307 results about "Maximum eigenvalue" patented technology

Disclosed is a method for configuring transmission antenna groups in a mobile communication system with multiple transmission antennas and multiple reception antennas. The method includes the steps of: receiving a first matrix having a size of ‘Nt×Nt’, which corresponds to Nt number of transmission antennas, fedback from the mobile terminal, generating combinations of sub-matrixes of the first matrix corresponding to Nr number of reception antennas, and calculating sums of maximum eigenvalues; determining combinations of the sub-matrixes having a maximum value from among the calculated sums of eigenvalues to be transmission antenna groups for the Nt number of transmission antennas; and transmitting different relevant symbols from each other through the transmission antenna groups.

The invention discloses a detection method for an abnormal behavior of a vehicle based on spectrum clustering. The detection method comprises the following steps: obtaining a space-time track of a moving target through video tracking; removing abnormality and preprocessing, thereby obtaining a normal track; constructing an image for the track, thereby obtaining an undirected image corresponding to a track sequence; calculating similarity among tracks, thereby obtaining a similarity matrix; performing Laplace transformation on the similarity matrix, thereby obtaining a Laplace matrix; clustering the feature vector matrix of the front k maximal feature values; after performing mode learning on a motion track, obtaining motion modes of the target under a normal state; if a new track meets one of the motion modes, i.e. a normal motion mode, confirming that the traffic is normal; and if not, confirming that the vehicle abnormally runs, namely, the traffic abnormality occurs. According to the detection method, through the clustering learning for the vehicle track, the monitoring for the abnormal behavior of the vehicle is realized, the abnormal lane change is detected and the basis for automation of traffic management is supplied.

The invention discloses a multisource information fusion method in an evidence high-conflict environment, which comprises the following steps of: determining the mutual support degree of evidences according to evidence distances, further calculating the relative discount factor of each evidence by taking a characteristic vector corresponding to the maximum characteristic value of a evidence support degree matrix pattern as a weighting vector of the evidence, wherein the evidence with the maximum weighing coefficient is a key evidence; and discounting each evidence information, and fusing the information of multisource sensor data by using the D-S (Dempster / Shafer) rule so as to increase the accuracy and precision of a monitoring proposition judgment under an evidence high-conflict status.The method can effectively eliminate the uncertainties of multisource information, increase the accuracy of information evaluation and overcome the limitation that correct evaluation results can not be acquired by using the traditional D-S evidence theory under the evidence high-conflict status, and has the advantages of simple algorithm, high convergence rate and the like.

A method for tracking coronary artery motion includes constructing (11) a centerline model of a vascular structure in a base phase image in a sequence of 2D images of coronary arteries acquired over a cardiac phase, computing (12), for each pixel in a region-of-interest in each subsequent image, a velocity vector that represent a change in position between the subsequent image and base phase image, calculating (13) positions of control points in each phase using the velocity vectors, and applying (14) PCA to a P×2N data matrix XT constructed from position vectors (x, y) of N centerline control points for P phases to identify d eigenvectors corresponding to the largest eigenvalues of XXT to obtain a d-dimensional linear motion model {circumflex over (α)}p, in which a centerline model for a new image at phase p+1 is estimated by adding {circumflex over (α)}p to each centerline control point of a previous frame at phase p.

The embodiment of the invention provides a knowledge graph data processing method and device. The method comprises the steps of selecting a currently-processed target entity from other entities; obtaining a first feature vector of an initial entity and obtaining a second feature vector of the target entity; calculating a feature value corresponding to the target entity according to the first feature vector and the second feature vector; for the maximum feature value, adopting the first feature vector and the second feature vector corresponding to the maximum feature value for updating the initial entity and classification information and relation information of the other entities. According to the knowledge graph data processing method and device, the entities and relation information in a knowledge graph are converted into vectorization, prediction of the classification information and relation information of the entities is convenient, and more intelligent semantic analysis and processing are further performed, semantic analysis and processing are automated, construction is achieved not based on rules, the labor and maintenance cost is lowered, and the application range is wider.

