Papers by Keyword: Singular Value Decomposition (SVD)

Abstract: A detailed description of the basic principles of MOESP subspace system identification methods is introduced. The extended state space equation which is formed by Hankel matrix is used in subspace identification, and then the augmented observation matrix and state space vector estimation are all got by SVD decomposition. The simulation results on a multi input/output system demonstrate the effectiveness and feasibility of the method.

Abstract: As a good time and frequency domain localization features, the method of wavelet ridge became one of effective ways to identify the pulse modulation. Aiming at the problem of the low intra pulse recognition rate of the wavelet ridge method in the situation of low signal-to-noise ratio, this paper presented a method of fast wavelet ridge intra pulse modulation based on singular value decomposition. The basic idea of this method:The first signal for fast Morlet wavelet transform; determined the singular value decomcomposition filtering threshold; According to the iterative method to extract the wavelet ridge was to calculate the instantaneous signal teristics; then to identify signal according to the different characteristics of signal instantaneous frequency. The simulation results show that: In less than or equal 0dB SNR situation, this method can improve the recognition rate of common pulse modulation.

Abstract: To weaken the noise disturbance of GRM and improve the matching precision and matching probability of inertial/geomagnetic system, this paper proposed a method for denoising based on SVD. Firstly, from the perspective of information entropy, the singular entropy is introduced and the inner link between singular entropy and signal-to-noise ratio (SNR) is analyzed. Secondly, the method based on the asymptotic characteristic of the probabilities associated with the different singular values order (SVO) is proposed. Lastly, by utilizing practical GRM, the denoising analysis about the proposed method is demonstrated and later simulation experiments of GMN are accomplished. Simulation results show that the method is feasible and reliable.

Abstract: The paper mainly discusses the digital watermark algorithm for 3D grid model. Traditional algorithm cannot coordinate the contradiction between invisibility and robustness of digital watermark. In order to improve the security, this paper proposes a 3D grid digital watermark algorithm based on singular value decomposition (SVD). Firstly, we set the matrix formed by distance from the vertexes to center of 3D mesh model as the vertex modulus matrix. Then the matrix is decomposed by SVD, and finally the watermark is embedded into the diagonal matrix to get the 3D mesh model containing watermark signal. The simulation results show that the algorithm effectively solves the problem of invisibility and robustness of watermark, and the watermark formed here can resist various attacks which enforce the protection of 3D mesh models Rights.

Abstract: A blind watermarking algorithm for 2D CAD drawings based on singular value decomposition (SVD) is presented. In this paper, the handles of lines are sorted to rule out additional entities and relative vertex coordinates are divided into small matrixes for larger watermark capacity. Then, quantization modulation is used to embed watermark so it can be extracted without original drawings. The experiments show that the proposed algorithm has good invisibility, larger watermark capacity and is also robustness to various geometric and entity addition attacks.

Abstract: The description of syndromes and symptoms in traditional Chinese medicine (TCM) is extremely complicated. And how to diagnose the patient's syndrome in a better way is the primary objective of clinical health care workers all the time. It was a good attempt to diagnose patient's syndrome by combining Latent Semantic Analysis and the feature of TCM knowledge----both syndromes and organs have the same clinical manifestation collection that are symptoms. In this paper, correlative degrees would be computed and sorted in a certain latent semantic space which was constructed by syndromes and organs . According to the result of correlative degrees computing, the classifying could be done by choosing the highest correlative degree as the belonging class. The experimental results show that this method performs quite well.

Abstract: In view of the modal parameter identification when only has the output signals of the system under ambient excitations has difficulty, a new method which can identify the structural modal parameters exactly based on singular value decomposition of the power spectrum is put forward. This method is used in the modal parameter identification of a cable stayed bridge under ambient excitations, and the identification frequency is compared with the finite element computation frequency. The results indicated that this method has overcome the subjectivity in modal selection of frequency domain pick-peaking method, choose eigenfrequency and identify close modal accurately and objectively. With the advantages of practical, processing simply and fast.

Abstract: HHT is widely used to analyze nonlinear and non-stationary signals. But how to extend boundaries of signals in decomposition processes is a key problem of HHT. A new technique based on response surface method (RSM), which establishes the recursive relations between sample points of signals, is presented to deal with this difficult problem. Besides, the boundary extension problem arising from HHT can be described by mathematical least squares problem but traditional gradient algorithms may diverge when the Hessian matrix of the object function of the least squares problem is non-positive. It has been proved that the generalized inverse of the linear equations (derived from the linear least squares problem) by singular value decomposition is the solution of original linear least squares problems. Thereby the divergence problem is also solved. Analysis results with respect to simulation signals and measured signals show that the method with new boundary extension technique performs successfully for HHT.

Abstract: In order to obtain the MEMS micro structure's high resolution measurement dynamic parameters ,this paper ,by obtaining stroboscopic imaging technology of micro structure of moving image sequence, proposed one kind based on the phase correlation technique and singular value decomposition technique combining sub-pixel measurement method. By the method of phase correlation, image space coordinate can transform into the frequency domain to the coordinates of the parameter space. And then through the singular value decomposition technique to obtain the correlation matrix, using the least-squares fitting, get the sub-pixel level displacement measurement results. The experimental results show that, with this method of measuring MEMS micro structure plane motion amplitude can reach sub-pixel accuracy, and can be effectively reduced by uneven illumination effect on the measuring result, thereby increasing the measurement results of stability and reduces the measuring error.

Abstract: We present an efficient stratified optimization approach for self-calibration of a camera in the case that its focal length and the principal point location are unknown. Generally we can assume that the two views are of the same focal length, and the pixels are nearly perfectly rectangular, also it is possible to know the aspect ratio rather accurately. In our approach, we use singular value decomposition to solve a modified Kruppa Equation to derive the focal length with the supposition that the principal point is at the center of the image, and perform an exhaustive search for the principal point near the center of the image to minimize a cost function. We can get a much accurate result with the optimized principal point location.