利用改进的稀疏表示处理FBG传感信号

发布时间：2018-02-04 14:23

本文关键词： 压缩感知(CS) 稀疏表示正交匹配追踪(OMP) 信号去噪 FBG传感信号　出处：《光电子.激光》2014年10期 　论文类型：期刊论文

【摘要】：基于压缩感知(CS)的正交匹配追踪(OMP)算法,须以稀疏度确定为先验条件,在实际应用中稀疏度不易确定的情况下,本文提出了稀疏度确定方法和二次正交匹配追踪(TOMP)算法。先引入熵权法采用多指标融合并结合饱和值点法确定稀疏度,然后利用所提方法对实验信号进行重构。实验仿真结果表明:与同类算法相比,本文所提TOMP算法增加0.1s运行时间降低了12~22%的重构误差,更好折中处理了重构误差和时间;与不同类算法相比,本文所提方法重构的信号信噪比(SNR)最大可提升22dB,且均方根误差(RMSE)降低0.7,因此去噪效果更优。
[Abstract]:The orthogonal matching tracking (OMP) algorithm based on compressed sensing (CS) must take the sparse degree as a priori condition, and it is difficult to determine the sparse degree in practical applications. In this paper, the sparse degree determination method and Quadratic orthogonal matching tracking algorithm are proposed. Firstly, the entropy weight method is introduced to determine the sparsity with the combination of multi-index fusion and saturation point method. The simulation results show that compared with the similar algorithms, the proposed TOMP algorithm increases the running time by 0.1 s and reduces the reconstruction error by 12 ~ 22%. Better compromise deals with reconstruction error and time; Compared with different kinds of algorithms, the SNR (signal to noise ratio) can be increased by 22 dB and RMSE is reduced by 0.7, so the denoising effect is better.
【作者单位】：重庆邮电大学工业物联网与网络化控制教育部重点实验室;重庆邮电大学光纤通信技术重点实验室;
【基金】：国家自然科学基金(61275077) 重庆市基础与前言研究计划(cstc2013jcyjA40052)资助项目
【分类号】：TN911.4
【正文快照】： 1引言光纤传感系统中,解调技术是决定其测量结果的关键。光纤传感信号在采集和传输过程中会引入大量噪声,严重影响了信号解调精度。因此,对含噪传感信号的处理是十分必要的。小波具有多尺度、多分辨率的特性,在信号处理中去噪效果好,而受到广泛关注。具有代表性的有小波阈值去

【参考文献】