115results about How to "Improve decoding efficiency" patented technology

The invention relates to the technical field of remote controller awakening, and discloses a remote controller awakening method, device and equipment and a storage medium, and the method comprises the steps: carrying out the multi-cascade coding processing of a key signal of a remote controller, obtaining the multi-cascade coding data, and generating a session key; performing adaptive carrier modulation to obtain a dynamic modulation signal; performing multi-band emission control processing and adaptive power control to obtain an emission control signal; performing low power consumption management on the remote controller system, creating a multi-stage dormancy strategy and a dynamic power consumption control scheme, and controlling the signal processor to receive the multiband optical signal; performing parallel processing and adaptive decoding on the multi-band optical signal to obtain decoded multi-cascade coded data; and based on the session key, performing multi-factor identity verification processing on the decoded multi-cascade coded data to obtain a secure wake-up instruction, thereby improving the reliability and security of wake-up, optimizing the power consumption management strategy, and enhancing the signal processing capability.

The invention discloses a rate division multiple access communication network total rate optimization method and system, and belongs to the technical field of communication. The method comprises the following steps: constructing a downlink rate segmentation multiple access communication system model based on an intelligent reflecting surface; the communication system model comprises a base station, an intelligent reflecting surface and a user side; grouping users in the user side, and constructing a hierarchical rate segmentation multiple access model based on user grouping; the hierarchical rate segmentation multiple access model comprises a corresponding relation between the user group and information flow generated by the base station; and according to the hierarchical rate segmentation multiple access model and the intelligent reflecting surface, assisting downlink communication between the base station and the user side. According to the method and the device, the layered rate division multiple access model is constructed by grouping the users, and the system performance is improved by combining the intelligent reflecting surface passive beam forming and the rate division multiple access active beam forming, so that the problems of spectrum resource shortage and signal interference can be solved, and the spectrum efficiency is greatly improved.

The invention provides a visual question and answer method based on knowledge retrieval enhancement, and relates to the field of computer vision and natural language processing, and the method comprises the following steps: respectively encoding a multi-modal query and a knowledge text by using a multi-modal encoder; calculating the similarity between the multi-modal query and the feature vector of the knowledge text by using a vector dot product, and retrieving to obtain a plurality of related external knowledge; the method comprises the following steps of: extracting the most useful visual representation from a frozen image encoder by using a lightweight visual Transform module; using a T5 encoder to independently encode the joint input of a plurality of knowledge paragraphs, images and questions; the first N most representative marks are connected in series and input into a T5 decoder for joint decoding, and finally a correct answer is generated. According to the method, the problems of visual information loss and insufficient knowledge utilization in the existing method are solved, and the performance of the system in the knowledge-intensive visual question and answer task is remarkably improved.

The invention relates to the technical field of communication, and particularly provides a terminal signal, a signal processing method and device, a passive tag and a product. The terminal signal provided by the invention is obtained by encoding a first code element and a second code element; wherein the length of the first code element is the same as that of the second code element; the code pattern of the first code element is a full high level; the code pattern of the second code element comprises at least one low level and a plurality of high levels. The two code elements provided by the invention at least only comprise one low level, the proportion of the high level is higher, and the terminal signal provided by the invention can provide longer charging duration for the terminal under the condition of ensuring that the terminal receives the same signal, thereby effectively ensuring that the terminal can obtain enough working electric quantity.

The invention relates to a decoding method and device for adaptive threshold adjustment, a decoder and a medium. The method comprises the following steps: initializing soft information, and setting a threshold value; according to the received soft information, performing iterative updating on the check node based on a threshold value, and generating check information and a classification result; in an iteration process, adaptively adjusting a threshold value according to a classification result, and judging whether to update the check node according to the classification result; the updated check node transmits the check information to the connected variable node, and the variable node performs corresponding updating according to the received check information to obtain a variable node updating result; and performing decoding judgment according to the variable node updating result, if the judgment does not pass, continuing iterative processing until a preset number of iterations is reached, and outputting a decoding result. According to the method provided by the invention, the calculation complexity can be reduced, the decoding speed is increased, the decoding performance is improved, and the robustness is high.

