454 results about "Information compression" patented technology

Systems, methods, and instrumentalities are disclosed herein for pre-processing for channel state information (CSI) compression in wireless system. A WTRU may receive configuration information that indicates a reference signal and a data processing model for channel state information (CSI) compression. The WTRU may determine CSI associated with a channel using the reference signal. The WTRU may determine a channel condition associated with pre-processing. The WTRU may select a pre-processing type from a plurality of pre-processing types based on the data processing model and the determined channel condition associated with pre-processing. The WTRU may pre-process the CSI associated with the channel based on the selected pre-processing type. The WTRU may generate compressed CSI by compressing the pre-processed CSI using the data processing model for CSI compression. The WTRU may send the compressed CSI to a network.

Systems and methods are disclosed herein for spatial domain Channel State Information (CSI) compression for coherent joint transmission. In one embodiment, a method performed by a User Equipment (UE) comprises receiving information for a CSI report configuration that configures: (a) a plurality of Non-Zero Power (NZP) Channel State Information Reference Signal (CSI-RS) resources for channel measurement, wherein each of the plurality of NZP CSI-RS resources is associated with a different Transmission Configuration Indicator (TCI) state or unified TCI state; (b) a single NZP CSI-RS resource for channel measurement, wherein the single NZP CSI-RS resource comprising a plurality of sets of CSI-RS ports wherein each set of CSI-RS ports within the single NZP CSI-RS resource is associated with a different TCI state or unified TCI state; or (c) both (a) and (b).

The invention relates to a control method (2000, 4000, 5000) for a compressed air supply system (1200) to which a control device (1300) is assigned, comprising the steps: - Receipt (2100) of a supply requirement (Bv) by the control unit (1300), - Controlling (2200) a compressed air transmitter (200) by the control unit (1300) to supply compressed air (110) at a compressed air connection (1) when the control unit (1300) receives a supply requirement (Bv), - Determine (2300) a freezing risk (R I ) by the control unit (1300) depending on environmental information (I U ) and / or an operating status information (I B ) of the compressed air transmitter (200), - Actuation (2400) of a main line switching valve (25) by the control unit (1300) to block the pneumatic main line (12) in a first operating mode (B1), in the event that the risk of freezing (R I ) below a predefined limit (W), - Guiding (2500) compressed air (141) through the branch line (14, 14.1, 14.2) to the compressed air supply connection (2) in the filling direction (B) in the first operating mode (B1) to supply the compressed air consumer (300). The invention further relates to a control device for carrying out the method and a vehicle with such a control device.

The invention discloses a multi-modal multi-scale retrieval enhancement generation method, system and device applied to external knowledge questions and answers and a medium. The method comprises question perception, multi-modal multi-scale query fusion coding, dense recall and answer generation. Analyzing a key query phrase from the question through a fine-tuned instruction language model, and accurately positioning a region of interest corresponding to the phrase in an image by using an open set visual positioning model; multi-source information is compressed and distilled into an optimal query vector through a deep fusion network integrating multi-head self-attention and an information bottleneck theory; executing a maximum inner product search to recall related knowledge; guiding the large language model to synthesize all information to generate a final answer; the system, the equipment and the medium directly perform feature fusion in the vector space based on the method, so that challenges such as information loss and cascading errors caused by a traditional normal form can be effectively dealt with, high correlation and high accuracy of retrieval knowledge are ensured, and accurate and reliable image-text questions and answers are realized.

The invention discloses a regional medical information sharing method based on artificial intelligence, and relates to the field of information processing, and the method comprises the steps: obtaining regional medical information; extracting text information in the regional medical information; dividing the text information into a key text and a conventional text; text encryption of the key text is completed, and text encryption information is obtained; performing information compression encryption on the non-text information to obtain non-text encrypted information; performing global encryption on the conventional text, the text encryption information and the non-text encryption information to obtain medical encryption information; and uploading the medical encrypted information to a medical information sharing terminal. According to the method and the device, the medical information can be effectively shared in a state that the medical information is not leaked.

