。结合具有加性干扰的稀疏信号模型，通过对感知矩阵进行能量约束，最小化信号的后验协方差矩阵的迹，实现感知矩阵的优化设计。仿真不同信号稀疏度和重构算法时，感知矩阵优化对信号重构误差和重构时间的影响；分析信号先验信息存在偏差时，感知矩阵优化对重构效果的影响。仿真结果表明，优化后的感知矩阵能够更好地获取稀疏信号中的重要信息，信号重构精度的均方误差减小约15～25 dB，重构时间减少约40%。
关键词:感知矩阵优化/
稀疏贝叶斯理论/
稀疏信号模型/
结构性噪声
Abstract:To solve sparse signal processing problem with structural noise interference, a method of sensing matrix optimization design based on sparse Bayesian theory is proposed. Combining the sparse signal model with additive interference, the design of the sensing matrix is realized by minimizing the trace of the posterior covariance matrix and the energy constraint of sensing matrix. The effects of sensing matrix optimization on the reconstruction error and reconstruction time are simulated using difference sparse signal and reconstruction algorithms, and the effects of the sensing matrix optimization on the reconstruction effect are analyzed when there is a bias in the prior information. The simulation results show that the optimized sensing matrix can obtain the important information in the sparse signal, the mean square error of the signal reconstruction accuracy is reduced by about 15～25 dB, and the reconstruction time is reduced by about 40%.
Key words:Sensing matrix optimization/
Sparse Bayesian theory/
Sparse signal model/
Structural noise
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=e82bccd5-0b79-4e81-b89e-e4e6cbcd5e4e
