113 results about "Singular matrix" patented technology

A method and system for beamforming communication in a wireless communication system that includes a wireless initiator and a wireless responder is provided. A channel matrix is estimated at the responder. The singular value decomposition of the channel matrix yields the right singular matrix, which is further deconstructed into certain components. The right singular matrix components are quantized in a vector fashion and fed back to the initiator for reconstruction and beamforming communication.

The present invention is to provide a wireless communication system and a receiving device each capable of restraining a circuit scale and obtaining an excellent error rate performance. In a wireless communication system in which a transmitting device and a receiving device communicate in an orthogonal frequency division multiplexing system, the transmitting device includes a partition unit partitioning a transmission data sequence into a plurality of data streams and a mapping unit mapping the plurality of data streams to each of transmitting antennas by using, for a precoding matrix, columns, corresponding to a stream count of the plurality of data streams, of a right singular matrix acquired by singular value decomposition of a channel matrix, and the receiving device includes a decoding unit that Viterbi-decodes the received signals by weighting a path metric by using a weighting coefficient corresponding to a signal-to-noise ratio obtained from the present channel matrix.

The invention discloses a pre-coding method, a pre-coding system and a construction method of a pre-coding code book. A transmitting terminal and a receiving terminal are used for storing the information of the pre-coding code book together; a code word is selected from the pre-coding code book by the receiving terminal according to an estimated channel matrix; the sequence number of the code word is fed back to the transmitting terminal; and symbol blocks transmitted to the receiving terminal are pre-coded through the code word by the transmitting terminal according to the code word found by the sequence number, wherein column vectors in at least eight code word vectors or at least eight code word matrixes in the pre-coding code book are obtained on the basis of an eight-dimensional vector selected from an eight-dimensional vector set, and the eight-dimensional vector set is obtained by computing partial or all four-dimensional vectors of u1, u2, u3, u4, u5, u6, u7, u8, s, m and n. The pre-coding method is used for the pre-coding of eight-antenna MIMO systems when the channel matrix is a singular matrix, and can enhance the pre-coding property. The four-dimensional vectors used for computing the code word can be stored only for saving storage space.

The invention relates to a method for automatic identification and detection of defects in a composite material. The method comprises steps of: detecting the composite material to generate an infrared image by using infrared thermal wave nondestructive testing equipment; conducting phase space reconstruction on the infrared sequence image to determine defect position of the composite material and segment defect area of the image; conducting phase space reconstruction on the infrared sequence image with defect area and carrying out singular value decomposition to obtain a singular matrix, and left and right projection matrixes; carrying out matrix reconstruction again on the two projection matrixes; extracting algebraic characteristics of time information and space information of the defect through singular value decomposition; constructing mixing characteristic vector as characteristic symptom of the defect; and utilizing results from a nerve network classifier to complete the identification and classification determination. The method of the invention can realize automatic identification and detection on defect in the composite material, carry out rapid detection on damage type of the composite material and provide rapid detection means according to usage condition of the composite material, and has critical reality meaning and research value.

The invention discloses a power system transient stability simulating method based on implicit numerical integration. Compared with an existing power system transient stability numerical simulation implicit trapezoidal integration method, the power system transient stability simulating method employs a power-angle integration formula with a smaller local truncation error, namely, enables a non-linear differential equation set for describing a power system transient process to be expressed as a linear portion and a non-linear portion. An accurate analysis expression of a state transition matrix is obtained by reasonably selecting a system matrix of the linear portion as a singular matrix, and a group of implicit integration formulas is obtained by leading linear integrable functions to be approximate to the non-linear portion of the differential equation set. The local truncation error of the power-angle implicit integration formulas of the generator refers to O (h5) which is larger than a local truncation error O (h3) of implicit trapezoidal integration, the calculated quantity of integration each time is equivalent to that of the implicit trapezoidal integration. By means of the high-precision implicit integration formulas, iteration times of each integration step under the same iteration precision condition are decreased, so that the simulated calculated quantity is remarkably decreased.