A method of closed-loop multi-stream wireless communication between a transmitter (1) comprising a transmit antenna array of N transmit antenna elements and a reciever (3) comprising a receive antenna array (4) of M receive antenna elements, wherein a plurality of distinct data steams (x1 X2) are transmitted from the transmit antenna array to the receive antenna array and the data streams are weighted by respective complex weighting matcices before being applied to the transit antenna array. The distinct data steams are separated and estimated at the reciever. The distinct data streams (x1 xG) are applied to respective sub-groups (6, 7) of the transmit antenna elements at least one of which comprises a plurality of the transmit antenna elements each of the sub-groups comprising at least Nd transmit antenna elements, where M is greater than or equal to (N / Nd). The complex weighting matrices (v1 to vn) are functions of the respective transmission channels (hij) of the data streams (x1, to xG) including the respective sub-groups of transmit antenna elements. Nd is preferably greater than or equal to two. Each of the complex weighting matrices is calculated to be substantially equal to the eigenvector corresponding to the largest eigenvalue of the matirx H H, where H is the matrix of the equvalent channel including the respective sub-groups of transmit antenna elements (6, 7) seen by the corresponding data stream HH is the Hermitian transform of the matrix H. The number of the transmit antenna elements in each of the sub-groups is preferablly re-configurable during operation as a function of channel conditions.

The invention discloses a coordinated beamforming method for coordinated multiple point transmission (CoMp) downlink cells in a long term evolution-advanced (LTE-A) system, which mainly solves the problems of low throughput and co-channel interference of edge users among cells. The method comprises the following realization steps that: (1) a base station utilizes pilot frequency to obtain a channel matrix from the base station to users and resolves a related matrix; (2) the base station performs singular value decomposition on the related matrix according to a maximum signal-to-noise ratio criterion to resolve an eigenvector corresponding to a maximum eigenvalue of the related matrix; (3) a codon with the shortest distance from the eigenvector is found in a codebook space; (3) the base station finds a codon in a codebook according to a criterion of minimum user leakage signal; (4) whether the codons are the same is judged, if so, a beamforming vector of the users is found, and otherwise, operation of the next step is performed; and (5) a codon, the sum of the distances from which to the two codons is smallest, is found in the codebook space and taken as the beamforming vector of the users. The method can improve the average throughput of the edge users of the cells when effectively controlling leakage signal energy of the edge users of the cells, which inhibits the co-channel interference among the edge users of neighboring cells.

The invention discloses an uplink system resource allocation method based on beam forming; in allocation period, BS can obtain channel state information of each SS uplink in each time slot in a frame by the channel detection technology, and for each SS, the eigenvalue of MIMO channel matrix is decomposed in each time slot. The biggest eigenvalue after decomposition and corresponding eigenvectors form channel power gain matrix of each SS in each time slot; BS utilizes marginal effect and improved greedy algorithm to allocate time slots for each SS; each SS extracts allocated time slot information from a received downlink subframe and allocates power and bit; each SS sends the information of allocated power and bit to BS by a control part of an uplink subframe; BS is used for processing the received data to obtain the original data sent by SS. The invention also discloses a corresponding device, comprising a self-adaptive beam forming, sub-carrier allocation module as well as a power and bit allocation module. The method is of low complexity and good real-time property.

The invention discloses a power grid abnormal state detecting method based on a maximum feature value of a sample covariance matrix. The power grid abnormal state detecting method based on the maximumfeature value of the sample covariance matrix comprises the following steps: step 1, constructing a data source matrix Xs; step 2, acquiring a sliding window matrix X; step 3, standardizing the sliding window matrix X; step 4, acquiring a sample covariance matrix S; step 5, solving a maximum feature value as shown in specification of the sample covariance matrix; step 6, power grid state abnormalout-of-limit judgment: judging whether the maximum feature value as shown in specification is greater than a threshold value as shown in specification, if the maximum feature value is greater than the threshold value, determining that the state of a power grid is abnormal, and giving an alarm, and if the maximum feature value is not greater than the threshold value, determining that the abnormalstate does not exist at present, and returning to step 2 to continue carrying out a state abnormity detecting flow. The potential invalidation problem caused when a traditional mean spectral radius method detects the abnormal state of the power grid in a low signal-to-noise ratio scene is solved, meanwhile, calculation consumed time of a traditional method for detecting the abnormal state of the power grid on the basis of a random matrix theory is saved by simplifying a calculating link, and the calculating efficiency is improved remarkably.

