对于一类非线性信号的去噪问题,该文提出一种基于奇异值分解(Singular Value Decomposition, SVD)的有效迭代方法。对现有奇异值差分谱方法在两类不同非线性信号上的去噪效果进行了对比,指出在信号不具有明显特征频率、非周期性变化时这一方法并不适用,并分析了现象产生的原因;然后针对该类信号的特点重新定义了Hankel矩阵结构,给出有效奇异值的确定方式,并通过SVD多次迭代过程实现对该类信号的有效去噪。对实际飞行数据去噪的实验结果表明,该方法对提出的一类信号对象不仅去噪效果良好,而且可提高运算效率。
To solve a class of nonlinear signal denoising, an effective iteration method based on the Singular Value Decomposition (SVD) is proposed. When the signals have no obvious characteristic frequency and non-periodic change, the current difference spectrum method is not applicable by comparing the results on the two class of nonlinear signal, and then the corresponding reason is analyzed. According to the signal feature, the structure of the Hankel matrix is defined again and the valid singular values are determined. The effective denoising is realized by the repeated iteration which is based on the SVD. The results of the flight data demonstrate that the proposed method can effectively reduce the noise and improve the computing efficiency as well.
查翔,倪世宏, 张鹏. 一类非线性信号去噪的奇异值分解有效迭代方法[J]. 电子与信息学报, 2015, 37(6): 1330-1335.
Zha Xiang, Ni Shi-hong, Zhang Peng. Effective Iteration Method of a Class of Nonlinear Signal Denoising Based on Singular Value Decomposition. JEIT, 2015, 37(6): 1330-1335.
Massari C, Brocca L, Ciabatta L, et al.. A Wiener-wavelet based filter for de-noising satellite soil moisture retrievals[C]. Proceedings of the EGU General Assembly, Vienna, Austria, 2014: 1123-1148.
[2]
Donoho D L. De-noising by soft-thresholding[J]. IEEE Transactions on Information Theory, 1995, 41(2): 613-618.
[3]
He Wang-peng, Zi Yan-yang, Chen Bin-qiang, et al.. Tunable Q-factor wavelet transform denoising with neighboring coefficients and its application to rotating machinery fault diagnosis[J]. SCIENCE CHINA Technological Sciences, 2013, 56(8): 1956-1965.
[4]
Maryam A and Rodrigo Q Q. Automatic denoising of single-trial evoked potentials[J]. NeuroImage, 2013, 66(10): 672-680.
[5]
Pascal P M, Christian B, and Florence B. Denoising NMR time-domain signal by singular-value decomposition accelerated by graphics processing unites[J]. Solid State Nuclear Magnetic Resonance, 2014, 61(5): 28-34.
[6]
Li Zhen-xing and Dai Wei-xiao. Local mean decomposition combined with SVD and application in telemetry vibration signal processing[J]. Applied Mechanics and Materials, 2013, 347(2): 854-858.
[7]
Ajit R, Anand R, and Arunava B. Image denoising using the higher order singular value decomposition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(4): 849-862.
[8]
Cheng D H, Xu Q J, and Yao W F. Regional wind energy resource forecasting based on SVD and support vector machine[J]. Advanced Materials Research, 2014, 247(12): 1070-1072.
Zhao Xue-zhi, Ye Bang-yan, and Chen Tong-jian. Difference spectrum theory of singular value and its application to the fault diagnosis of headstock of lathe[J]. Journal of Mechanical Engineering, 2010, 46(1): 100-108.
Lü Yong-le and Lang Rong-ling. Noise reduction method for flight data based on singular value decomposition[J]. Computer Engineering, 2010, 36(3): 260-262.
[11]
Zhao H M, Shen H, Fu Y, et al.. Using singular value decomposition and high order spectrum for bearings fault diagnosis[C]. Proceedings of the IEEE Transportation Electrification Conference and Expo Asia-Pacific(ITEC Asia-Pacific), Beijing, China, 2014: 1-4.
[12]
Zhao Xue-zhi and Ye Bang-yan. Selection of effective singular values using difference spectrum and its application to fault diagnosis of headstock[J]. Mechanical Systems and Signal Processing, 2011, 25(5): 1617-1631.
Lei Da and Zhong Shi-sheng. Aircraft engine health signal denoising based on singular value decomposition and empirical mode decomposition methods[J]. Journal of Jilin University, 2013, 43(3): 764-770.
Zhao Xue-zhi, Ye Bang-yan, and Chen Tong-jian. Extraction method of faint fault feature based on wavelet-SVD difference spectrum[J]. Journal of Mechanical Engineering, 2012, 48(7): 37-48.
[15]
Brand M. Incremental singular value decomposition of uncertain data with missing values[C]. Proceeding of the 2002 European Conference on Computer Vision, Copenhagen, Denmark, 2002: 1-12.