The invention discloses a hybrid precoding method of a large-scale MIMO system applied to a millimeter wave band. The method comprises the following steps: S1: obtaining, by a base station, channel state information (CSI); S2: performing SVD on a downlink channel to obtain a right singular matrix V, using a phase angle of a former NRF column of the V as each element in an analog precoding matrix FRF, wherein NRF represents the number of radio frequency links; S3: using former Ns columns of the right singular matrix V subjected to the SVD on the downlink channel as an optimal precoding matrix Fopt, and obtaining a digital precoder (the formula is described in the specification) according to the Fopt and the FRF by using a least-squares solution (LS), wherein Ns represents the number of datastreams; and S4: performing first time precoding on the data streams through the FBB obtained in the step S3, then performing phase adjustment on a signal, namely, the FRF in the step S2, by using aphase shifter, and finally transmitting the signal through an antenna. Compared with the prior art, the hybrid precoding method disclosed by the invention has the advantages that no optimization method or complicated iterative algorithm is needed, so that the complexity is greatly reduced, and higher spectral efficiency can be achieved.

The invention discloses a channel estimation method for an orthogonal frequency division multiplexing (OFDM) system under an interference environment. A column singular matrix is multiplied by a corresponding orthogonal projection matrix to obtain a zero matrix so that the vector obtained in the step (5) is equivalent to T=QIp, wherein the element Ip (k) in the vector Ip is a result of dividing an interference signal and noise by a local pilot signal. Because the orthogonal projection matrix is a linear matrix, the statistical property of each element T (k) of the vector T is the same as thatof the element Ip (k), the variance estimation value obtained in the step (5) is the sum of the interference signal variance and the noise variance, and optimal estimation of channel time domain impulse response is obtained by using the step (7). The method has low calculation complexity, and the estimated performance of the method is approximate to the ideal channel estimated performance in the absence of interference signals; and meanwhile, aiming at block pilot patterns and uniformly-spaced comb pilot patterns, the method is insensitive to the power of the interference signal, only needs one OFDM symbol, and has good instantaneity.

The invention provides a calculation method for a vamp glue spray trajectory. The calculation method comprises the steps that firstly, a set of soles and vamps are scanned through a linear laser scanner, and two sets of point cloud data are obtained; secondly, arrays of original point cloud pairs are made to be identical in size through preprocessing; thirdly, rough matching is achieved through an SVD singular matrix decomposition method, accurate registration is achieved through an ICP algorithm after the point cloud pairs are identical in direction and become close to each other, and coordinate conversion relations between the point cloud pairs are obtained. Meanwhile, due to the practical production demands, local fitting deformation of a sole profile point cloud set is achieved through a Bezier curved line algorithm, the accurate calculation of the vamp glue spray trajectory is achieved, and thus glue spray automation is achieved.

The invention discloses a position error compensation method for a distance recognition superfluous kinematics parameter-based robot. Superfluous parameters in kinematics parameters are removed firstly to determine recognizable kinematics parameters of the robot; then, calculation is performed to obtain a parameter error calculation model, and compensation is performed by aiming at a parameter error, so that the problem of a singular matrix can be avoided, the calculated amount is effectively reduced, and the accuracy is improved; moreover, a compensation effect is obvious; the standard deviation of a distance error of the robot is greatly reduced, and the positioning accuracy of the robot is greatly improved. The position error compensation method for the distance recognition superfluous kinematics parameter-based robot, disclosed by the invention, can be widely applied to the field of robot research.