The invention relates to a cooperative transmission method of uplinks of secondary users in a cognitive radio system. The cognitive radio system comprises primary users, secondary users and a base station serving both the primary users and the secondary users. The cooperative transmission method comprises the following steps: selecting cooperative users undergoing cooperative forwarding from idlesecondary users according to the channel information; setting beamforming factors for the selected cooperative users according to the spectrum access mode of a secondary source user; and finding out the normalized eigenvector corresponding to the maximum eigenvalue to serve as the beamforming vector of a receiving array antenna of the base station by decomposing the H2H1PsW (H2H1PsW)H eigenvalue.Compared with the existing technical schemes, the technical scheme adopted by the invention has the following advantages: the cooperative transmission of the uplinks is realized through integration and reutilization of the idle secondary user resources, thus improving the transmission performance; and the scheme is simple and easy to realize and can meet the requirements of the cognitive radio system for high flexibility and instantaneity.

The invention provides a solving method which coordinates a cognitive radio system to enhance robustness detection. The method for detecting a frequency spectrum hole in cognitive radio mainly comprises the following steps: independently sampling a frequency spectrum environment which needs to be detected by a plurality of cognitive users of a cognitive system in different geographical positions so as to obtain perceptive signals; sending perceptive data obtained by means of sampling by the cognitive user through a control channel in the cognitive system to a cognitive based station; receiving a plurality of perceptive signals and splitting the perceptive signals into a plurality of shorter perceptive signal sections with identical slot time by the cognitive based station, and obtaining a perceptive signal matrix through data infusion; and calculating the perceptive signal matrix and a covariance matrix, maximum and minimum eigenvalues of the covariance matrix and eigenratio of the maximum eigenvalues and minimum eigenvalues, comparing the eigenratio with a decision threshold, and judging whether the frequency spectrum hole exists or not. The method for detecting the frequency spectrum hole in the cognitive radio can effectively improve systematic detecting performance under the condition that the number of cooperative users are less.

The invention discloses a frequency spectrum perceiving method based on the estimation of a signal arrival direction, which comprises the steps: in a cognitive radio system to which smart antenna technology is applied, the direction and the angle of signal transmission are used as a frequency spectrum opportunity, and a master user signal is subjected to two-step perception: firstly, the master user signal is detected by using a method based on a generalized likelihood ratio test, namely that the ratio of the maximum eigenvalue to the sum of the eigenvalues of a sample covariance matrix is used as test statistics; the test statistics are compared with a decision threshold which is set according to a given false alarm probability so as to decide whether the master user signal exists or not; if a master user does not exist, then the channel can be used, and if the master user exists, then a classical MUSIC (Multiple Signal Classification) algorithm is further used to estimate the arrival direction of the master user signal. A cognitive user can select other directions for communication, which do not cause disturbance to the master user, and the frequency spectrum perceiving method can effectively perceive time, frequency and the frequency spectrum vacancy of a space domain, perceive the frequency spectrum vacancy of an angle domain, and increase the frequency spectrum opportunity.

A largest eigenvalue is determined among eigenvalues corresponding to a correlation matrix indicating correlations between a plurality of channels receiving incoming radar waves from an object that reflects a radar wave as a reference eigenvalue λ1. A ratio Rλi (=10 log 10(λi / λ1)) is calculated of each eigenvalue λ2 to λN to the reference eigenvalue λ1. Eigenvalues among the reference eigenvalue λ1 and the eigenvalues λ2 to λN of which the eigenvalue ratio Rλi is greater than a noise threshold TH are identified as eigenvalues in signal space. Eigenvalues of which the eigenvalue ratio Rλi is equal to the noise threshold TH or less are identified as eigenvalues in noise space. The number of eigenvalues identified as the eigenvalues in signal space is counted as the number of arrival signals.