The invention relates to the technical field of emergency communication, in particular to a Beidou short message-based multi-mode emergency communication terminal, which comprises a Beidou communication module, a multi-source positioning unit, an information coding processor and a main control unit, the Beidou communication module is used for realizing short message communication with a Beidou satellite and supporting a BDS-3 protocol; the physical SOS button is provided with a mistaken touch prevention structure and comprises a mechanical pressing assembly with a triggering pressure threshold value larger than 2N and continuous pressing time verification logic; the multi-source positioning unit is used for acquiring position information including longitude, latitude and height; the information coding processor is configured to compress the collected help-seeking information to be within the capacity limit of a Beidou short message and generate a standard message of an emergency communication protocol; a standby power supply switching circuit is arranged in the main control unit, the main control unit is used for coordinating the work of each module and managing a power supply system, and the power supply system comprises a replaceable lithium thionyl chloride battery and a solar charging panel; and the low-power-consumption main control unit controls the Beidou communication module to transmit information to a Beidou communication satellite.

The invention relates to the technical field of information management, in particular to a safety production informatization data management method and system, and the method comprises the following steps: obtaining a synchronous monitoring record, screening field integrity and numerical value qualified information, marking and classifying abnormal items, filing qualified and abnormal data, constructing a time sequence, additionally recording and correcting abnormal contents, and storing the abnormal contents. Weight hierarchical storage information is extracted, low-frequency data is compressed, a field structure is optimized, and a management scheme is generated. According to the method, the monitoring information is preliminarily screened according to field integrity and numerical standard, the data distribution efficiency is improved, abnormal information is classified and labeled, the structure recognition precision is enhanced, missing supplementary pushing and numerical correction are realized through time sequence comparison, the data continuity is guaranteed, hierarchical storage is performed according to information weight, attributes are clearly reserved, and low-frequency data compression and integration are performed; and the storage burden is relieved, the field structure is automatically expanded and rearranged, the format uniformity and the management specification are improved, and refined processing and efficient circulation of data are realized.

The invention discloses a sparse dynamic 3D Gaussian splash method based on global-local feature extraction. According to the method, a global-local feature extraction module, an inter-frame feature flow aggregation network and depth prior are combined with a 4D pseudo pose supervision mechanism for cooperative work, and high-quality dynamic scene reconstruction and rendering are realized under the condition of a sparse camera visual angle. According to the method, the multi-view image information is compressed to the feature space for processing, so that the calculation complexity is reduced, and the geometric structure and the dynamic features of the scene are effectively reserved; the fused depth prior information and the 4D pseudo pose provide accurate geometric constraints, and the inter-frame feature flow aggregation network optimizes the time consistency of the dynamic scene. The method is particularly suitable for processing a complex dynamic scene under a sparse view angle, the sense of reality and the real-time performance of virtual view synthesis are remarkably improved, and a dynamic view with a more accurate geometric structure, higher time continuity and richer details can be generated.

The invention discloses a motion intention analysis method and device which are used for performing motion intention recognition on electroencephalogram signals. The method comprises the following steps: acquiring an electroencephalogram signal at a human brain electrode position, and performing information compression processing on the electroencephalogram signal to obtain potential state information of the electroencephalogram signal; the electroencephalogram signals are obtained based on a plurality of electrode signals; the potential state information is used for representing spatial-temporal characteristics of the electroencephalogram signals, the spatial-temporal characteristics are used for representing collection moments and collection channels of the multiple electrode signals, and each electrode signal in the multiple electrode signals corresponds to one collection channel; then, feature data of the electroencephalogram signals are extracted from the potential state information, the motion intention corresponding to the electroencephalogram signals is analyzed according to the feature data, it is not needed to judge whether cognitive distraction exists or not on the electroencephalogram signals, motion intention recognition can be directly conducted on the electroencephalogram signals, and therefore the motion intention recognition efficiency can be improved.