A transmitter beamforming technique for use in a MIMO wireless communication system determines a calibration factor and then applies the calibration factor to a transmit beamforming steering matrix developed using implicit beamforming, i.e., using an estimate of a forward channel disposed between a transmitter and a receiver based on a measurement of the reverse channel disposed between the receiver and the transmitter. The beamforming technique first determines descriptions of both the forward and reverse channels, determines an estimate of the forward channel from the description of the reverse channel, determines right singular matrixes which model the forward channel and the estimated forward channel and then develops a calibration factor from the determined right singular matrixes. The beamforming technique then applies the determined calibration factor to a steering matrix which is calculated using a standard implicit beamforming technique, i.e., assuming that the forward channel can be described as the transpose of the reverse channel. The use of this beamforming technique provides superior beamforming results when using implicit beamforming without having to take the necessary steps to determine a description of the actual forward channel each time a new steering matrix is to be calculated.

The invention discloses a non-uniform array design and direction of arrival (DOA) estimation method and mainly aims at solving the problems that arrangement is inflexible and computation complexity isrelatively high in the prior art. A realization process of the method comprises the following steps: according to maximum freedom degree of a nested array, determining head and tail position coefficients of a non-uniform array; calculating position coefficients of all the virtual array element positions meeting the non-uniform array and contained in a differential synthetic matrix; according to the position coefficients, finally obtaining positions of array elements of the non-uniform array; according to received data, calculating a data covariance matrix and vectorizing, so that received data r of a virtual differential synthetic array is obtained; performing redundancy elimination on the r and sorting to obtain received data of a virtual array, and then obtaining a non-singular matrix vof the virtual array; constructing a linear operator, and obtaining a signal subspace by estimating the linear operator; and constructing a selection matrix to obtain a rotation matrix, and finally estimating a direction of arrival by virtue of the rotation matrix. The method disclosed by the invention has the advantages that array configuration is flexible, characteristic decomposition does notneed to be performed on the data covariance matrix and spectral peak search does not need to be performed on the whole airspace angle under the same conditions.

Byte or symbol organized linear block codes are optimized in terms of reducing the number of ones in their parity check matrices by means of symbol column transformations carried out by multiplication by non-singular matrices. Each optimized symbol column preferably, and probably necessarily, includes a submatrix which is the identity matrix which contributes to low weight check matrices and also to simplified decoding procedures and apparatus. Since circuit cost and layout area are proportional to the number of Exclusive-OR gates which is determined by the number of ones in the check matrix, it is seen that the reduction procedures carried out in accordance with the present invention solve significant problems that are particularly applicable in the utilization of byte organized semiconductor memory systems. Reduced weight coding systems are also generated in accordance with weight reducing procedures used in conjunction with modified Reed Solomon codes. The decoding and encoding methods and apparatus are extendable to the inclusion of any number of check symbols.

The encoding method for LDPC code based on LU decomposition for optimized search matrix is realized by software pretreatment and hardware encoding. Wherein, eliminating correlative rows of check matrix, realigning the matrix; selecting nonsingular matrix with non-correlative column composite algorism for decomposition with simple LU algorism; optimizing search by diagonal element asymptotic one of nonsingular matrix; finally, selecting nonsingular matrix with lowest density, and reducing delay of whole coding system by time interval control parallel process hardware structure. This invention holds advantages in prior art, reduces calculation amount near Greedy algorism, and fit to any LDPC code special to abnormal code and normal node with correlative rows.

The invention discloses a face recognition method comprising the following steps of: establishing a face training database including a plurality of face training images, structuring a projection matrix based on all the face training images of the face training database, and projecting all the face training images in the face training database onto the projection matrix to obtain the characteristic matrixes of all the training images, respectively; projecting an image to be recognized to the projection matrix to obtain the characteristic matrix of the image to be recognized, calculating a distance between the characteristic matrix of the image to be recognized and the characteristic matrix of each training image, and selecting the type of the characteristic matrix of the training image having the shortest distance away from the characteristic matrix of the image to be recognized as the type of the image to be recognized. Therefore, no matrix inversion operation is needed in the method, and the restraint of the inversion operation of a singular matrix is avoided; and the method is higher in robustness and wider in application range, compared with Two-Dimensional Linear Discrimination Analysis (2DLDA).

