Dissertation > Excellent graduate degree dissertation topics show

Blind Source Separation Algorithm Based on ICA

Author: LiuHaiPeng
Tutor: TianXueMin
School: China University of Petroleum
Course: Control Science and Engineering
Keywords: Blind Source Separation Independent Component Analysis Kurtosis Newton's method Seismic signals Denoising
CLC: TN911.7
Type: Master's thesis
Year: 2011
Downloads: 71
Quote: 0
Read: Download Dissertation

Abstract


Blind source separation is a no or very little about the premise of source signals and mixed prior knowledge of restoring source signals from a set of mixed-signal technology, which in speech processing, image processing , underwater acoustic signal processing , are widely used in the field of radar signal processing , seismic signal processing , biomedical , fault diagnosis and even financial data analysis . This paper studies the blind source separation based on independent component analysis algorithm and its application in seismic signal denoising . The main contents are as follows: summarizes the basic theory and algorithm of blind source separation problem of the blind source separation problem model , blind source separation algorithm prior knowledge , followed by a detailed discussion of several common based independent component analysis for blind source separation algorithm, and finally simulation experiments to compare these separation performance of several algorithms . Miscellaneous series of mixed-signal blind source separation problem , blind source separation algorithm based on Newton's law , the introduction of independent meta-analysis of the algorithm is given an improved switch criteria , using the kurtosis of the random variable can not be realized for some algorithms to distinguish the type of signal , different signals to different non-linear function , containing both super-Gaussian signal and sub ??- Gaussian signal miscellaneous mixed-signal blind source separation . Also verified through simulation of the algorithm for blind source separation Hybrid mixed-signal , and improved convergence speed and separation performance than the original Newton method and extended information maximization method . Blind source separation algorithm seismic signal denoising is divided into multi-channel and single-channel seismic signal denoising two parts . For part of the multi-channel noise as part of the source signal , and then use the blind source separation algorithm seismic signal denoising ; single channel part , by the singular value decomposition of the single-channel seismic signal preprocessing , generating another road signal , then blind source separation algorithm for processing the two signals , in order to achieve a single-channel seismic signal denoising , simulation results show that the method is feasible .

Related Dissertations

  1. Research on Electromagnetic Acoustic Transducer Detection System Based on FPGA,TH878.2
  2. Research on Methods of Medical Ultrasound Image Denoising,TP391.41
  3. Design of Testing Equipment for Electronic Products,TN06
  4. Thickness of plate electromagnetic ultrasonic detection and data processing technology research,TG115.28
  5. The Research of Denoising and Edge Detection Method in Image Measurement,TP391.41
  6. Research on Video Denoising Algorithm Based on Temporal-spatial Domain,TP391.41
  7. Study on the Recognition and Classification System of Cervical Exfoliated Cells Based on Neural Network,TP391.41
  8. Research on the Denoising and the Restoration of Frame Interpolation of X-band Nautical Radar Sea-surface Image,TP391.41
  9. Algorithm of Poisson-Noise Removing Based on ICA and Its Application on CT Imaging,TP391.41
  10. Image matching method and application,TP391.41
  11. Research on Moving Object Detection Based on Independent Component Analysis,TP391.41
  12. Image Denoising and Enhancement based on Curvelet Transform,TP391.41
  13. Research on Process Monitoring Based on ICA and Wavelet Transform Methods,TP274
  14. Heat Fault Diagnosis of Transmission Line Joints Based on Infrared Detection Technology,TP274.52
  15. Research and Implement on Diagnoses of Common Bile Duct Stones Based on Image Processing Technology,TP391.41
  16. Practical Online Brain-Computer Interface System Based on Motion-onset Visual Responses,R318.0
  17. The Enhancement of Fault Detection and Diagnosis in Rolling Element Bearings Using Spectral Kurtosis and EMD,TH165.3
  18. Based on the time-frequency analysis of the mechanical failure of blind source separation method,TH165.3
  19. Mechanical fault diagnosis method based on kernel function method,TH165.3
  20. The the HHT analysis and application of mechanical vibration signal,TH113.1
  21. Research on Anti-jamming Technology of MMW-Radar,TN974

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Communication theory > Signal processing
© 2012 www.DissertationTopic.Net  Mobile