The invention discloses a three-dimensional point cloud geometric information compression method based on implicit neural representation, and the method comprises the steps: constructing a trunk structure of an implicit neural network through a plurality of sine representation network layers which are connected in series, and introducing a variable-scale position coding mechanism on this basis, the method enables a network to obtain higher geometric reduction precision while keeping a compression ratio, and comprises the following steps: (1) inputting space coordinates of divided voxels into a position coding module with adjustable scale parameters; (2) feeding a coding result into an implicit neural network constructed by a network layer based on sine representation, and outputting the occupancy probability of the voxel through an activation function; (3) in a training stage, the model continuously optimizes parameters, so that the output probability distribution is highly consistent with a real occupied label; (4) after model training is completed, a method of combining an AdaRound second-order quantization optimization strategy and quantization perception training is introduced, network weight is finely adjusted, and quantization errors are reduced; (5) in a reasoning stage, judging whether the voxel is occupied or not according to a preset threshold value, and when the prediction probability exceeds the threshold value, regarding the voxel as occupied; and (6) all voxels judged to be occupied are aggregated, and reconstruction of the geometric structure of the point cloud is completed.

The invention relates to a method and device for obtaining channel quality, a storage medium and a chip. The method comprises: a terminal device receiving a pilot signal sent by a network device through a downlink channel; acquiring a first channel matrix according to the pilot signal; compressing the first channel matrix according to a channel state information (CSI) compression model and channel state information (CSI) compression parameters to obtain a compressed target channel matrix; and sending the target channel matrix to the network equipment, so that the network equipment determines the channel quality of the downlink channel according to the target channel matrix. Wherein the first channel matrix is used for representing the channel quality of a downlink channel; the CSI compression model comprises a channel encoder, and the channel encoder comprises a plurality of sub-encoders; different sub-encoders correspond to different CSI compression parameters; the target channel matrix is used for indicating the network equipment to determine the channel quality of the downlink channel.

The invention discloses an image processing method, a cloud server, a cloud desktop terminal, a chip system and a medium, and the method comprises the steps: obtaining a to-be-synchronized grid associated with a change region of a current rendered image, and coding the image data of the to-be-synchronized grid according to preset coding information to obtain compressed image data; according to the grid image information of each to-be-synchronized grid, the compressed image data and the position in the current rendered image, respectively constructing each sub-image load; combining the sub-image loads and adding the coding information as image header information to obtain to-be-synchronized strip image information; and enabling the cloud desktop terminal to update a display image according to the to-be-synchronized strip image information, according to the method, only the to-be-synchronized grids needing to be synchronized are transmitted, the mode of transmitting a complete picture is replaced, and the data transmission amount is further reduced in combination with the mode of single image header information, so that the rendering speed is increased, and the image synchronization efficiency is improved.

This disclosure provides a user equipment (UE) and methods for channel state information (CSI) compression. Processing circuitry of the UE obtains a plurality of first channel matrices that each indicates CSI of a communication channel between the UE and a base station (BS) at a different time during a time period. The processing circuitry compresses each of the plurality of first channel matrices into a respective compressed channel matrix through one or more convolutional neural networks (CNNs), and extracts a feature correlation over time from the plurality of first channel matrices through one or more recurrent neural networks (RNNs). Based on the plurality of compressed channel matrices and the extracted feature correlation over time, the processing circuitry determines a feature vector that is to be sent to the BS for CSI feedback.

The invention discloses an underwater vortex optical communication anti-turbulence method based on a photoelectric hybrid deep neural network, and the method comprises the steps: carrying out the all-optical information compression at a front end through the diffraction deep neural network through the fusion of a diffraction deep neural network and a convolutional neural network, and transmitting the processed information to a rear end of the convolutional neural network, a model task is completed through training, a predicted coefficient of a Zernike polynomial is output, prediction and compensation of ocean turbulence are achieved, and the mode purity of vortex beams is improved. The front-end diffraction deep neural network is utilized to effectively improve the problem that the rear-end convolutional neural network is difficult to process massive information, the photoelectric hybrid deep neural network model is high in integration, high in operation speed and accurate in light field signal processing, the power overhead for achieving turbulence resistance through the machine learning technology is saved, operation is easy, implementation is easy, and the method is suitable for popularization and application. The method has a wide application prospect in the aspects of vortex beam underwater communication and the like.