The invention provides a magnetic resonance perfusion imaging postprocessing method and a system thereof. The method comprises the following steps of perfusing image filtering to an original brain, removing a brain edge and acquiring a filtered brain perfusion image; receiving an area selection instruction of a user, acquiring a selected area of a middle artery area in the filtered brain perfusion image and outputting a weighted artery input curve assessing an artery input function; according to an improved Gamma function, fitting the weighted artery input curve and a concentration time curve respectively so as to obtain the optimized artery input curve and the concentration time curve; according to an improved singular matrix decomposition method based on a non-parametric model, solving a relative quantification parameter of perfusion. Through carrying out weighted optimization on AIF and a solving matrix, sensitivity to a noise is reduced; a simplified and effective Gamma function is used to carry out fitting, a non-linear problem is converted into linear solution, and a post-treatment speed is accelerated.

The invention provides a colorful digital image repairing method based on compressed sensing, comprising the following steps: step 1, pre-treating an original damaged image and assigning a value of a damaged part of the image to be 0; step 2, dividing the image into blocks and analyzing to generate a singular matrix of the damaged part of each image block; step 3, multiplying the image blocks obtained in step 2 by the singular matrix to obtain a signal to be collected; then, representing the signal to be collected by using a CS sparse base to obtain a corresponding sparse signal; and finally, multiplying the sparse image by a randomly-generated Gaussian measurement matrix to obtain an observation signal; and step 4, utilizing a logarithm barrier function method based on total variation to recover a CS observation signal obtained in step 3 to obtain a recovered image. According to the colorful digital image repairing method based on compressed sensing, a specific structure which is not attached to the image has a similar recovering effect aiming to different images and can be used in different environments.

The invention relates to a public key code encrypt method, based on neural net chaotic attractor, which comprises: inputting key, coding attractor, coding clear text and generating ciphertext. Since the safety of said invention is based on the decomposition hardness of high-order singular matrix, but not based on the decomposition hardness of single large number factor as RSA method, it has high safety, high encrypt speed, and the applications in modern group encrypted communication and business encrypted communication.

The invention belongs to statistic pattern recognition and machine learning technical fields, especially is a linear feature extraction for k near neighbor classification. The invention provides that an adjacency image model and an identifying adjacency matrix representing the image model can be formed by a plurality of classes of data points under action of a local attractive force and a local expulsive force. A symmetrical matrix is derived from a target function, and as the result by performing spectral analysis on the matrix, the maximum precision of near neighbor classification can be obtained when the data is mapped to a space which is formed by tensing the eigenvectors corresponding to the negative eigenvalue of the matrix. The method determines the optimized dimension number of a new space suitable for classification by spectral analysis rather than debugging in experiments. Furthermore, the method also has the advantages of non-parameter, non-iteration, preventing local minimum and singular matrix.

The invention discloses a hyper-spectral image classification method based on singular value decomposition and neighborhood space information, comprising the following steps: inputting a training sample matrix of each category, and carrying out singular value decomposition on the training sample matrixes to get a right singular matrix corresponding to jth-category training samples; for data in the training sample matrixes, using a least squares method to calculate the residual of the data corresponding to each category, comparing the residuals of the data corresponding to all the categories, and classifying the data to the category corresponding to the smallest residual; and repeating the steps to get the category of each training sample, comparing the category of each training sample with the original category, getting a parameter making the classification accuracy rate of the training samples highest through iterative comparison, classifying each data in a test sample matrix, and outputting a classification result matrix. Through the classification method, the precision of classification is improved, the time for classification is shortened, and the level of automation of hyper-spectral image classification is improved.