The invention relates to a sound source direction positioning method based on a difference array. The sound source direction positioning method based on a difference array includes the steps: utilizing a plane microphone array to record input signals, and calculating a covariance matrix; further performing decomposition of characteristic constants to acquire the characteristic vector of the maximum characteristic constant and take the characteristic vector of the maximum characteristic constant as the target sound source, and calculating the covariance matrix of the target sound source; designing two orthogonal N-step dual difference array wave beam forming; based on the weight of wave beam forming and the covariance matrix of the target sound source, calculating the sound source direction candidate item; utilizing the sound source direction candidate item to calculate the sound source direction candidate items in other three quadrants, and forming a sound source direction candidate item group together; and utilizing the wave beam forming method for delay-and-sum to perform wave beam forming, and comparing the energy of the wave beams formed by the four sound source direction candidate items, and selecting the sound source direction candidate item with the maximum energy as the target sound source direction. The sound source direction positioning method based on a difference array has the advantages of high efficiency, low computational complexity and high accuracy, and is a new positioning method.

The invention provides a user electric equipment safety assessment method based on an improvement section analytic hierarchy process. The method comprises the following steps that a safety assessment index system of user electric equipment is established, and an expert grades a safety assessment index; the indeterminacy value of a judgment section is converted into a determinacy type value with the improvement section analytic hierarchy process; a section judgment matrix is constructed, and the maximum eigenvalue and eigenvector of the matrix are calculated; consistency of the section judgment matrix is tested, and weights of all safety assessment indexes are determined. According to the user electric equipment safety assessment method based on the improvement section analytic hierarchy process, the user power utilization safety can be effectively assessed through the assessment result, and a powerful basis is provided for preventing the human casualty accident of power utilization safety; compared with a common analytic hierarchy process calculation result, the feasibility of the method is shown, and a certain reference is provided for the user power equipment safety assessment.

The invention discloses a direct positioning method based on outer radiation source time delay and Doppler frequency, which overcomes the problem that the prior art does not fully utilize the information of a direct wave signal of an external radiation source. The direct positioning method comprises steps of performing time synchronization on double-channel reception systems of L observation stations, calculating the Fourier coefficient of the data received by the double channels of each observation station, transmitting the obtained array signal frequency domain data to a center station by the observation station, constructing a gauss maximum likelihood function for the data of the conversion value frequency domain and extracting the information matrix containing the echo time delay, the Doppler and the direct wave time delay information. The accurate positioning of the object can be obtained through setting a grid searching scope, calculating the maximum characteristic value corresponding to the data information array on the geography grid and searching the coordinate corresponding to the maximum value in the range of the grid. Compared with the traditional two-step positioning algorithm, the invention reduces the loss of the positioning information, and the positioning accuracy is dramatically improved.

The invention discloses a parallel community discovery method and device, and relates to the field of data mining. The method disclosed by the invention comprises the following steps: reading original social network data, converting the original social network data into an adjacency matrix way, and storing the original social network data on a HDFS (Hadoop distributed file system); calculating a stiffness matrix D and a Laplacian matrix of the adjacency matrix of a picture stored on the HDFS on a computing cluster configured with a Hadoop environment; carrying out the parallel Lanczos numerical value solving of a characteristic valve and a characteristic vector to the Laplacian matrix to obtain the characteristic vectors corresponding to the first K maximum characteristic values of the matrix; constructing a characteristic vector matrix for normalizing to obtain a standardized characteristic vector matrix and extract characteristics; taking each line as a point, and clustering the points into K types by a clustering method; according to the corresponding relationship of the points, equivalently dividing individuals in an original community into K types to finish the classification of the communities. The invention also discloses a parallel community discovery device. A technical scheme of the invention exhibits good adaptability to large-scale data.