The invention belongs to the field of secure multi-party computing, and discloses a method for combining an RS (Reed-Solomon) code list decoding algorithm, cyclic redundancy check and a secure multi-party computing scheme, which improves the security of secure multi-party computing. And theoretically, the algorithm combining RS code list decoding and CRC exists, so that the secure multi-party computing scheme can resist at most one half of hostile attackers. Meanwhile, a specific method combining an RS code list decoding algorithm and CRC is given, the limitation of the unique decoding radius of the RS is broken through, the error correction capability of the RS code is improved with high probability, and the security of secure multi-party calculation is improved. The method specifically comprises the following steps: calculating a CRC check code of information and combining the CRC check code with the information to obtain a combined code word; performing RS coding on the code word; carrying out list decoding on the attacked code word; each code word in the list is verified, and original information and a secret value are obtained in a probabilistic mode.

The invention discloses a traveling salesman problem solving method based on a quantum self-attention neural network, and belongs to the field of quantum computing. A classic encoder-decoder structure is adopted, a quantum self-attention layer is designed in the encoder and comprises a quantum multi-head attention layer and a quantum full-connection feed-forward layer, and a variable component sub-circuit is used in a sub-layer to replace a full-connection layer of a classic neural network. Firstly, an encoder receives an input data set, node features are embedded into a quantum self-attention layer, and a complex relation between nodes is extracted through a quantum multi-head attention mechanism. The decoder then stepwise outputs the sequence of nodes to form an optimal path according to the embedding generated by the encoder. In the model training stage, a REINFORCE deep reinforcement learning algorithm is utilized, the total path length serves as an optimization target, and model parameters are updated through strategy gradients. Compared with the existing algorithm, the method has the advantage that the performance is well improved.

The invention provides a cascade coding and decoding method based on RS-convolutional codes in an FSO system. Belongs to the technical field of atmospheric laser communication. The method comprises the following steps of: firstly, inputting a source data sequence into a sparse code encoder for blocking and RS (Reed-Solomon) coding to generate an RS code word containing error correction redundancy; and then, the RS code words are rearranged by using an interleaving technology, so that the occurrence of continuous errors is reduced. And then, after the interleaved RS code words are subjected to parallel-serial conversion, the RS code words are input into a dynamically-adjusted convolutional encoder, encoding parameters are dynamically adjusted according to the current communication environment, and a redundancy-enhanced encoding sequence is generated. After signals are sent and received, the interleaved RS code words are recovered through iterative decoding, and finally, the RS decoder corrects symbol errors and recovers source data by using dynamically adjusted decoding parameters. In the whole process, data transmission in the FSO system is realized through an RS-convolutional code cascade coding and decoding method, and the reliability and accuracy of data transmission are improved.

The invention discloses a content missing mail analysis method and device, and relates to the technical field of emails, the method comprises the following steps: carrying out text content extraction and classification on a mail text to obtain the category of the text content; the category comprises a plain text format, a webpage format and an attachment; taking the text content of which the category is the plain text format and the webpage format as a to-be-decoded text; decoding the to-be-decoded text according to a corresponding coding mode to obtain a decoded character string; performing segmented character set detection on the decoded character string according to a set byte length, and adding a character set corresponding to a character set detection result meeting a set condition into a character set list; and performing character conversion on the decoded character string by traversing a character set in the character set list until a decoding result passes semantic coherence judgment, thereby obtaining an original mail text. According to the method and the device, the effective analysis of the mail with the missing content is realized.

The present invention discloses an image encoding / decoding method and device. The image decoding method according to the present invention includes: obtaining symmetric motion vector difference mode available information from a bitstream; obtaining first prediction direction zero motion vector difference information from a bitstream; obtaining symmetric motion vector difference mode information of a current block from a bitstream based on the symmetric motion vector difference mode available information and the first prediction direction zero motion vector difference information; obtaining first prediction direction reference picture index information, second prediction direction reference picture index information and first prediction direction motion vector difference value based on the symmetric motion vector difference mode information; and generating a prediction block of the current block by using at least one of the first prediction direction reference picture index information, the second prediction direction reference picture index information and the first prediction direction motion vector difference value, wherein the step of obtaining the reference picture index information of the first prediction direction, the reference picture index information of the second prediction direction and the motion vector difference value of the first prediction direction is obtained by deriving from the bitstream instead of decoding the bitstream when the symmetric motion vector difference mode information of the current block indicates the symmetric motion vector difference mode.

The present application relates to a method, device, computer equipment, storage medium and program product for processing video stream data. The method comprises: upon receiving a stream mode start signal sent by a cloud host, initializing a rendering engine and a hardware decoder, receiving video stream data sent by the cloud host, calling a hardware decoder to decode the video stream data, obtaining decoded data, rendering the decoded data to an external texture structure through a rendering engine, and rendering the external texture structure to a visible area of ​​a cloud computer client through a rendering engine, and upon receiving a stop signal sent by the cloud host, stopping the process of calling a hardware decoder to decode the video stream data, and releasing decoding resources through the hardware decoder. The use of this method can ensure real-time processing of video stream data, improve decoding efficiency, and reduce resource consumption.