A method for estimating a coil sensitivity map for a magnetic resonance (MR) image includes providing a matrix A of sliding blocks of a 3D image of coil calibration data, calculating a left singular matrix V∥ from a singular value decomposition of A corresponding to τ leading singular values, calculating P=V∥V∥H, calculating a matrix S that is an inverse Fourier transform of a zero-padded matrix P, and solving MHcr=(Sr)Hcr for cr, where cr is a vector of coil sensitivity maps for all coils at spatial location r, andM=((11…100…0………00…0)(00…011…1………00…0)…(00…000…0………11…1)).

Improved variable dimension vector quantization-related (“VDVQ-related”) processes have been developed that provide quality improvements over known coding processes in codebook optimization and the quantization of harmonic magnitudes that can be applied to a broad range of distortion measures, including those that would involve inverting a singular matrix using known centroid computation techniques. The improved VDVQ-related processes improve the way in which actual codevectors are extracted from the codevectors of the codebook by redefining the index relationship and using interpolation to determine the actual codevector elements when the index relationship produces a non-integer value. Additionally, these processes improve the way in which codebooks are optimized using the principles of gradient-descent. These improved VDVQ-related processes can be implemented in various software and hardware implementations.

The invention provides a hardware implementation method for channel equalization of an MIMO-OFDM (multiple-input multiple-output and orthogonal frequency division multiplexing) system. A Givens rotation-based QR decomposition method is provided and is applicable to a channel equalization module of the MIMO wireless system under the IEEE802.11ac protocol. A traditional matrix inversion hardware implementation method based on a block inversion idea involves mass second-order matrices and a calculation module of a complex multiplication and is only available to non-singular matrices. When the traditional QR decomposition method is used in matrix inversion, CORDIC (coordinated rotation digital computing) and SQUARE ROOT are involved, and many hardware resources are consumed. The invention provides a modified algorithm; matrix QR decomposition is achieved by a modified Givens rotation method to obtain a matrix inversion result. The hardware implementation method has the advantages that matrix inversion can be performed in a pipeline form, the use of coordinated rotation digital computing and square root is avoided, expenditure on the hardware resources is reduced, and a MIMO channel equalization process is implemented efficiently.

The invention discloses a satellite network flow prediction method based on space-time correlation. The method comprises the following steps: extracting satellite space-time correlation flow; reducingrelated flow dimensions of singular matrix decomposition, and extracting features; and establishing a satellite network traffic prediction model based on the gradient boosting regression tree. According to the method, singular matrix decomposition is carried out on the collected space-time flow to obtain the space-time related flow after dimension reduction, the space-time related flow serves asprediction input of a gradient boosting regression tree, then training and testing are carried out, and finally an accurate prediction value is output. According to the method, a new model is constructed by the gradient boosting regression tree in the gradient descending direction, the algorithm convergence method is optimized by improving the learning rate, in addition, the model is continuouslyupdated by minimizing the expected value of the loss function, so that the model tends to be stable, and finally, a future value is predicted by using test data for verification. Decision support is provided for planning of satellite network flow, and the method has a good application prospect.

The invention provides a pre-coding method and device according to a complex matrix, belonging to the field of singular value decomposition, wherein the method comprises the following steps: a step of constructing a real matrix: constructing the real matrix shown in the specification corresponding to the complex matrix H=A+Bi; a step of diagonalizing: diagonalizing the real matrix X through Jacobi iteration so as to obtain a diagonalizable matrix of the real matrix, wherein all non-diagonal elements in the diagonalizable matrix are not more than a pre-set threshold value; a step of calculating the matrix: calculating a singular value diagonalizable matrix D and a right singular matrix V of the complex matrix according to a Jacobi rotational transformation matrix used in the diagonalizable matrix; and a step of pre-coding: pre-coding a signal to be transmitted by an MIMO transmitter utilizing the right singular matrix V by taking the right singular matrix V as a filter matrix. By means of the technical scheme provided by the invention, the SVD processing time of the complex matrix can be reduced; and increasingly high real-time requests can be satisfied.