The invention discloses an information compression sensing pushing method and system, a camera carries an area number and an equipment type to join in a unified multicast group distributed by a management server, a camera gateway router and a multicast router in a multicast tree establish multicast forwarding table items carrying the area number and the equipment type, and thus a multicast forwarding tree is formed. After the camera recognizes the pedestrian, the face features are reported to the multicast group, after the multicast router receives the face reporting message, the face reporting message is sent to the upstream interface, the same face feature vector of the same area number is only reported once, and waste of server calculation consumption and network burden is avoided. And after receiving the face report message, the management server extracts the face feature and the area number, and pushes the information corresponding to the area number to the pedestrian corresponding to the face feature, thereby realizing accurate information push based on the area number.

The invention provides a communication method. The method can be applied to equipment provided with an encoder or the encoder. The method comprises the steps that N pieces of channel information are acquired, the N pieces of channel information are determined based on reference signals on K antenna port groups, N is a positive integer, and K is a positive integer smaller than or equal to N; compressing the N pieces of channel information by using a first encoder to obtain a first sequence which is a joint feedback quantity of the N pieces of channel information; and sending the first sequence. According to the method, channel information of an analog beam is jointly compressed and fed back by using an encoder, for example, N pieces of channel information are compressed into a first sequence, so that the overhead of feeding back the channel information is reduced.

The invention provides an information compression and pruning RAG database construction method. The method comprises the implementation steps that node objects and a hierarchical clustering tree are constructed; compressing the node object text information in the clustering tree; and pruning the clustering tree after the partial node object text information is compressed based on a QR-like decomposition method. According to the method, the sentence unit text information in the sub-nodes in the clustering tree is compressed, the embedded vectors of the compressed sub-nodes are updated, and the most valuable external information can be screened and stored according to the actual requirements of the large language model, so that invalid data storage in an RAG database is reduced, and the reliability of the RAG database is improved. According to the method, text information stored in the RAG database is guaranteed to have gain all the time, meanwhile, pruning is carried out on a clustering tree obtained after partial node object text information is compressed based on a QR-like decomposition method, redundant child node objects with high semantic similarity are effectively recognized and deleted, and the storage overhead of the node objects in the RAG database is reduced.

A remote sensing image-text retrieval method based on information processing comprises the steps of preprocessing before modal input, building of an image-text retrieval model and characterization alignment, and the step of building of the image-text retrieval model comprises designing a target information enhancement module and a noise information compression module. The target information enhancement module aims at features of multiple scales to integrate detail information into visual features through the detail guide information attention module so as to enrich detail expressions of images, and the noise information compression module compresses graphic and text information so as to filter redundant information. According to the remote sensing image-text retrieval method, the problems that in an existing remote sensing image-text retrieval method, remote sensing image-text information capture is insufficient, and redundant information is excessively concerned are solved, and the remote sensing image-text retrieval precision is improved.

The invention provides a communication method and a communication device. The method comprises the following steps: a first communication device compresses N pieces of second channel information based on X pieces of first channel information to obtain M pieces of compressed information, for example, the ith compressed information is obtained by compressing Pi pieces of second channel information based on the jth piece of first channel information, and the acquisition time of the first channel information is earlier than the acquisition time of the second channel information; the first communication device sends the ith compressed information, and the first communication device also sends the jth first channel information; the second communication device restores Pi pieces of second channel information based on the i-th piece of compression information and the j-th piece of first channel information that may be received at different times. Through the mode, the channel information (namely the second channel information) can be fed back in time, and the feedback overhead can be reduced.