The present invention discloses a method for recognizing EEG speech imagery based on dynamic hypergraph learning. The method is as follows: 1. EEG data acquisition; 2. Preprocessing and feature extraction; 3. Establishing an objective function; 4. Solving the variables in the objective function to obtain a feature projection matrix and a speech imagery category corresponding to the measured EEG data. 5. Analyzing the feature projection matrix to obtain the importance of each channel and each frequency band, and reconstructing the sample features based on the importance as the model input to obtain the final recognition result. The present invention uses a hypergraph to construct an EEG data model, mines high-order information between data, better characterizes the correlation of data, and dynamically and adaptively updates the vertex weights, hyperedge weights and hypergraph structure in the hypergraph during the label learning process, thereby improving the learning ability and decoding accuracy of the model. In addition, the dynamically updated feature projection matrix projects the original features into a low-dimensional subspace, thereby improving the robustness of the model.

The invention discloses an attribute information decoding and coding method, device and equipment, and belongs to the technical field of communication, and the decoding method comprises the steps: carrying out the decoding and inverse quantization of to-be-decoded data, and obtaining a reconstruction transformation coefficient corresponding to each node; the reconstruction transformation coefficients comprise a first reconstruction transformation coefficient corresponding to the brightness component, a second reconstruction residual transformation coefficient corresponding to the first chrominance component and a third reconstruction residual transformation coefficient corresponding to the second chrominance component; determining a second reconstruction transformation coefficient according to the first reconstruction transformation coefficient and the second reconstruction residual transformation coefficient; executing at least one of the following steps: determining a third reconstruction transformation coefficient corresponding to the second chrominance component according to the second reconstruction transformation coefficient and a third reconstruction residual transformation coefficient; determining a third reconstruction transformation coefficient according to the first reconstruction transformation coefficient and the third reconstruction residual transformation coefficient; and determining a third reconstruction transformation coefficient according to the first reconstruction transformation coefficient, the second reconstruction transformation coefficient and the third reconstruction residual transformation coefficient.

Embodiments of the present application provide a speech recognition method and device, electronic equipment and storage medium, at least applied to the field of artificial intelligence and speech recognition, wherein the method comprises: performing vector coding processing on the audio feature vector of the speech to be recognized to obtain an audio coding vector; performing classification processing on the audio coding vector to obtain a prediction probability distribution of each predicted character in a preset vocabulary corresponding to each speech frame in the speech to be recognized; performing pruning processing on the audio coding vector based on the prediction probability distribution to obtain a pruned audio coding vector; and performing speech recognition on the speech to be recognized based on the pruned audio coding vector to obtain a speech recognition result. Through the present application, the decoding calculation amount in the speech recognition process can be reduced, the decoding efficiency can be improved, and thus the speech recognition efficiency can be improved.

The invention relates to the technical field of voice recognition, in particular to a voice recognition method and device, electronic equipment and a storage medium, and aims to improve the voice recognition efficiency. The method comprises the following steps: acquiring to-be-recognized voice, encoding the to-be-recognized voice through a target encoder, and acquiring candidate encoding features corresponding to the to-be-recognized voice; obtaining a blank frame probability corresponding to each voice frame in the to-be-recognized voice according to the frame distribution information of the to-be-recognized voice, and screening out at least one target frame from the to-be-recognized voice based on the obtained blank frame probability; the frame distribution information represents probability distribution of each voice frame in a dictionary space; screening out a target candidate coding feature corresponding to at least one target frame from the candidate coding features; and decoding based on the target candidate coding features to obtain a text recognition result corresponding to the to-be-recognized voice. According to the invention, the frame rate compression is carried out through the frame distribution information, only part of important frames are decoded, and the speech recognition efficiency can be effectively improved.

The invention relates to a video stream data processing method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: under the condition that a stream mode starting signal sent by a cloud host is received, initializing a rendering engine and a hardware decoder, receiving video stream data sent by the cloud host, calling the hardware decoder to decode the video stream data to obtain decoded data, and rendering the decoded data to an external texture structure through the rendering engine, and the external texture structure is rendered to the visual area of the cloud computer client through the rendering engine, the process of calling the hardware decoder to decode the video stream data is stopped under the condition that a stop signal sent by the cloud host is received, and decoding resources are released through the hardware decoder. By adopting the method, the real-time processing of the video stream data can be ensured, the decoding efficiency is improved, and the resource consumption is reduced.

