Dissertation > Excellent graduate degree dissertation topics show

Blind signal separation application technologies

Author: GaoLiLi
Tutor: HeSongBai
School: University of Electronic Science and Technology
Course: Circuits and Systems
Keywords: ICA BSE BSS PCA
CLC: TN911.7
Type: Master's thesis
Year: 2011
Downloads: 74
Quote: 0
Read: Download Dissertation

Abstract


The technology of Separate Blind Signal is a theory and method that is development relatively late in the field of signal processing,but it has rapidly become an important part of this field and it’s development gradually maturing and sysematic.In this paper,it expounds the research status,application area and basic research methods of blind signal,while summarizes the classical and isolated techniques.Through analysising the independent component of separation techniques in the blind signal,it compares with the traditional algorithm in the fields of convergence speed and separation precision and so on.It builds the corresponding mathematical model from two aspects,blind source separation(BSS) and blind source extraction(BSE) respectively by the methods of principal component analysis(PCA),independent component analysis(ICA) and the natural gradient method.Applying theories of statistical signal and matrix,combining with the gradient algorithm,making Newton’s iterative method,etc,it respectively put forward two improved algorithms based on the models of BSS and BSE.The fundamental purpose of blind source separation is making the final output relatively independent.There are two basic methods,one is building the input matrix A,getting a lower degree of the X matrix,which make the output Y independently.the second is to build a mixing matrix W,through the rotation matrix of W,calculating their relavancy,iteratively until making the output Y independently.The basic idea of improving the algorithm is as follows:(1)the model BSS:Building a mathematical model→finding out the rotation matrix of the mixing matrix→calculating the positive corelation of the results of the matrix calculation→making the output independently by designing steps and calculating repeatly;(2)the model BSE: Analysising the building mathematical model→extracting the initial signal to make the output relatively independent→getting the output matrix by calculating repeatedly→making the output independently by the mixing matrix.The improved algorithm uses in composite fault diagnosis of induction motor. Through the stability of convergence and adaptability comparative study, turn out that optimization algorithm in the above has obvious advantages.Results Analysis from two aspects: one is Pure theory; secont is the test in the application of the induction motor and the composite fault diagnosis. Theoretical Analysis shows that The improved algorithm has better precision and better astringency etc.Practical tests prove that got better stability.Finally, based on the broad application fields of Separation technology and good application prospect,In addition, some seinsfrages,the technology of several research was prospected.

Related Dissertations

  1. The Research of the Fault Diagnoses Algorithm for the Liquid Rocket Engine Testing Bed Based on PCA-SVM,V433.9
  2. Research on Recognition of Fabric Weave Pattern Based on Space-Frequency Domain,TS101.923
  3. Application of Functional MR Imaging in Diagnose for Prostate Cancer,R737.25
  4. Blind Source Separation Algorithm Research with Communication Signals,TN911.7
  5. Multivariate Process Monitoring and Fault Diagnosis Based on Distribution Characteristics of the Data,TP274
  6. Design and Implementation of Telecom’s PSTN Service Pretreatment System,TP311.52
  7. Neural Mechanisms of Motor Function with Structural Equation Modeling in Stroke Study,R318.0
  8. Temperature-responsive Expression of Two Phz Gene Clusters is Differently Regulated by Sensor RetS in Pseudomonas Aeruginosa Strain M18,Q75
  9. Research on ECG Feature Extraction and Classification Method,TN911.7
  10. Research on Feature Extraction Methods in Color Face Images,TP391.41
  11. Implementation and Application Design of Nanjing Telecom BSS Obstacle Automatic Processing System,TP311.52
  12. Role of Heat Shock Protein 27 in Regulating Mouse Embryonic Stem Cells Differentiation Induced by Icariin in Vitro,R285.5
  13. ICA of Complex FMRI Signal Based on Kurtosis Maximization Algorithm,TN911.6
  14. Clinical Effection on Abnormal Expression of the Angiopoietin-1&-2 and Receptor Tie-2 in Human Intracranial Aneurysms,R651.1
  15. Remote Sensing Image Fusion Technology,P237
  16. Application of Independent Component Analysis in fMRI Data Processing,TP274
  17. Blind Source Separation of Vibration Signals of AC Motor Speed Control System,TM921.51
  18. GAD-Ab, ICA, IAA in LADA Expression and Clinical Significance,R587.1
  19. The Dimension Reduction for High-dimensional Multispectral Space,TP391.41
  20. The Control Method Research for Prosthetic Hand Based on EEG,TP391.4
  21. Research on Analysis & Recognition Method for Multi-motor ERD/ERS Signal in BCI System,TN911.6

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