114results about How to "Prevent error propagation" patented technology

A method for communicating video data on a wireless channel in a packet-switched network includes the steps of operating at a wireless terminal a compression in packets on the video data during a video coding operation, detecting wireless channel conditions and adapting control parameters of the video coding operation to the detected wireless channel conditions. The compression operation is a robust header compression operation and the step of adapting control parameters of said video coding operation is performed in dependence of information about the wireless channel conditions detected on a feedback channel made available in a decompression step associated with the compression operation.

The invention relates to multimedia communication technique, disclosing a multimedia communication method and the terminal design and implementation thereof, improving transmission reliability and communication quality. And the invention adopts ERRTP to provide a transmission layer packing form able to carry error resilient encoding solution related information on the basis of the existing RTP protocol, making multimedia data label the corresponding error resilient encoding solution related information while transmitted on ERRTP, thus merging error resilient mechanism into the transmission layer; in accordance with H.264 NALU structure, the invention provides a special ERRTP packing method and protocol header modification solution, able to integrate all NALU header bytes in the same ERRTP packet into its header, thus expressing important information of NALU in the ERRTP header and raising transmitting efficiency.

A method of inhibiting error propagation in a bidirectional digital compressed image data communication system is provided. A method of processing transmission data to inhibit error propagation in a digital image data communication system, includes (a) inputting an image frame from an external source, (b) checking for feedback error information including the location of an erroneous block on a compressed image frame detected during decoding by a decoder, the feedback error information received via a communication network, (c) if it is determined in step (b) that there is feedback error information, intracoding an erroneous block, the location of which is included in the feedback error information, and its search range, which is referred to encode the erroneous block using an intercoding method, among the image frame input in step (a), thereby constituting a compressed image frame, and (d) transmitting the compressed image frame constituted in step (c), via a communication network. When an error is detected by a decoder, it is fed back to an encoder, and the encoder uses a technique of compulsorily intercoding an erroneous block and the search range of the block with reference to feedback information, thereby inhibiting error propagation in a digital image data communication system.

The invention discloses a low-complexity polarization code decryption SCL algorithm based on segmented verification assistance, and the algorithm selects a parity check code, repeatedly uses the parity check code in a decryption process, and achieves the performances of SCL-CRC24. Moreover, compared with a conventional scheme, the algorithm is better in low signal to noise ratio anti-noise performance and error rate. In addition, the spatial complexity of the algorithm is lower than that of the SCL-CRC24, the time complexity is greatly reduced, and the decoding speed is greatly improved. Compared with a CRC-24 verification algorithm sacrificing multiple information bits, the algorithm employs a parity check method, enables verification elements to be distributed in the information bits, is repeatedly used in the decoding process, and is lower in time complexity than the prior art.

The embodiment of the invention discloses an image recognition method, device and equipment, a computer readable storage medium and a confusion perception convolutional neural network. The confusion perception convolutional neural network comprises a prediction classifier, a confusion perception model, a correction classifier group and a probability average layer, and the prediction classifier, the confusion perception model, the correction classifier group and the probability average layer are trained by using a training sample set and are used as traditional convolutional neural network classifiers. The confusion perception model is constructed based on a confusion matrix obtained by cross validation of a prediction classifier on the training sample set. Each correction classifier is obtained by using a confusion perception model as a decision-making system and training confusion class sample data with fuzzy boundaries in the training sample set. The probability average layer outputsa classification result of the to-be-identified image according to the category probability output by the prediction classifier and the category probability output by the target correction classifier, and the target correction classifier is a correction classifier selected by the confusion perception model according to the prediction category of the prediction classifier. The accuracy of image recognition can be improved.

An emulator for emulating a wireless network comprised of a plurality of RF nodes is comprised of a programmable controller for emulating the movements of the plurality of RF nodes within an emulated space. The controller provides both information and control signals based on the emulated movements. A programmable logic core receives a plurality of signals from the plurality of RF nodes and emulates signal propagation based on the information from the controller. A plurality of signal generation and conversion cards are interposed between the programmable logic core and the RF nodes. The signal generation and conversion cards are responsive to the control signals. Because of the rules governing abstracts, this abstract should not be used to construe the claims.

The invention relates to a processing method of word segmentation, part-of-speech tagging, entity identification and syntactic analysis, discloses a combined processing method of word segmentation, part-of-speech tagging, entity identification and syntactic analysis and belongs to the technical field of natural language processing. The core ideal of the method is characterized in that firstly, a combined model is constructed step by step; secondly, existing entity data and component syntax tree data are utilized to construct combined syntax tree data; thirdly, training data is extracted from the combined syntax tree data; fourthly, the training data is utilized to train the combined model; finally, the trained combined model is used for analyzing Chinese sentences to be analyzed to obtaina combined syntax tree serving as an analysis result. The method effectively avoids an error propagation problem and is an analysis method based on transfer, and the execution speed of the method is ensured.

An apparatus and method is disclosed for reducing error propagation in a decision feedback equalizer by constraining the values of feedback filter tap coefficients in a feedback filter of the decision feedback equalizer. The feedback filter tap coefficients that are constrained are calculated using a constraint cost function and a related constraint function. The constraint condition of the present invention reduces error propagation in a decision feedback equalizer by preventing error from circulating in a feedback loop of a feedback filter within the decision feedback equalizer.