The invention discloses a TLS-ESPRIT-based harmonic and inter-harmonic detection method. The method comprises the following steps: arranging voltage signal sampling data x0, x1, and the like till xN-1into an LxM-dimension sampling data matrix X; estimating the number of signal frequency components k through the Gerschgorin disk method; carrying out singular value decomposition on the sampling data matrix X so that X is equal to R sigma U<T>, and acquiring the first k rows of a right singular matrix U to obtain a signal subspace Us; obtaining frequencies of components in sampling signals through the total least square method; and obtaining amplitudes and phase angles of the components in the sampling signals through the least square method. The method disclosed by the invention has the advantages of a shorter time window, a higher frequency resolution, higher stability and higher detection precision; and compared with the traditional windowed interpolation algorithm, the algorithm canachieve higher precision during inter-harmonic detection of a dynamic process of a power distribution network, and can more effectively monitor the dynamic process of a system.

The invention discloses an image denoising method based on multi-resolution singular value decomposition. The method comprises the following steps of step1, adjusting a size of an original image, and carrying out singular value decomposition on an adjusted image data matrix to acquire a left singular matrix; step2, carrying out product on a transposed matrix of the left singular matrix and the adjusted image data matrix to acquire a new image data matrix; step3, carrying out size adjusting on each row of data of the new image data matrix; step4, taking a low frequency portion as an original image and repeatedly carrying out at least once process from the step1 to the step3; step5, using a threshold denoising rule to process a matrix corresponding to a high frequency portion, recombining a processing result into the image data matrix with the same size with the original image and acquiring a denoised image. In the invention, when the noise is removed, simultaneously, high frequency information of the image can be well kept, and a process is simple and is easy to carry out.

The invention includes steps: using approximate solution of minimum mean square error instead of Wiener solution as an initial value; according to above basis, estimating noise power, and determining value of delta2; constructing broad sensing correlation matrix XíñXH+ delta2I; further, using Wiener solution obtains weight value Wú¢(XíñXH+delta2I) -1íñXíñdH. Advantages of the invention are: avoiding instability of weight value caused by singular matrices, extending algorithm applicability of intelligent antenna, and simplifying system structure, raising operation speed, tracking userí»s quick moving in real time, and resisting multiple interferences. Comparing with prior art, the invention overcomes disadvantages of large quantity of calculation, complex structure of system etc.

The invention provides a topic model-based software defect detection method and system. The topic model-based software defect detection method comprises the steps: enabling a static source code to generate a corresponding abstract syntax tree in space construction, enabling the abstract syntax tree to be mapped to a digitalized feature vector space, and constructing an original matrix; in semanticanalysis, constructing a defect mode library, carrying out singular matrix decomposition dimensionality reduction on the defect mode library to form a reconstructed matrix, carrying out cosine similarity calculation on the reconstructed matrix, and obtaining a defect detection result; and adding a machine learning algorithm for processing a natural language into software defect detection, and detecting hidden defects from semantic-level detection codes, so that the means of high-security software code defect detection is enriched, and whether a given code has defects similar to known defectsin a defect mode library or not is automatically detected, and the quality of software is further improved.

The invention discloses a channel prediction method of an MIMO closed-loop transmission system, and belongs to the field of digital communication. The method comprises the steps of predicting the channel state of the next moment according to the channel states of the former two moments, and then feeding back to a sending end for precoding; using the MIMO system to obtain the statistical information of a channel through the measurement for a period of time, wherein the statistical information comprises the Doppler frequency offset and a correlation matrix of a sending end and a receiving end; using a receiver to perform singular value decomposition on the channel matrixes at the current moment and the previous moment, extracting a plurality of columns of the right singular matrix to construct two new matrixes, and modeling as two adjacent points of the Grassmannian manifold; based on the geodesic theory of Grassmannian manifold, constructing a geodesic through the series matrix transformation, predicting the channel state of the next moment, and finally fedding back to the sending end. Compared with a traditional prediction method, the method is closer to the real channel state at the next moment, and the chord distance error performance is better.