49 results about "Markov transition" patented technology

Systems and methods for the design or implementation of statistical filters for use in the spectral shaping of transmissions are disclosed. A desired power spectrum may be mapped to find pole locations that approximate the desired spectrum. These pole locations may then be mapped to the edge or inside of an equilateral polygon lying inside a unit circle, the equilateral polygon having the same number of sides as the order of the statistical filter desired and one vertex mapped to unity, to yield a set of eigenvalues. These eigenvalues may be the eigenvalues of a stochastic matrix the elements of which may be the Markov transition probabilities for use in a statistical filter designed to achieve the desired power spectrum. Use of a statistical filter employing these Markov transition values may be utilized to shape UWB or other signals to achieve the desired power spectrum.

The invention discloses a tracking algorithm for high maneuvering targets. Under the colored noise condition, a target movement model and an observation model are established; the maneuvering targets are tracked by adopting an interactive multi-mode (IMM) algorithm-based Kalman filter; and a 'current' statistical model which is self-adaptively adjusted by acceleration in the IMM algorithm is combined with CV and CA models. The integral IMM algorithm performance is improved; by using system mode information in the current measurement, a Markov transition probability is calculated on line in real time, so that an accurate posterior estimate is obtained, and the model fusion precision is improved.

A three-dimensional urban traffic road network global state prediction method under a large data environment is disclosed. Aiming at a coupling multi-layer large-scale urban traffic road network, a global traffic mode containing coupling information is defined, a method of extracting a main road network characteristic is provided, a high dimension state time sequence is converted into a discrete state type sequence, and difficulties of high dimension data in processing and prediction aspects are simplified. From aspects of traffic periodicity, delay performance and other characteristics, a historical state transfer information database containing a multielement traffic characteristic is constructed, mass data information and a data driving idea are fully used and a Markov transition probability theory is applied so as to realize multi-step global traffic state prediction. Through the global traffic prediction, a traffic state of each area can be acquired one time. Compared to an existing traffic prediction method, by using the method of the invention, a long-term prediction result is acquired rapidly and a good prediction output effect is possessed.

Methods for dimensionality reduction of large data volumes, in particular hyper-spectral data cubes, include providing a dataset Γ of data points given as vectors, building a weighted graph G on Γ with a weight function w_ε, wherein w_ε corresponds to a local coordinate-wise similarity between the coordinates in Γ; obtaining eigenvectors of a matrix derived from graph G and weight function w_ε, and projecting the data points in Γ onto the eigenvectors to obtain a set of projection values Γ_B for each data point, whereby Γ_B represents coordinates in a reduced space. In one embodiment, the matrix is constructed through the dividing each element of w_ε by a square sum of its row multiplied by a square sum of its column. In another embodiment the matrix is constructed through a random walk on graph G via a Markov transition matrix P, which is derived from w_ε. The reduced space coordinates are advantageously used to rapidly and efficiently perform segmentation and clustering.

The invention discloses a method for classifying encrypted traffic and a server, and a computer readable storage medium. The method comprises a training phase and a classification phase, wherein the training phase comprises: obtaining and labeling the encrypted traffic of encrypted applications to obtain a training set; respectively extracting, from the training set, message type sequences of eachencrypted application and converting the sequences into corresponding coding sequences, and calculating the representative length sequence of the applications according to the packet length sequencesof the applications; constructing the Message type Markov transition matrix by using the coding sequences of each application, and constructing the length Markov transition matrix according to the representative length sequence; inputting the coding sequence and the representative length sequence respectively into corresponding transfer matrixes, to generate fingerprints corresponding to the encrypted traffic of the applications; and inputting each fingerprint into the classification model for training to obtain a classification model; and the classification phase comprises: for the data stream to be classified, inputting the fingerprint of the data stream into the trained classification model for classification.