The present disclosure relates to a coding method, device, electronic equipment and storage medium. The decoding method comprises: writing a motion vector index corresponding to a target motion vector, an offset direction index corresponding to a target offset direction, an offset step index corresponding to a target offset step, and a multi-function index into a coding bitstream corresponding to an image frame; the offset direction index is used to indicate that the offset direction is a horizontal X-axis direction or a vertical Y-axis direction; wherein, in the case that the prediction direction of the current coding unit is bidirectional, the multi-function index is used to represent a candidate coding unit index; in the case that the prediction direction of the current coding unit is unidirectional, the multi-function index is used to represent the positive or negative direction of the X-axis or Y-axis in the offset direction. The method can effectively reduce the influence of MMVD optimization on decoder parallelism and improve decoding efficiency.

Provided are an image encoding and decoding method and a recording medium storing a bit stream. The present invention relates to an image encoding and decoding method. To this end, the image decoding method comprises the following steps: determining a reference sample of a current block; filtering the reference sample based on the characteristics of a region including the reference sample; and performing intra-frame prediction by using the filtered reference sample.

Disclosed is a decoder. At least one computing unit of the decoder is configured to: acquire soft bit information to be decoded of a variable node n in a basic matrix; determine whether the number of decoding iterations i is less than a decoding iteration threshold; if yes, determine whether the number of decoding layers k is smaller than the number of rows a of the basic matrix; if it is determined that k is smaller than a, determine the soft bit information of the variable node n in a kth layer according to the soft bit information of the variable node n in a k-1th layer, and determine variable node information in a k+1th layer according to the soft bit information of the variable node n in the k-1th layer; determine check node information in the k+1th layer according to the variable node information in the k+1th layer; update k to k+1, re-determine whether k is smaller than a; if it is determined that k is greater than or equal to a, update i to i+1, and re-determine whether i is smaller than the decoding iteration threshold until i equals to the decoding iteration threshold. Also disclosed are a decoding method and a computer storage medium.

This invention discloses a layered decoding method and apparatus for LDPC frames based on mutual information, relating to the field of digital signal transmission in the physical layer of wireless communication. The method includes the following steps: 1) Calculating the LLR value of a given codeword to be decoded, and performing Q-ratio specific-point quantization to maximize mutual information on the entire frame's LLR value; 2) Calculating the PP value of the Q-ratio specific-point quantized LLR value according to Bayes' formula; 3) Layering the PP values ​​of the LLR values ​​of the data to be decoded according to the coefficients of the parity check matrix H; 4) Calculating the information within the information node; 5) Performing parity check node update calculations according to the PMS decoding algorithm to obtain the information value outside the parity check node; 6) Updating the information node using the information value outside the parity check node; 7) Symbol decoding; 8) Calculating the syndrome according to H. If the result of the syndrome is 0 or the maximum number of iterations is reached, outputting the three-frame codeword result; otherwise, returning to step 3) for the next decoding iteration. This invention can effectively improve the decoding processing speed of each frame.

The invention provides a greedy confidence method and device applied to low-density parity check codes, and the method comprises the steps: initializing a decoder, and obtaining a variable node and a check node after initialization processing; transmitting a message obtained based on the received signal to a check node through the variable node; a greedy optimization strategy is adopted to carry out partial node updating and iteration in the message transmission process of the variable nodes and the check nodes, and an optimal analysis result is obtained; and obtaining decoding information according to the optimal analysis result, and outputting a decoding result for the decoding information. According to the invention, the greedy algorithm is applied to the BP algorithm, the BP algorithm is optimized, messages of part of variable nodes and check nodes are updated according to a greedy optimization strategy, the complexity of the BP algorithm is reduced, the iteration time is effectively reduced, the decoding time delay is reduced, and the LDPC decoding efficiency is improved.

