People

师资队伍

研究序列

曾媛--研究助理教授

个人主页

I am a self-motivated and hardworking individual, with solid experience in Digital Signal Processing and Artificial Intelligence. I am interested in doing research on Multi-channel Signal Processing, Noise Reduction, Speech Enhancement, Machine Learning and Image Processing.

教育经历

2010-2015 Ph.D. in Signal and Information Processing, Department of Intelligent Systems, Delft University of Technology, the Netherlands

  • Research Topic: Distributed Speech Enhancement in Wireless Acoustic Sensor Networks, 7 scientific articles and 1 patent have been published, Supervisors: Richard Hendriks and Inald Lagendijk.

2007-2010 M.Sc. in Signal and Information Processing , Department of Signal and Information Processing, Northwestern Polytechnical University,  China

  • Graduate Topic: Wireless Sensor Networks of Airplane Condition Monitoring , 3 scientific articles have been published, Supervisor: Jiangcang Ma.

2002-2006 China Youth Volunteer, Guangxi, China

  • Received award of “China Youth Volunteer Service Bronze Medal”.

2002-2006 B.Sc. in Biomedical Engineering, Department of Electronic and Information, Nanchang Hangkong University, China

工作经历

Postdoctoral Researcher; Delft University of Technology   2015-Present

Project: Machine Learning Algorithms for Image Reconstruction, funded by the Netherlands Organization for Scientific Research NOW and Van Gogh Museum (No. 323.54.004).       

  • Investigated benefits and drawbacks of existing machine learning algorithms for image segmentation and image reconstruction.
  • Analyzed valuable information in images and Designed image processing systems to improve image reconstruction performance via machine learning algorithms, i.e., deep learning and convolutional neural networks.
  • Conducted performance analysis on reconstruction accuracy and computational speed.

 

Ph.D. Candidate; Delft University of Technology       2010-2015

Project: Distributed Optimization Algorithms for Speech Enhancement in Wireless Acoustic Sensor Networks, funded by China Scholarship Council (No. 2010629038).   

  • Developed novel distributed noise reduction algorithms to improve speech quality and speech intelligibility, and evaluated the performance of the algorithms by both mathematical analysis and experimental analysis.
  • Developed new distributed algorithms to solve average consensus problems in wireless sensor networks and provided analysis to show the reduction of communication cost and the improvement of robustness.
  • Developed distributed algorithm to estimate inverse of the correlation matrix for beamforming, and improved noised reduction performance.
  • Analyzed the effect of clock synchronization on speech enhancement algorithms, and evaluated the use of three different clock synchronization algorithms in the context of multi-microphone noise reduction.

 

Research Assistant; Northwestern Polytechnical University        2009-2010

Project: Wireless Sensor Networks of Airplane Condition Monitoring, funded by Graduate Starting Seed Fund of University (No. Z200938).  

  • Developed and designed a management system of wireless sensor networks to process the observed information and monitor wireless sensors.
  • Organized and led meetings of team members for system debug and test.

 

Graduate Student; Northwestern Polytechnical University   2008-2009

Project: Blind Processing Research of Aero-engine Vibration Signal, funded by National Nature Science Foundation (No. 60672184).    

  • Developed feature extraction model of the aero-engine vibration signals, and the accuracy of aero-engine fault diagnosis is improved via the proposed model and learning algorithms.

 

Teaching Assistant     2010

Northwestern Polytechnical University, China

  • Prepared and graded homework solutions for the bachelor course “Digital Signal Processing”.
  • Assisted bachelor students in solving project problems and writing theses.

代表著作与论文

  1. Zeng. “Distributed Speech Enhancement in Wireless Acoustic Sensor Networks”, ISBN: 978-94-6186-423-9, 2015.

Patents:

  1. Heusdens, G. Zhang, R. C. Hendriks, Y. Zeng and W. Kleijn. “Distributed Beamforming Based On Message Passing”. U.S. Patent, No. US 9584909 B2, 28 Feb, 2017.

Journals:

  1. Zeng, J. C.A. van der Lubbe, M. Loog and E. Postma. “Convolutional Neural Networks for Pixel-wise Image Reconstruction”, IEEE Trans. on image processing, to be submitted.
  2. Zengand R. C. Hendriks. “Distributed Estimation of the Inverse of the Correlation Matrix for Privacy Preserving Beamforming”, Elsevier signal processing, 107: 109-122, Feb. 2015.
  3. Zeng and R. C. Hendriks. “Distributed Delay and Sum Beamformer for Speech Enhancement via Randomized Gossip”, IEEE/ACM Trans. Audio, Speech and Language processing, 22(1):260-273, Jan. 2014.
  4. Ma and Y. Zeng. “Fault diagnosis method for Aero-engine based on Independent Component Analysis and Discrete Hidden Markov Model”, International Journal of Plant Engineering and Management, 2010.
  5. Ma, Y. Zengand G. Zhang. “Application of Feature Basis Function ICA in the Aero-engine Fault Diagnosis” ,Computer Measurement and Control, 2009.
  6. Ma, G. Zhang and Y. Zeng. “Application of the Combination of Blind Source Separation with Hilbert-Huang Transform in Aero-engine vibration signal analysis” have published in Mechanical Science and Technology for Aerospace Engineering, 2009.

联系方式

邮箱:zengy3@sustc.edu.cn

电子系微信公众号

首页

地址:广东省深圳市南山区学苑大道1088号
          第二科研楼五层

邮箱:eee@sustc.edu.cn

© 2016 All Rights Reserved. 粤ICP备14051456号