The invention relates to spatial information compression and delivery, and belongs to the field of information delivery. The invention provides a device for delivering event information to a nervous system after spatial compression and a method for fine regulation and control of large-scale neurons. The device comprises an imaging system for respectively coding event information into a plurality of patterns, and a plurality of optic nervous systems for regulating and controlling the activity of a single neuron in a nervous system by using a minimum information unit coded in the patterns, and the pattern projection system is used for respectively projecting the plurality of patterns coded by the imaging system to each optic nerve system. According to the device and the method, the problems that an existing neural electrode is low in information transmission efficiency, immunological rejection exists between the neural electrode and a nervous system, and nervous tissues are mechanically damaged are solved, fine regulation and control of all neurons in a large-scale neuron cluster are achieved, and high-bandwidth input of multi-dimensional event information to a whole brain area of the nervous system is achieved.

The invention discloses a federated learning image classification method based on time extension gradient cumulative average. Each client uses a local data set to train a local model to obtain gradient information of a current communication round, and uploads the gradient information to the central server after compression; the client continues to perform local training in a time period from the time when the client starts to upload the compression gradient information of the current communication round to the time when the server completes the compression gradient information aggregation of the current communication round; the central server side aggregates the compression gradient information to obtain average compression gradient information of the current communication round, and issues the average compression gradient information to the client side to update a local model of the (t + beta) th communication round; and repeating the steps until the training is completed, and inputting the to-be-detected image into the local model to obtain an image classification result. The method shows excellent performance under various data sets and different compression levels, has higher convergence speed and model precision, and solves the problem of low communication efficiency caused by network delay and bandwidth limitation in image classification of federated learning.

The invention discloses an efficient image analysis and question answering system for intelligent customer service, and aims to solve the commercialized problems of slow response and high cost when an existing AI customer service processes user picture questions. In order to improve customer service interaction experience, the system greatly reduces the amount of visual data input into a core language agent through an innovative visual information compression engine, for example, the amount of the visual data is compressed to one fourth of the original amount. By means of the key optimization, the intelligent customer service staff can respond to picture inquiry of customers in real time, and customer loss caused by waiting is avoided. Meanwhile, the system ensures high-precision identification of commodity details through a unique training strategy, provides reliable and accurate answers for customers, and successfully solves the problem that speed and precision are difficult to consider in customer service application in the industry. Finally, a practical solution with instant response and high reliability is provided for the fields such as e-commerce intelligent customer service and the like needing real-time interaction, and commercialized popularization of the advanced AI technology is powerfully promoted.

The present application relates to a log processing method, a control apparatus, and a computer storage medium. The method comprises: acquiring a plurality of abnormal log files, file names of at least a part of the abnormal log files comprising identifiers compressed on the basis of log information; on the basis of a preset condition, performing a deduplication operation on the abnormal log files to obtain deduplicated abnormal log files and a data set table, the data set table comprising statistical data in the deduplication operation; and, on the basis of the data set table and a preset rule, sending the deduplicated abnormal log files to a specified backend for processing, the specified backend being specified on the basis of the preset rule. In the implementation of the technical solution of the present application, an efficient and automated log deduplication mechanism is provided. Such a method significantly increases the speed and accuracy of log processing and reduces the waste of resources caused by duplicate or highly similar logs.

The invention discloses a hierarchical network topology estimation method based on an information bottleneck theory, and the method comprises the steps: carrying out the merging of end-to-end traffic time sequences, introducing comparative learning, generating random perturbation for the merged end-to-end traffic time sequences in a time domain and a frequency domain, and carrying out the estimation of the hierarchical network topology. The end-to-end traffic time sequence is comprehensively represented from the perspective of time-frequency consistency analysis through comparative learning, end-to-end traffic time sequence embedding is obtained, and finally, the length of a shared path is measured by adopting the Euclidean distance embedded by the end-to-end traffic time sequence, and the network topology is recovered from bottom to top. According to the method, an information bottleneck theory is adopted, information related to topology estimation is reserved, noise is compressed, end-to-end flow time sequences are combined based on confidence, time sequence data are effectively represented from a time domain and a frequency domain at the same time by adopting comparative learning, features capable of reflecting variation difference and consistency between different time sequence data are extracted, and the time sequence data are effectively represented. And the length of the shared path is better measured, so that the network topology structure is more accurately recovered.

