基于LSTM特征提取的有限新息率畸变信号重构

doi:10.12263/DZXB.20201109

电子学报 ›› 2022, Vol. 50 ›› Issue (1): 217-225.DOI: 10.12263/DZXB.20201109

基于LSTM特征提取的有限新息率畸变信号重构

刘狄¹, 钱慧¹, 王中风²

^1.福州大学物理与信息工程学院，福建福州 350108
^2.南京大学电子科学与工程学院，江苏南京 210023

收稿日期:2020-10-10 修回日期:2021-06-16 出版日期:2022-01-25 发布日期:2022-01-25
作者简介:刘狄男，1994年10月生，福建福州人.2017年和2021年分别在武汉大学和福州大学获得工学学士和工学硕士学位.主要从事智能信息处理方面的研究工作. E-mail:526920885@qq.com
钱慧（通信作者）女，1977年11月生，湖北荆州人.2004年和2012年分别在华中师范大学和福州大学获得工学硕士和工学博士.目前任福州大学副教授，主要从事低功耗信号采样理论以及模数混合系统设计方面的研究工作. E-mail:qianhui@fzu.edu.cn
王中风男，1963年10月生，安徽省安庆人.教授、博士生导师、IEEE会士.1988年和1990年在清华大学获得工学学士和工学硕士学位.2000年在美国明尼苏达大学获得工学博士学位.目前任南京大学特聘教授，主要从事信号处理系统的低功耗设计方面的研究工作. E-mail:zfwang@nju.edu.cn
基金资助:
数字福建物联网工程应用实验室建设项目(82917002);工业物联网感知识别技术创新平台项目(82317241)

Reconstruction of Distorted Signal with Finite Innovation Rate Based on LSTM Feature Extraction

LIU Di¹, QIAN Hui¹, WANG Zhong-feng²

^1.College of Physics and Information Engineering，Fuzhou University，Fuzhou，Fujian 350108，China
^2.School of Electronic Science and Engineering，Nanjing University，Nanjing，Jiangsu 210023，China

Received:2020-10-10 Revised:2021-06-16 Online:2022-01-25 Published:2022-01-25

摘要/Abstract

摘要：

有限新息率（Finite Rate of Innovation，FRI）采样利用已知的信号波形结构实现信号的亚奈奎斯特率采样，在宽带信息系统应用中具有广泛的前景.但是，在实际的信息系统中，信号波形结构常常因噪声、远距离传输等非理想因素而发生畸变，从而导致FRI重构失败.本文依据波形再生的原理，提出了一种基于长短时记忆（Long and Short-Term Memory，LSTM）自动编码器的FRI重构方法.该方法利用LSTM自动编码器取代FRI采样系统中的采样核函数，通过离线训练获取畸变信号的未知波形结构，从而将波形序列投影为狄拉克特征序列，实现了波形畸变信号的FRI采样及重构.结果表明，本文的方法可以借助经典的零化滤波器有效地重构由于多径效应而发生畸变的FRI波形信号.

关键词: 亚奈奎斯特率采样, 有限新息率, 波形再生, 长短时记忆自动编码器, 零化滤波器

Abstract:

The finite rate of innovation(FRI) samples signal at rate of sub-Nyquist rate by using the known waveform structure, which has a wide application prospect in wideband information systems. However, in the real-world information system, the signal waveform structure is often distorted by the non-ideal factors, such as noise and long-distance transmission, which leads to fail to reconstruct the FRI waveform. According to the principle of waveform regeneration, an FRI reconstruction method based on long and short-term memory(LSTM) is proposed in this paper. This method replaces the sampling kernel of FRI sampling system by an automatic LSTM encoder, and the distorted waveform with unknown structure is obtained by off-line training. Thus, the waveform sequence is projected to a Dirac signature sequence. The FRI sampling and reconstruction of waveform distortion signal are realized. The results show that the proposed method can effectively reconstruct the FRI signals, which distorted by the multipath effect, by exploiting the standard annihilating filter.

Key words: Sub-Nyquist sampling, finite rate of innovation(FRI), waveform reproduction, long and short-term memory(LSTM) encoder, annihilation filter(AF)

中图分类号:

TN79

刘狄, 钱慧, 王中风. 基于LSTM特征提取的有限新息率畸变信号重构[J]. 电子学报, 2022, 50(1): 217-225.

LIU Di, QIAN Hui, WANG Zhong-feng. Reconstruction of Distorted Signal with Finite Innovation Rate Based on LSTM Feature Extraction[J]. Acta Electronica Sinica, 2022, 50(1): 217-225.