The invention discloses a robust adaptive beamforming method based on shrinkage estimation covariance matrix reconstruction. The method comprises the following steps: performing linear shrinkage estimation on a sampling covariance matrix; performing feature decomposition on the sampling covariance matrix obtained by linear shrinkage estimation to obtain a noise subspace; calculating average noisepower in an expected small-angle interval according to the MUSIC spectrum, and removing the average noise power in the process of reconstructing an accurate expected signal covariance matrix; carryingout feature decomposition on the reconstructed expected signal covariance matrix, taking a feature vector corresponding to a maximum feature value as an initially estimated expected signal steering vector, and combining a modulus constraint and a noise power constraint of the preliminarily estimated expected signal steering vector into a Laplace constraint to obtain a corrected expected signal steering vector; and forming a robust adaptive wave beam by utilizing the sampling covariance matrix obtained by linear contraction and the expected signal steering vector. The main lobe wave beam and the output signal to interference plus noise ratio of the wave beam formed by the method are closer to ideal values, and the method has higher anti-interference capability.

The invention discloses a power distribution equipment state evaluation method based on variable weight and fuzzy comprehensive evaluation. The power distribution equipment state evaluation method comprises the steps of 1, selecting evaluation indexes; step 2, index weight primary distribution: adopting an analytic hierarchy process, establishing a comparative judgment matrix by comparing the importance between every two evaluation factors, and finally calculating a feature vector under the maximum feature value to obtain the weight of each index; step 3, calculating the health state score ofeach index; step 4, weight redistribution: carrying out weight redistribution, wherein xi is the health state score value of the ith index; step 5, performing fuzzy decision index normalization processing; step 6, calculating a membership function; step 7, establishing a fuzzy evaluation matrix; step 8, determining an evaluation result by utilizing a membership degree maximum principle. The problem that in the prior art, when power distribution equipment is evaluated, other indexes with small weights cannot correctly reflect the health state of the equipment due to the fact that the indexes with large weights have too large influences on the equipment is solved.

Disclosed is a wave beam figuration method and apparatus, in order to solve that the existing wave beam figuration method is unable to implement the performance optimization in a multiple-array system, because of independences between arrays. The method comprises: calculating and obtaining channel estimation results hka, ka=1...Ka, n=1...N of each aerial in the multiple-array aerial system, wherein, N represents array total, N>1, Ka is the aerial unit total in the n array; calculating the combined channel space covariance matrix R of each aerial array based on the hka; calculating the characteristic vector corresponding to the maximum characteristic value of R; performing the transformation only with amplitude variation for the webb; performing the figuration transmission by using the figuration vector transformed by the webb, because of the relationship between aerial arrays, the combined wave beam figuration processing can be performed for each aerial array to enable the performance of the wave beam figuration in the multiple-array system to be optimized obviously.

A method of identifying spherical objects in a digital image is provided. The image comprises a plurality of 3D surface points. The method includes computing, at each point in a domain of the image, a gradient of the image; computing an elementary structure tensor at each point in the domain of the image; determining a structure tensor for each point in the domain of the image; finding the eigenvalues of the structure tensors; and calculating an isotropy measure for each structure tensor, wherein said isotropy measure is defined by a ratio of a smallest eigenvalue of said structured tensor by a largest eigenvalue of said structure tensor, wherein a spherical object correspond to an isotropy measure equal to unity.

A method for deriving coefficients for a time domain equalizer function (24) as implemented by a digital signal processor (35) in a DSL modem (20) is disclosed. A transmitting modem (10), such as at a central office, issues a pseudo-random training sequence that is received by the receiving modem (20). Correlation matrices are derived by the digital signal processor (35), from which sets of eigenvalues and eigenvectors are derived. A flatness constraint on the frequency response of the time domain equalizer is established, and included with a flatness scaling factor (λ) into a minimization cost function. One or more values of the flatness scaling factor (λ), preferably between minimum and maximum eigenvalues, are evaluated in the cost function, to derive the optimum filter for the time-domain equalizer. The flatness constraint ensures that the time-domain equalizer is not subject to near null conditions and large variations in its frequency response.