The embodiment of the invention provides a data processing method of a network communication system and related equipment, and the method comprises the steps: obtaining a to-be-decoded data vector sequence and a to-be-decoded matrix sequence from a plurality of batches of coding packets transmitted by a data transmitting end through network coding, and enabling the to-be-decoded matrix sequence to comprise a plurality of to-be-decoded matrixes; performing row simplification on all matrixes to be decoded in the matrix sequence to be decoded to obtain a simplified matrix sequence, and recording a corresponding row transformation operation sequence; and performing simplified iteration on the simplified matrix sequence until the variable columns of all the to-be-decoded matrixes are solved or decodable variable columns do not exist, and the row transformation operation sequence is used for performing transformation operation on the to-be-decoded data vector sequence to obtain decoded data, so that the solving efficiency of the to-be-decoded data vector is improved, and the decoding efficiency of the to-be-decoded data vector is improved. And the decoding efficiency and the decoding reliability of the matrix sequence to be decoded are improved.

Provided is an image encoding / decoding method and device. The image decoding method disclosed herein may include the following steps: obtaining maximum transform block size information; deriving the maximum transform block size based on the maximum transform block size information; and determining whether to apply a secondary inverse transform to the current block based on the maximum transform block size and the horizontal size and vertical size of the current block.

This invention provides a data processing method and mobile device based on a brain-computer interface. By fusing two multimodal neural signals, EEG and fNIRS, and performing precise alignment based on timestamps, the method obtains the current application scenario and the proportion of computational resources required for feature extraction from the aligned signals. Furthermore, it utilizes the spatiotemporal separation convolutional layers and attention layers of a lightweight neural network model to dynamically adjust feature extraction and weight allocation according to the scenario, rather than processing all brain regions and frequency bands indiscriminately, thereby improving the adaptability and decoding efficiency of the mobile device in different application scenarios.

The invention relates to an audio and video call method and system, and a storage medium. The method comprises the following steps: collecting audio and video data; compressing and encrypting the collected audio and video data; transmitting the compressed audio and video data between the clients; and a receiving end receives the data and decodes the data, wherein the decoded audio data are subjected to real-time audio and video rendering and then are respectively output to audio equipment and display equipment. According to the method, the WebRTC can be efficiently integrated in the Unity engine, and the stability, synchronism and safety of audio and video transmission are ensured.

Disclosed are an image decoding method, an image encoding method, and a recording medium storing a bitstream. The decoding method includes determining a search area for updating motion information of a current block; generating an SAD list including a plurality of SAD candidates for the search area; deriving change amount motion information of the current block based on the SAD candidates in the SAD list; and updating the motion information of the current block based on the change amount motion information, wherein, when a width and a height of the current block are greater than or equal to 8, the updating of the motion information of the current block based on the change amount motion information, the derivation of the change amount motion information is performed based on a comparison result between a reference SAD candidate and a predetermined threshold, the reference SAD candidate means an SAD candidate corresponding to a non-directional offset, and in a case where the reference SAD candidate is greater than or equal to the predetermined threshold, the derivation of the change amount motion information includes identifying an SAD candidate having a minimum value among the SAD candidates belonging to the SAD list; and determining the change amount motion information based on an offset corresponding to the identified SAD candidate.

The application discloses a high-parallelism storage-computation integrated polar code decoding method based on coset decoding, and belongs to the technical field of channel coding and decoding. The method comprises the following steps: constructing an error pattern group and a check constraint subgroup; performing coset division on the error pattern group to obtain a set of mutually disjoint cosets; screening a limited number of minimum cosets based on the minimum Hamming weight of the cosets; and performing parallel decoding processing on the paths corresponding to the minimum cosets in the storage-computation integrated architecture. k The number of paths in the traditional multi-bit decoding is reduced from 2 to a preset number L through coset division, and the parallel execution of the likelihood information updating and the path metric calculation is realized in combination with the calculation-type storage array, so that the decoding performance is ensured, the calculation complexity and the hardware resource consumption are significantly reduced, and the decoding throughput and the energy efficiency are improved.

The invention provides a decoding method and a decoding system based on grid ROI setting. The decoding method comprises the following steps: acquiring a row number M and a column number N corresponding to a grid ROI; acquiring a bar code image; uniformly dividing the barcode image into M * N ROIs (Region Of Interest); and performing multi-thread parallel decoding on the image information corresponding to the M * N ROIs. According to the method, the row number M and the column number N corresponding to the grid ROI are obtained, the row number M and the column number N are consistent with the row number and the column number of the materials distributed at regular positions on the production line, the bar code image is uniformly divided into the M * N ROIs, and the bar codes in the M * N ROIs are subjected to multi-thread parallel decoding, so that the ROI region is simple to draw and is not easy to make mistakes, the decoding accuracy is ensured, and the decoding efficiency is improved. And the decoding time is effectively saved, and the decoding efficiency is improved.