图/表 8

参考文献 17

1	戴琼海, 付长军, 季向阳. 压缩感知研究[J]. 计算机学报, 2011, 34(3): 425-434.
	DAI Q H, FU C J, JI X Y. Research on compressed sensing[J]. Chinese Journal of Computers, 2011, 34(3): 425-434. (in Chinese)
2	钱慧, 杨超. 频谱互质重排亚奈奎斯特率采样方法[J]. 电子学报, 2017, 45(10): 2506-2510.
	QIAN H, YANG C. Sub-Nyquist sampling on spectrum co-prime permutation[J]. Acta Electronica Sinica, 2017, 45(10): 2506-2510. (in Chinese)
3	王亚军, 李明, 刘高峰. 复杂脉冲序列的有限新息率采样方法[J]. 电子与信息学报, 2013, 35(7): 1606-1611.
	WANG Y J, LI M, LIU G F. Sampling complex pulse streams with finite rate of innovation methods[J]. Journal of Electronics and Information Technology, 2013, 35(7): 1606-1611. (in Chinese)
4	VETTERLI M, MARZILIANO P, BLU T. Sampling signals with finite rate of innovation[J]. IEEE Transactions on Signal Processing, 2002, 50(6): 1417-1428.
5	DONOHO D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
6	BHANDARI A, BLU T. FRI sampling and time-varying pulses: Some theory and four short stories[C]//Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing. New Orleans, LA, USA: IEEE, 2017: 3804-3808.
7	MARAVIC I, VETTERLI M. Sampling and reconstruction of signals with finite rate of innovation in the presence of noise[J]. IEEE Transactions on Signal Processing, 2005, 53(8): 2788-2805.
8	PAN H, SCHEIBLER R, BEZZAM E, et al. FRIDA: FRI-based DOA estimation for arbitrary array layouts[C]//Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing. New Orleans, LA, USA: IEEE, 2017: 3186-3190.
9	BLU T, DRAGOTTI P L, VETTERLI M, et al. Sparse sampling of signal innovation[J]. IEEE Signal Processing Magazine, 2012, 25(2): 31-40.
10	MULLETI S, SEELAMANTULA C S. Paley-wiener characterization of kernels for finite-rate-of-innovation sampling[J]. IEEE Transactions on Signal Processing, 2017, 65(22): 5860-5872.
11	DOGAN Z, GILLIAM C, BLU T, et al. Reconstruction of finite rate of innovation signals with model-fitting approach[J]. IEEE Transactions on Signal Processing, 2015, 63(22): 6024-6036.
12	ZHANG Y, DRAGOTTI P L. On the reconstruction of wavelet-sparse signals from partial fourier information[J]. IEEE Signal Processing Letters, 2015, 22(9): 1234-1238.
13	BAECHLER G, SCHOLEFIELD A, BABOULAZ L, et al. Sampling and exact reconstruction of pulses with variable width[J]. IEEE Transactions on Signal Processing, 2017, 65(10): 2629-2644.
14	RUDRESH S, NAGESH S, SEELAMANTULA C S. Asymmetric pulse modeling for FRI sampling[J]. IEEE Transactions on Signal Processing, 2018, 66(8): 2027-2040.
15	LEUNG V C H, HUANG J J, DRAGOTTI P L. Reconstruction of FRI signals using deep neural network approaches[C]//Proceedings of the 2020 IEEE International Conference on Acoustics, Speech and Signal Processing. Barcelona, Spain: IEEE, 2020: 5430-5434.
16	TARAR M O, KHALID Z. Reconstruction of finite rate of innovation spherical signals in the presence of noise using deep learning architecture[C]//Proceedings of the 28th European Signal Processing Conference. Amsterdam, Netherlands: IEEE, 2021: 1487-1491.
17	REDDY P S, PREMKUMAR A, SAIKIRAN B, et al. Finite rate of innovation signal reconstruction using residual neural networks[C]//Proceedings of the 2020 IEEE 4th Conference on Information & Communication Technology. Chennai, India: IEEE, 2021: 1-6.

参数类型	数值
脉冲结构	高斯脉冲
脉冲宽度	0.01
幅度	0~1范围内随机
脉冲	在0~1范围内以0.01间隔为基础随机分析
周期	1 s
脉冲畸变模式	路径数
新息率	10
样本采样率	1/21
样本长度	21
第二条路径反射系数	0.3
第二条路经延迟时间	0.05
第三条路径反射系数	0.2
第三条路径延迟时间	0.1
第四条路径反射系数	0.1
第四条路径延迟时间	0.15

参数类型	数值
脉冲结构	高斯脉冲
脉冲宽度	0.01
幅度	0~1范围内随机
脉冲	在0~1范围内以0.01间隔为基础随机分析
周期	1 s
脉冲畸变模式	路径数
新息率	10
样本采样率	1/21
样本长度	21
第二条路径反射系数	0.3
第二条路经延迟时间	0.05
第三条路径反射系数	0.2
第三条路径延迟时间	0.1
第四条路径反射系数	0.1
第四条路径延迟时间	0.15

基于LSTM特征提取的有限新息率畸变信号重构

Reconstruction of Distorted Signal with Finite Innovation Rate Based on LSTM Feature Extraction

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 17

相关文章 1

编辑推荐

Metrics