The invention discloses a full-automatic fracture steel plate model personalized reconstruction method. The method comprises the following steps of broken bone image separation; broken bone three-dimensional reconstruction; broken bone axis extraction, wherein an initial sample matrix Z of three-dimensional models of a first broken bone and a second broken bone is generated, the sample matrix is subjected to centralization to obtain a matrix X, characteristic values of a covariance matrix C and a covariance matrix C of the matrix X are calculated, and characteristic vectors corresponding to the maximum characteristic values are selected as broken bone axes of the first broken bone and the second broken bone; broken bone section segmentation, wherein characteristic points are extracted from the three-dimensional models of the first broken bone and the second broken bone, the characteristic points are divided into the head part and the tail part according to a length buffer value and a mesh model, and four sets of head and tail characteristic points of the first broken bone and the second broken bone are formed separately; broken bone registration; virtual steel plate pre-bending, wherein point selection is conducted on spliced broken bone fracture surface attachments, and the size and shape of the steel plate model are determined; a normal vector value of each triangular patch is calculated, thickening is conducted on each plane to a certain extent according to the normal vector direction, the gaps are filled up, and finally the simulated steel plate three-dimensional data is obtained.

The invention discloses an edge detection method of a color textile texture image oriented to the textile industry. The method comprises the following steps: computing color differences between an upper half and a lower half, and a left half and a right half of a window within a sliding window by taking a pixel point as a center according to a fact that a human visual system is affected by surrounding adjacent pixel points when observing image colors; taking the color differences as partial derivatives in horizontal direction and vertical direction of the pixel point, wherein, the partial derivatives in the horizontal direction and the vertical direction of three color components of the pixel point form a 3*2 matrix D; computing a maximum characteristic value of the matrix D<T>D and a corresponding characteristic vector, and respectively taking the maximum characteristic value and the corresponding characteristic vector as a gradient size and a gradient direction of the pixel point; and computing and thinning an edge graph of the image according to the gradient size and the gradient direction of the pixel point. The method can resist against the interference of texture noise in the color textile image, improve the edge detection accuracy of an art pattern in the image, and provide basic edge information for further pattern editing and design, thus achieving an automatic tracing function oriented to the textile industry.

The invention discloses a precoding treatment method for a multi-input multi-output (MIMO) broadcast channel, and a precoding treatment device and a base station thereof. The method comprises the following steps: utilizing a unit matrix as a decoding matrix of a receiver o obtain a receiver initial decoding matrix; calculating an initial precoding emission matrix corresponding to each user data stream according to the receiver initial decoding matrix and the minimum mean square error standard; utilizing the multiplication of an emission channel matrix and the initial precoding emission matrix as an equivalent channel corresponding to the user, and performing the singular-value decomposition of the equivalent channel corresponding to the user and then selecting a right eigenvector with the maximum eigenvalue from the decomposition results; utilizing the multiplication of the initial precoding emission matrix and the conjugate transpose of the right eigenvector as a first precoding emission matrix; performing the iterative calculation of the final precoding emission matrix and the decoding matrix of the receiver by utilizing the first precoding emission matrix until the preset condition comes into existence. The precoding treatment method increases the convergence rate of calculating the precoding emission matrix and the receiver decoding matrix.

The present invention discloses an improved spectral clustering method based on a swarm intelligence algorithm. Feature vectors corresponding to former 2k maximum feature values of a Laplacian matrixare selected as source data of a cluster, and the swarm intelligence algorithm is employed to select high-quality initialization center points for cluster operation so as to improve the maximum accuracy of the cluster and the stability of multi-times cluster results. The Cuckoo search algorithm is introduced to search initialization center points, a fitness function in the process of the Cuckoo search algorithm employs an error square sum function and is applied into the spectral clustering to take data points of the minimum error sum square obtained through search as the initialization centerpoints. The Levy flight strategy in the Cuckoo search algorithm is introduced into the particle swarm optimization algorithm, in a condition that the convergence rate of the particle swarm optimization algorithm is slow, the Levy flight strategy is employed to generate step lengths with small frequencies and occasionally large step lengths, and speed updating formulas for different aspects are introduced in different step lengths.