The invention provides an event extraction method. The method includes obtaining original corpora, labeling elements in sentences in the form of [label-element], dividing the marked corpus into a training set and a test set; mapping the training set into a vector by a pre-training language model; obtaining word embedding vectors, inputting the word embedding vector into a neural network model; outputting sequence label information by the neural network model; establishing a loss function based on the sequence label information, evaluating the event extraction model by using the test set to finally obtain an event extraction model, adjusting the training set and the test set structure for multiple times, selecting the event extraction model with the best evaluation result as an optimal model, and inputting a new original corpus into the optimal model to obtain an event extraction result; the event extraction method provided by the invention belongs to a supervised neural network extraction method, is combined with the fine-grained contextualized word vector in the field, and more conforms to the scene of event extraction in the professional field.

The invention relates to a method for constructing a protocol data unit (PDU) of a media access layer (MAC layer) of a wireless communication system. The method comprises the following steps: constructing an MAC PDU in different sections, wherein one MAC Sub-PDU comprises four parts, that is, a synchronous code of N bit, an MAC head, an MAC Payload and a cyclic redundancy check code (CRC) of K bit. The invention further relates to an SDU (Service Data Unit) on the MAC layer and control information unit recovery method. By adopting the method, the throughput rate of the system is increased, and the precious shortage resource is saved.

The invention discloses a system and a method for compressing satellite images with low bit rate. The system comprises a control module, a wavelet transformation module, a CCSDS scanning coding module and a code stream splicing module. In the system and the method, image data are divided into a plurality of 32X32 (or 64X64 and 128X128) small images; then, the small images are transmitted into the wavelet transformation module, the scanning coding module and the code stream splicing module for obtaining compressing code streams. The wavelet transformation module, the scanning coding module and the code stream splicing module are used for line production among image blocks; and the scanning coding module is used for the line production along bit planes. The compressing effect of the system and the method for compressing the images is between JPEG 2000 and EZW; when the compression is carried out at a low bit rate, the compressing effect is equivalent to JPEG2000; a decoding re-synchronization mechanism is adopted for preventing error codes of space data from diffusing in transmission, with low power consumption; and extensively used parallel operation and a line structure increase the processing speed of a chip and meet the macro compressing requirement of the satellite images.

The invention discloses an entity relation extraction method and device. All entities having target relations with each sub-entity are stored in an obtained complete knowledge base, that is, attributevalues of all the stored sub-entities are complete, language data matched with an alternative relation mode obtained through analysis as well as entity pairs contained in the language data are determined through a corpus on the basis, the alternative relation has no semantic drift if the entity pairs exist in the knowledge base and can be determined as the target relation mode, and then the entity pair of a target relation is extracted according to the target relation mode. Understandably, the alternative relation mode has semantic drift if the entity pairs do not exist in the knowledge base,then the alternative relation mode is abandoned, and error propagation caused by iteration on the basis of the alternative relation mode is avoided.

The invention relates to a signal reception and detection method. The method comprises the following steps: an adjacent cell with a pilot signal for disturbing data signals in the cell is a disturbed cell; only the data signals in the cell in the different time-frequency resource positions with the pilot symbols in the disturbed cell are received and detected; and the first default value is adopted as the reception and detection result of the data signals which are not actually received and detected. In the invention, the data in the same time-frequency resource position with the pilot symbols in the disturbed cell is disturbed, therefore, the undisturbed data, i.e. only the data in the different time-frequency resource positions with the pilot symbols in the disturbed cell is actually received and accepted. The intensive interference of pilot frequency of the adjacent cell to the UE on the edge of the cell is avoided, the decoding performance of the UE is enhanced, and the demodulation errors are reduced; and the error propagation caused during the soft-decision iterative decoding is also avoided, so the data receptivity of the UE is further improved. The invention also provides a UE as well as a method and a device for signal transmission.

In order to realize effective picture compression without significantly impairing the resolution, it is proposed that the picture to be compressed be subdivided into a plurality of picture areas, for example into picture blocks each having 16×16 pixels, and the picture areas be subdivided into a plurality of layers, in particular into three layers, which correspond to different sampling patterns. The pixels (13-15) of a layer are in each case predicted from adjacent pixels of the other layers, in particular of the already previously processed and coded layers, the prediction area being coded. The invention is suitable in particular for use in video coders or video decoders with motion-compensated prediction, in order to compress the reference picture to be stored in a reference picture memory (5).

Linear predictive system with classification of LP residual Fourier coefficients into two or more overlapping classes, and each class has its own vector quantization codebook(s). The use of strong and weak predictors minimizes codebook size by only quantizing the difference between Fourier coefficients of a frame and the Fourier coefficients predicted from a prior frame. The choice of using either a strong or weak predictor adapts to the prior choice of predictor so that a strong predictor following a weak predictor is changed to a weak predictor to insure attenuation of error propagation as arise from frame erasures.

The invention relates to a multi-sequence frequency hopping anti-interference communication method based on pseudo-random linear frequency modulation, which comprises the following steps: a frequencymodulation slope is replaced with a controllable pseudo-random number in a linear frequency modulation signal; On the basis, the multi-sequence frequency hopping technology is combined with the multi-sequence frequency hopping technology; The carrier frequency of each hop is mixed with one pseudorandom linear frequency modulation signal, it is ensured that the extended bandwidth of the pseudorandom linear frequency modulation signal is smaller than the channel bandwidth, and a communication anti-interference waveform is generated. Based on multi-sequence frequency hopping, the multi-sequence frequency hopping anti-interference communication method based on pseudo-random linear frequency modulation is provided by taking further improvement of the anti-interference capability of multi-sequence frequency hopping communication as a research starting point, and can be applied to practice as an emergency communication means.

A method (100) and system (300) is described for processing video data comprising a plurality of images. The method and apparatus is for obtaining for labelling of a plurality of objects or regions in an image of a sequence of images followed by label propagation to other images in the sequence based on an inference step and a model.