The invention discloses a system security access control method based on deep learning and dynamic risk awareness, which comprises the following steps of: generating a behavior state vector by collecting user identity, operation, equipment state and environment information, constructing an access behavior map, and establishing an access behavior map through a Hebbian learning rule; the method comprises the following steps: dynamically updating connection strength between behaviors in combination with historical behavior data, after receiving an access request, selecting an association path from a map, calculating an activation potential value, selecting an optimal path as an access decision basis by applying a Winner-Take-All selection mechanism, executing an access control operation, generating a risk mark and a behavior evolution mark, and establishing an access control strategy; the method is used for adjusting a selection threshold and an atlas weight. And finally, compressing and coding the related information, and recording the information in an access log. According to the method, intelligent judgment on the access behavior is realized by using deep learning and a dynamic risk perception mechanism.

This disclosure provides a user equipment (UE) and methods for channel state information (CSI) compression. Processing circuitry of the UE obtain a plurality of first channel matrices that each indicates CSI of a communication channel between the UE and a respective one of multiple transmission-reception-points (TRPs). The processing circuitry compresses each of the plurality of first channel matrices into a respective feature vector through one or more convolutional neural networks (CNNs), and concatenates the plurality of feature vectors into a joint feature vector. The processing circuitry compresses the joint feature vector into a compressed joint feature vector through one or more fully connected neural networks (FCNNs).

A method, implemented in a wireless transmit and receive unit (WTRU), may include receiving one or more reference signals, selecting a first codebook-based precoding matrix based on the one or more reference signals and determining a second precoding matrix based on the one or more reference signals. The method may include determining a differential precoding matrix based on the first codebook-based precoding matrix and the second precoding matrix. The method may include determining a compressed differential precoding matrix based on the differential precoding matrix. The method may include transmitting feedback information indicating the compressed differential precoding matrix.

A monitoring system can generate compressed storage system monitoring data segments using monitoring data obtained from a storage system. The monitoring system can obtain storage system monitoring data and generate a segment of the storage system monitoring data. The monitoring system can determine time information, a compression technique, and a compression parameter set for the segment. In response to the user query, the monitoring system can identify the compression technique and the compression parameter set based on a time interval of the query and the time information and perform at least one of: reconstructing and providing at least a portion of the segment using the time information, the compression technique, and the compression parameter set; or providing the compression parameter set and the indication of the compression technique for reconstruction of the portion.

Provided in the embodiments of the present application are an information compression method and apparatus, and a terminal device and a network device. The method comprises: according to the priority of first channel state information (CSI) in CSI that is born on an uplink channel, a terminal device determining that it is necessary to perform compression on the first CSI, wherein the first CSI is AI / ML-based CSI, and the uplink channel is used for bearing the first CSI; and the terminal device performing compression on the first CSI by means of adjusting a first model, wherein the first model is used for generating the first CSI.

The invention provides a multi-modal large-model long-sequence information compression retrieval method, and relates to the technical field of data processing, and the method comprises the following steps: receiving multi-modal long-sequence original data of a target patient; all modal data in the multi-modal long sequence original data are uniformly mapped to the same hidden space through a pre-trained multi-modal large model, and a unified semantic representation sequence of semantic association between modals is obtained; performing time sequence feature analysis on the unified semantic representation sequence, regarding feature points in the semantic representation as a spatial point set, determining boundary distribution by constructing convex hulls of feature clusters, identifying key feature clusters in the semantic representation, and determining a key time sequence interval based on a convex hull distribution mode of the key feature clusters. According to the method, the problems of weak cross-modal semantic association, easy loss of key time sequence information and low compression and retrieval efficiency in the existing multi-modal long sequence data retrieval are solved.