The invention relates to a prediction method and device for an opportunity network link, and a readable storage medium. The method comprises the steps: reconstructing a Markov transition probability matrix based on a random walk algorithm and second-order neighbor information, and obtaining an improved restart random walk similarity index according to the Markov transition probability matrix; segmenting the opportunity network data to obtain multiple network snapshots according to a time series analysis method, and performing the randomized walk of each of the segmented network snapshots to obtain a corresponding similarity matrix according to the improved restart random walk similarity index; establishing a deep belief network prediction model, taking the similarity matrix as the model input, and obtaining a lowest-energy deep belief network prediction model to predict the opportunistic network link after the iterative training test. The method provided by the invention can achieve the precise prediction of the network link.

The invention provides a user abnormal behavior detection method and system, and the method comprises the steps: obtaining historical transaction data, carrying out the training through a Markov modeland a probability suffix tree model according to the historical transaction data, and obtaining a Markov transition probability matrix and a probability suffix tree transition probability matrix; combining the Markov transition probability matrix and the probability suffix tree transition probability matrix through a linear weighted fusion method to obtain a fraud early warning transition probability matrix, and obtaining a fraud transaction early warning model according to the fraud early warning transition probability matrix; identifying the historical transaction data through a preset critical value and the fraud transaction early warning model, and adjusting the preset critical value according to an identification result to obtain a final critical value; and identifying to-be-detectedtransaction data according to the final critical value and the fraud transaction early warning model to obtain a detection result.

The invention relates to a power grid security risk prediction method. The method comprises steps that screened network security posture factors are processed; security posture state classification standards are determined; a hidden Markov model is constructed according to actual power grid conditions; hidden Markov model parameters are updated according to present security posture of equipment; security posture of an integral electric power network is calculated; a prediction risk value of the equipment and a weight prediction risk value are calculated. Through the method, the network security posture factors are deeply researched, the factors having greatest influence on the security posture are screened out, data processing workload is reduced, model training is carried out through utilizing the data relevant to the security posture factors, a problem of over-hugeness of a hidden Markov transition matrix is effectively alleviated, moreover, problems of non-high prediction precision and parameter determination difficulty are alleviated.

The invention discloses a method and a system for detecting illegal data implantation internal attack in a smart power grid. The method adopts a bidirectional channel architecture consisting of gathering, clustering and control centers. The method comprises: (1) a gathering classification module; (2) a system initialization module; (3) a local trust evaluation module; (4) a global trust evaluation module; (5) an internal attach detection module. According to the method and the system, a trust model and a behavior model detection method are focused, and a monitoring and detection system for the illegal data implantation internal attack of the smart power grid is realized by utilizing the intrinsic characteristics of physical coupling and data consistency of the smart power grid on the basis of a graph theory, biotic community, power grid system characteristic vector and characteristic value, markov transition probability matrix theory and technology. The system can resist the internal attack of the smart power grid with high imperceptibility and powerful attack capacity; the implantation attack of illegal data can be high efficiently and accurately detected in the transient stable state process caused by external disturbance or faults.

The invention belongs to the technical field of radar communication and particularly relates to a design method of a radar maneuvering target tracking waveform. The method includes: building a movement model for a maneuvering target, calculating filter update matrix weight, the mixed input state of the movement module and a corresponding estimation error covariance matrix, using a linear or nonlinear filtering algorithm to acquire the local unbiased filter estimation value and target state estimation error covariance matrix of each submodel, performing compound matrix updating to obtain optimal state estimation and fusion estimation error covariance matrix, acquiring transmission waveform rotation parameters on the basis, using fractional Fourier transformation to rotate the waveform set by a user to obtain a new measuring error ellipse and transmission waveform, and performing Markov transition probability matrix updating to achieve good tracking precision. By the design method applicable to the radar communication technology, the problem that maneuvering target tracking is poor in robustness and low in accuracy is solved.

The invention belongs to the technical field of data processing and discloses a method and system for improving mining profits. Two conditions that miner proof of work is the same and miner proof of work is different are analyzed according to a PoW consensus algorithm and a dilemma model, thereby obtaining a miner mining profit matrix; and a ZD-free policy model, a single-ZD policy model and a double-ZD policy model are established, and system mining profits are calculated according to a profit matrix and a Markov transition matrix when the proof of work is the same and the proof of work is different. Compared with [1,0,1,0], [0.1,0.1,0.1,0.1] (a single-ZD policy), a double-ZD policy provided by the invention has the advantage that the mining profits are improved by 68% and 287%; and whenthe proof of work is different, compared with the other two policies, the single-ZD policy has the advantage that the mining profits are improved by 35% and 355%, and compared with the other two policies, the double-ZD policy has the advantage that the mining profits are improved by 48% and 139%. According to the double-ZD policy, each miner is guaranteed to carry out cooperation mining, and further the system mining profits reach the maximum value.

The invention relates to a determination method and system for a wind power state number upper limit in a wind power generating algorithm based on MCMC. According to the method, first, historical wind power exerting data is collected and sorted out, the data is divided into exerting states in different numbers, a state transition frequency number and a state transition probability matrix of all wind power exerting data points under the exerting states in different numbers are calculated respectively, a chi-square test is performed on the exerting states to determine a critical exerting state number conforming to Markov transition characteristics, and the critical exerting state number serves as a dividable exerting state number upper limit. A theoretical basis is laid for profound research of wind power exerting time sequence modeling. Through the chi-square distribution test method, a maximum wind power state number needing to be divided can be determined, and the precision of a wind power sequence generated through the MCMC method is improved to the maximum extent.

The invention discloses a method and a device for diagnosing faults of rail locomotive network control systems. The method includes building fault state identification models on the basis of Bayesiannetworks and Markov transition matrixes; acquiring MVB (multifunctional vehicle bus) data packet on MVB in real time, acquiring occurrence frequencies corresponding to fault feature points in currentmonitoring standard calibration time delta t and acquiring fault feature occurrence probability and mapping probability of the fault feature points on the basis of the occurrence frequencies; computing prior probability of fault nodes, which correspond to fault features, in four fault states by the aid of prior probability transition models and ultimately computing posterior probability of the fault nodes, which correspond to the fault features, in the four fault states on the basis of prior probability and mapping probability parameters by the aid of probability conversation formulas. The fact that the fault states corresponding to the highest posterior probability are current states of the fault nodes is the highest in reliability.

The invention discloses a public bicycle system regulation and control method based on a Markov model. The public bicycle system regulation and control method comprises the steps of: (1) acquiring bicycle lending and returning statistics through a public bicycle information center; (2) dividing the acquired bicycle lending and returning statistics into a prior group and a posterior group, and calculating the statistics of the prior group by using a first-order Markov model parameterize method to obtain prior information used for subsequent inference; (3) combining the prior information with data of the posterior group, and utilizing Bayesian reasoning combined with the Markov model to obtain a Markov transition probability matrix; (4) inferring a public bicycle system OD matrix on the basis of the Markov transition probability matrix; (5) and conducting public bicycle system scheduling by using the OD matrix. The public bicycle system regulation and control method based on the Markov model utilizes the OD matrix statistical inference for conducting public bicycle system scheduling in real time, can obtain data easily, avoids solving the unclosed problem, effectively utilizes the prior information and improves accuracy of the obtained data.

Provided is a highly accurate converter and the like that makes up for the instability of circuit elements, by focusing on a relationship between the Markov chain and β conversion. A converter 1 that determines the decoded value of a sample value x based on L-bit number b_i (i=1, . . . , L) includes a decoding section 3 to determine the decoded value x_D for γ=1/β (where 1<β<2) using equation (eq1). Further, the converter 1 also includes a matrix estimation section 5 to determine the Markov transition matrix based on b_i. Unlike a conventional method that pays attention to the lower limit of an interval, the decoding section 3 using equation (eq1) pays attention to the center of the interval, and this point is a significant difference.

The invention discloses an AMR fundamental tone delay steganalysis method based on a difference value Markov transition probability characteristic in a fundamental tone delay sub-frame group. For thetwo steganographic algorithms for modifying fundamental tone delay in speech coding, the method provides an AMR steganalysis characteristic based on a difference value Markov transition probability ina fundamental tone delay sub-frame group. The method perform classification prediction by using a support vector machine, and achieves a steganalysis method oriented to AMR fundamental tone delay. Compared with two conventional steganographic methods oriented to fundamental tone delay, the method of the invention can achieve a detection rate more than 95% in case of a relative embedding ratio of50%.

The invention provides an identity authentication method and device based on gait behaviors, equipment and a storage medium. The method comprises the steps of acquiring gait inertia data of a target object; based on the gait inertia data, performing gait period feature extraction to obtain a gait period sequence; converting the gait period sequence into a visual two-dimensional graph, including: performing visualization processing on the gait period sequence to obtain a two-dimensional graph based on a recursive graph, a Markov transition field or a Gram angle field; and performing identity authentication based on the two-dimensional graph to obtain an identity authentication result of the target object. According to the method and the device, the stable gait period is obtained by performing feature extraction on the gait inertia data of the target object, the gait period sequence is formed, and then the gait period sequence is converted into the visual two-dimensional graph, so that the two-dimensional neural network can be effectively utilized, and the accuracy and efficiency of identity authentication are improved.

The invention discloses a target advanced predicting and tracing method. A motion state of a target can be predicted in advance under the condition of low observation noise influence. The target advanced predicting and tracing method comprises the steps of firstly building a motion model for a moving target; then carrying out advanced prediction, that is, priori estimation, for the motion state ofthe target in the next moment by adopting a square root volume kalman filter method; building a Markov transition probability model by counting an error sequence between the priori estimation value and an observation value before the current moment; predicting a priori prediction error at the next moment by using the Markov transition probability model and correcting an advanced prediction valueat the next moment by using the error. The advanced predicting and tracing method can predict the motion state of the target in advance and can provide a target state prediction value in advance for atracing control system; and the control system can act in advance based on the advanced prediction value to overcome tracing lag and can keep continuous tracing for the target based on the advanced prediction value when the target is shielded.

The invention relates to a Markov chain method for classifying coherent generator groups of a wind power plant. The method comprises the steps as follows: active data and reactive data of wind turbine generators in the wind power plant are processed to enable the data to have the Markov property; a random process matrix is built, Gaussian weight processing is performed, and a Markov transition matrix is obtained; the Markov transition matrix is subjected to spectral analysis to solve a feature value, a left feature vector and a right feature vector, and the number of the classified coherent generator groups is determined; a transition probability distance between any two wind turbine generators is calculated and compared with a set threshold value to determine whether the two wind turbine generators are classified in the same generator group; and the classification result of the coherent generator groups is corrected comprehensively. The classification method utilizes actual physical data of the wind power plant, the real characteristics of the wind power plant can be reflected better, the foundation is established for completely and accurately analyzing the influence of scale type wind power plant grid connection on grid operation, and the method has positive influence on improvement of grid operation level.

The invention relates to an MP3 audio steganalysis method and device, and the method comprises the steps: obtaining a QMDCT coefficient of an MP3 audio to be analyzed, and obtaining a QMDCT coefficient matrix, i.e., a matrix MQMDCT, and a corresponding QMDCT coefficient absolute value matrix, i.e., a matrix MAQMDCT after preprocessing; the Markov transition probability of the matrix MQMDCT and the Markov transition probability of the matrix MAQMDCT in the row and column directions are calculated, and intra-frame and inter-frame Markov features are obtained; respectively carrying out downsampling on the matrix MQMDCT and the matrix MAQMDCT, respectively calculating the Markov transition probability of each downsampling matrix in the row and column directions, and obtaining a block mean Markov feature; and using the truncation threshold T to obtain a feature vector F, and training through an SVM classifier to obtain an MP3 audio steganalysis detection model.

The invention relates to a data enhancement method for vibration signals of a converter transformer, and belongs to the field of electric power. The method comprises the following steps: firstly, a Markov transition field (MTF) matrix of a vibration signal sequence is calculated, and the matrix is converted into a two-dimensional feature map to retain the correlation of a time sequence; and on this basis, a self-attention module is introduced to optimize a generative network, which can accept information between long-distance pixel points and realize a global receptive field between the pixel points. Meanwhile, the identification network adopts convolution kernels of different sizes for parallel feature extraction so as to increase the range of a receptive field and obtain deeper features. According to the model, a two-dimensional image data set is established according to actually measured converter station vibration signals, effective training data can be further generated, and the problem that the converter transformer state recognition accuracy is low due to the fact that the number of data sets is unbalanced is solved.