Determination algorithm of optimal decomposition level based on singular spectrum analysis

Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (36): 137-141.

Previous Articles Next Articles

Determination algorithm of optimal decomposition level based on singular spectrum analysis

WU Long1, XING Likun2, CHEN Shuai1

1.College of Electrical and Information Engineering, Huainan Normal University, Huainan, Anhui 232001, China
2.College of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan, Anhui 232001, China

Online:2012-12-21 Published:2012-12-21

基于奇异谱分析的最优分解层数确定算法

伍龙1，邢丽坤2，陈帅1

1.淮南师范学院电气信息工程学院，安徽淮南 232001
2.安徽理工大学电气与信息工程学院，安徽淮南 232001

Abstract

Abstract: This paper, applying the singular spectral analysis theory to the wavelet threshold denoising algorithm, realizes an adaptive determination algorithm??for optimal level of decomposition. Comparing distribution of the singular spectral with noise signals under different SNR environments, it determines the optimal decomposition level based on the singular spectrum characteristics of wavelet coefficients. Being tested, the algorithm can adaptively determine the optimal decomposition level, according to noise interference of noise signals, effectively increasing voice enhancement effect and avoiding unnecessary waste of hardware resources.

Key words: optimal decomposition level, speech enhancement, singular spectral analysis, threshold denoising

摘要： 将奇异谱分析理论引入小波阈值算法中实现了一种基于奇异谱分析的自适应最优分解层数确定算法。通过对比不同信噪比下带噪信号的奇异谱分布情况，根据小波系数的奇异谱特性来确定最优分解层数。经测试，该算法可以根据带噪信号受噪声干扰情况自适应地确定最优分解层数，有效提高了语音增强效果并且避免了不必要的硬件资源浪费。

关键词: 最优分解层数, 语音增强, 奇异谱分析, 阈值降噪

WU Long1, XING Likun2, CHEN Shuai1. Determination algorithm of optimal decomposition level based on singular spectrum analysis[J]. Computer Engineering and Applications, 2012, 48(36): 137-141.

伍龙1，邢丽坤2，陈帅1. 基于奇异谱分析的最优分解层数确定算法[J]. 计算机工程与应用, 2012, 48(36): 137-141.

[1]	WANG Shiqi, ZENG Qingning, LONG Chao, XIONG Songling, QI Xiaoxiao. Multi-task Learning for Speech Enhancement and Detection [J]. Computer Engineering and Applications, 2021, 57(20): 197-202.
[2]	WANG Yan, JIA Hairong, JI Huifang, WANG Weimei. Feature Joint Optimization of Deep Belief Network for Speech Enhancement [J]. Computer Engineering and Applications, 2019, 55(9): 38-42.
[3]	JI Huifang, JIA Hairong, WANG Yan. Speech Enhancement Method for Improving Phase Spectrum Compensation [J]. Computer Engineering and Applications, 2019, 55(8): 48-52.
[4]	WANG Jie1, YANG Chengcheng1, MO Jiayong2, WANG Dunze1, WANG Xiexie1. A priori SNR estimator based on harmonic regeneration [J]. Computer Engineering and Applications, 2018, 54(7): 44-48.
[5]	WANG Hu, LI Jing, ZHAO Hengmiao, ZANG Yan, LI Chuntang. Speech enhancement algorithm of sparse low rank model and phase spectral compensation [J]. Computer Engineering and Applications, 2018, 54(5): 150-155.
[6]	WANG Bo，YU Fengqin，CHEN Ying. Speech enhancement based on nonsmooth nonnegative matrix factorization [J]. Computer Engineering and Applications, 2017, 53(7): 160-164.
[7]	WANG Xia1, WANG Dan1, WANG Guangyan2, ZHANG Yan1. Noisy face mask speech enhancement combining compressed sensing with EMD [J]. Computer Engineering and Applications, 2017, 53(18): 137-140.
[8]	YANG Zhengling1，2, ZHANG Xi1, ZHANG Jun1，2, YANG Zhao1, LIU Yadi1. Reduce sensitivity of wavelet denoising to outliers by arc-tangent function transformation [J]. Computer Engineering and Applications, 2017, 53(10): 241-245.
[9]	ZHANG Jianwei, TAO Liang, ZHOU Jian, WANG Huabin. Improved minima controlled recursive averaging algorithm based on improved spectrum smoothing strategy and speech enhancement [J]. Computer Engineering and Applications, 2017, 53(1): 153-157.
[10]	KONG Guojie, WANG Kai, SHEN Mingtang, ZHANG Fang, WU Ke. Adaptive total variation method with wavelet threshold filtering [J]. Computer Engineering and Applications, 2016, 52(8): 240-242.
[11]	HU Yonggang1, ZHANG Xiongwei1, ZOU Xia1, MIN Gang1，2, ZHANG Liwei1, WANG Jian3. Speech enhancement algorithm using ADMM sparse nonnegative matrix factorization [J]. Computer Engineering and Applications, 2016, 52(3): 108-112.
[12]	NING Kuangfeng, WANG Jingfang. DCT domain Wiener filter speech enhancement [J]. Computer Engineering and Applications, 2015, 51(8): 226-230.
[13]	CAO Longtao1, LI Ruwei1, BAO Changchun1, WU Shuicai2. Speech enhancement algorithm based on noise estimation of binary masking [J]. Computer Engineering and Applications, 2015, 51(17): 222-227.
[14]	GU Peng1, 2, ZHU Junhua1，2, DING Fei1，2. Method in noisy signal compressed sensing for speech de-noising [J]. Computer Engineering and Applications, 2015, 51(15): 216-220.
[15]	WANG Yulin1, TIAN Xuelong1，2, GAO Xueli1. DSP realization of modified speech enhancement algorithm based on adaptive filters [J]. Computer Engineering and Applications, 2015, 51(1): 208-212.

Determination algorithm of optimal decomposition level based on singular spectrum analysis

基于奇异谱分析的最优分解层数确定算法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics