100 results about "Pseudo data" patented technology

Skimming modules can be attached at self-service terminals, such as an automated teller machine or account statement printer, by manipulation with fraudulent intent by third parties who attempt to read the data from a magnetic stripe card introduced into the card reader (10). In order to counteract this problem effectively, a protective device (26, 28) is proposed that contains a protective generator (26) and an inductor (28) operatively connected thereto to generate an electromagnetic protective field (29) that is suitable for compromising the function of the spying device (SM), wherein the protective device (26, 28) generates an electromagnetic protective field (29) with a protective signal that simulates the type of signal that occurs when magnetic stripe cards are read. Through the protective field a special protective signal is induced in the potentially present skimming module to imitate the reading of the magnetic stripe card. The skimming card reader reads a pseudo magnetic stripe card, or is at least confused when spying on the genuine magnetic stripe card. In particular, the protective signal can contain pseudo data that are mixed with the actual card data at demodulation or suppress said data completely (capture effect) so that the third party ultimately obtains unusable data.

The invention discloses an epilepsy seizure prediction method based on a semi-supervised deep generative adversarial network. The method comprises the following steps: carrying out segmentation windowing processing on each EEG(electroencephalography) data of N original EEG samples through a sliding window in specified length, and during segmentation, carrying out superposition expansion processingon adjacent sub-windows to obtain S sections of sub-signals; carrying out short-time Fourier transform on the S sections of sub-signals to obtain spectrum signals, and converting the obtained spectrum data into an image for expression; inputting the processed picture information to a discriminator in the semi-supervised deep generative adversarial network for training and classification accordingto labeled training samples, training sample labels and unlabeled training samples; and inputting Z-dimensional noise data into a generator, and generating a pseudo-data image similar to distributionof the unlabeled data through the generator. The method gradually releases dependence of epilepsy seizure prediction on the labeled data through a semi-supervised learning mode, and obtains better classification accuracy.

The invention discloses an orchard planting monitoring system based on wireless sensor networks and a monitoring method thereof. The system comprises a plurality of wireless sensor networks comprisinga plurality of wireless sensor nodes, gateway nodes, a communication server connected with the gateway nodes and a host monitoring machine connected with the communication server. The method comprises the following steps: 1. self-organizing the plurality of wireless sensor networks by the plurality of wireless sensor nodes and transmitting monitoring data to the gateway nodes after collection atregular time, and then packing and transmitting the monitoring data to the communication server by the gateway nodes; 2. transmitting the monitoring data to the host monitoring machine by the communication server; and 3. pretreating the monitoring data and removing pseudo data by the host monitoring machine, and transferring an expert system for analyzing and processing the monitoring data and making a corresponding mentoring program. The invention has the advantages of reasonable design, low cost, simple and convenient operation, good monitoring effect and high degree of intelligence, provides the effective guiding scheme for peasant household and can carry out warning on the peasant household in real time.

In this invention, image data expressed by one component is encoded at a high speed by using a color image lossless encoder. To do this, a color conversion unit converts color image data read by a document reading unit into C, M, Y, and K data. In a color reading mode, a switching unit directly outputs the C, M, Y, and K data to an encoding unit. If the reading mode is a monochrome reading mode, the switching unit neglects the C, M, and Y components of the C, M, Y, and K data. Every time four K components are input, the switching unit supplies the four K components to the encoding unit as pseudo data of C, M, Y, and K color components. The encoding unit lossless-encodes the received C, M, Y, and K component data.

Improved privacy preservation techniques are disclosed for use in accordance with data mining. By way of example, a technique for preserving privacy of data records for use in a data mining application comprises the following steps/operations. Different privacy levels are assigned to the data records. Condensed groups are constructed from the data records based on the privacy levels, wherein summary statistics are maintained for each condensed group. Pseudo-data is generated from the summary statistics, wherein the pseudo-data is available for use in the data mining application. Principles of the invention are capable of handling both static and dynamic data sets

Method for more rapidly forming a stochastic model of Gaussian or Gaussian-related type, representative of a porous heterogeneous medium such as an underground reservoir, constrained by data characteristic of the displacement of fluids and punctual observations, these observations being uncertain and characterized by a probability law. It is based on a method for transforming static information, given in form of probability laws, into Gaussian punctual pseudo-data. This transformation technique has the advantage of being fully compatible with a gradual deformation method. In fact, it makes it possible to minimize an objective function J measuring the difference between the dynamic data (production data for example) and the corresponding responses simulated for the reservoir model considered. Minimization is carried out by combining realizations of a stochastic model on the one hand for the reservoir model and, on the other hand, for the Gaussian pseudo-data representative of the static information. During the minimization process, the combination coefficients are adjusted and the realizations allowing to reduce the objective function are identified. Application notably to petroleum reservoir development for example.

The invention provides an annotation data generation method and device and a computer readable storage medium. The method comprises following steps of S100: acquiring a data universal set and annotated annotation data sets included in the data universal set; S200: analyzing data characteristics of the annotation data sets and manufacturing pseudo-data sets satisfying the data characteristics according to the data characteristics; S300: based on a GAN neural network, expanding the pseudo-data sets so as to form an expanded data sets; S400: identifying whether data in the expanded data sets needs to be annotated, and screening annotated data so as to form a training data set; S500: carrying out neural network trainings on the training data set so as to form a training model; and S600: basedon the training model, cleaning data in the data universal set except for the data in the annotation data sets, annotating data satisfying the training model and arranging the data in the annotation data sets. Thus, based on little data, a training data set which highly matches the sample data and is highly random can be rapidly and highly efficiently generated, and data quantity of the annotationdata is improved.

A discriminator based on supervised learning includes a data expanding unit and a discriminating unit. The data expanding unit performs data expansion on unknown data which is an object to be discriminated in such a manner that a plurality of pieces of pseudo known data are generated. The discriminating unit applies the plurality of pieces of unknown pseudo data that has been expanded by the data expansion unit to a discriminative model so as to discriminate the plurality of pieces of pseudo unknown data, and integrates discriminative results of the plurality of pieces of pseudo unknown data to perform class classification such that the unknown data is classified into classes.

Disclosed are methods and systems for frustrating statistical attacks. A sender and receiver agree upon an encryption and integrity encoding scheme. For each “real” data packet, the sender generates integrity information, puts it in the packet, and then encrypts the packet using the agreed-upon scheme. Every now and again, the sender generates a “pseudo-data” packet encoded using a scheme different from the agreed-upon scheme. When the real-data and pseudo-data packets arrive at the receiver, they are decrypted using the agreed-upon scheme, and their validity is verified using the integrity information. Verification succeeds for real-data packets, and their data are accepted as valid. Verification fails for pseudo-data packets, however, and their data are discarded. An eavesdropper cannot differentiate pseudo-data from real-data and so cannot help but use all of the packets in its analysis. This mingling of pseudo-data with real-data frustrates the eavesdropper's statistical attempt to uncover the agreed-upon encoding scheme.

The invention provides a method for identifying calibration experiment pseudo-data in mining gas sensor monitoring data. Gas concentration data sequences are processes by a monitoring background through the following main steps that whether the increased value of two gas concentrations obtained successively is larger than a set threshold value or not is judged through comparison; the tour-inspection time t0 corresponding to the gas concentration value obtained in the previous time is determined, and the gas concentration maximum value and the tour-inspection time t1 obtained when the gas concentration maximum value occurs at the last time are determined within the tour-inspection threshold value range after the tour-inspection time t0; the gas concentration average value within the tour-inspection threshold value range after the t1 is calculated, and the tour-inspection time when it is determined that the gas concentration value is not larger than the average value within the threshold value range is denoted as t2; the difference value of the maximum value and the average value is calculated, and the difference value and the judgment threshold value are compared; and whether the gas concentration sequences between the t0 and t2 are the calibration experiment pseudo-data or not is determined according to set judging conditions. The pseudo-data generated by a gas sensor calibration experiment can be identified automatically and effectively in the gas concentration monitoring data sequences, and the automation level of coal mine safety monitoring is improved.

The invention discloses a pseudo industrial control processor and a data processing method and relates to the field of industrial control processors, The pseudo industrial control processor includes multiple CPU cores, a bus interconnect module, and a plurality of processor interfaces and a pseudo-functional module unit for realizing pseudo calculation and processing of input or output data of theprocessor. When data is output to the processor interface, pseudo judgment is performed on received downlink data, and correct downlink data are outputted to the processor interface according to thejudgment result; when the input data of the processor interface is received, the kernel identifier of the CPU core whose flow processing capacity satisfies the preset condition is determined, and thereceived uplink data is sent to the CPU core corresponding to the kernel identifier carried by the uplink data through the bus interconnection module. The invention relates to a pseudo industrial control processor, which can judge the output data of each CPU core, output the correct result, and introduce a pseudo data flow balancing mechanism to realize the load performance balancing of differentheterogeneous CPU cores.

Disclosed in the invention is an improved wavelet-analysis-based short-term power load prediction method. The invention relates to a wavelet-analysis-based short-term power load prediction method. With the method, a problem that the short-term power load can not be predicted accurately in the prior art can be solved. The method comprises the following steps: step one, selecting sample data; step two, carrying out pseudo-data monitoring on the selected sample data and correcting the pseudo-data; and step three, determining a sample set and carrying out normalization. Therefore, smooth and accurate sample data can be obtained and association with a predicted point load is high; and prediction becomes accurate and a prediction error is small. The load prediction technical level is improved reasonably; the planned electricity management can be realized well; the grid operation way and the generator set overhaul plan can be arranged reasonably; the coal and oil can be saved and the generating cost is lowered; the reasonable power supply building plan is formulated reasonably in real time; and the economic and social benefits of the power supply increase. The method provided by the invention is applied to the power load prediction field.

The invention relates to a machine translation automatic optimization method and system based on self-supervision, and belongs to the field of machine translation. The method comprises the following steps of 1, training a model by using a large-scale parallel corpus to enable the model to learn replacement operation; step 2, training the model by using artificially constructed pseudo data to enable the model to learn an insertion operation to obtain a trained model; and 3, performing candidate word prediction on each word and an inter-word gap in the to-be-optimized translation by using the trained model to finish post-editing operation of replacement and insertion. According to the method, the bilingual parallel corpus is used as the input of model pre-training, so that the model can obtain good optimization performance on different machine translation systems, and the model is a universal machine translation automatic post-editing model.

The object is to prevent a third party from specifying a condition of data communication executed by a radio terminal in a radio network using a radio communication device easily accessible to anyone such as WLAN. The radio base station 2 comprises a unit which determines whether to transmit a pseudo data frame assumed to be actual data communication according to a condition of data communication with the radio terminal 3 and a unit which stores information which is information for generating a pseudo data frame and based on actual data communication with the radio terminal 3, thereby generating and transmitting a pseudo data frame based on the stored information when starting pseudo data frame transmission.

A circuit for eliminating a skew between data and a clock signal in an interface between a semiconductor memory device and a memory controller includes an edge information storage unit which stores edge information output from the semiconductor memory device, a pseudo data pattern generating unit which outputs pseudo data including a pattern similar to actually transmitted data, a phase detecting unit which receives the edge information from the edge information storage unit and the pseudo data from the pseudo data pattern generating unit to detect a phase difference between the data and the clock signal and generate a corresponding detection result, and a phase control unit which controls a phase of the clock signal according to the corresponding detection result from the phase detecting unit, so as to eliminate a per-data input/output pin skew in a data write and read operation of the semiconductor memory device.

The invention discloses a machine translation model obtaining method and device, a text translation method and device and a storage medium. The invention relates to the field of natural language processing and deep learning. The method can comprise the following steps: obtaining training data that includes training data composed of bilingual parallel corpora and training data composed of pseudo bilingual parallel corpora, wherein the bilingual parallel corpus comprises a real source language text and a corresponding real target language text, and the pseudo bilingual parallel corpus comprisesa real target language text and a source language text which is converted into pseudo data in a predetermined mode; and training a machine translation model by utilizing the training data so as to obtain a target language text corresponding to the source language text to be translated by utilizing the machine translation model, and wherein Pinyin embedding is added in the input of the machine translation model. By applying the scheme of the invention, the accuracy of the translation result and the like can be improved.

The invention provides a parallel relatively prime array space-time expansion-based two-dimensional direction of arrival estimation method. The method comprises the steps of selecting and adding a special reference array element so that mutual correlation is performed o the reference array element by a sub-array of a relatively prime array to obtain correlation functions of different time lag andtake the self-correlation function as new receiving data; and performing pseudo-sampling to obtain a pseudo-data matrix by the new receiving data, vectorizing the pseudo-data matrix, converting a problem to a sparse optimization problem, and performing solving to obtain two-dimensional direction of arrival of a signal. The time-space information of the receiving data is fully utilized, a virtual array is expanded to the maximum extent, larger array aperture and higher freedom are achieved, and the direction of arrival estimation performance is better.