1857 results about "Codebook" patented technology

A method is provided for mapping the latent space of a Vector Quantized Variational AutoEncoder (VQ-VAE) to polynomial basis vectors. The method includes training a VQ-VAE model on a dataset to obtain a set of codebook vectors representing the latent space; defining a polynomial basis for the latent space, the polynomial basis containing terms up to a predetermined order; mapping each codebook vector to the polynomial basis by determining polynomial coefficients that represent each codebook vector in terms of the polynomial basis; and using the polynomial coefficients to reconstruct and manipulate latent space representations.

A system and method for real-time time series forecasting using a compound large codeword model with integrated supervisory neurons. The system processes diverse inputs through adaptive codebook generation and codeword allocation. A projection network fuses different data types, creating unified representations for a latent transformer-based machine learning core. The core contains local neural network regions of interconnected operational neurons, monitored by supervisory neurons. These supervisory neurons receive activation data from operational neurons, perform real-time statistical analysis, determine necessary structural modifications, and initiate their implementation during operation. This architecture enables efficient handling of multi-modal data, capturing complex relationships between different input types. The combination of adaptive codebook generation and the supervisory neuron system ensures responsiveness to evolving data patterns and task requirements. This approach provides more accurate and timely forecasts by leveraging diverse data types in a sophisticated, integrated manner, while continuously adapting its structure to maintain optimal performance.

The invention discloses a dynamic beam adaptive optimization method for the Internet of Vehicles, and the method comprises the steps: firstly obtaining the current position, speed and driving direction of a vehicle, and carrying out the prediction of a future trajectory through combining historical trajectory information and employing a Kalman filtering technology and the like; and selecting a candidate beam direction from the predefined beam codebook. The system receives channel quality feedback information of a communication link in real time, and dynamically updates a beam forming weight based on an improved recursive least square algorithm. According to the algorithm, a composite adaptive adjustment mechanism of an error size and an error change trend is introduced, a multi-objective optimization function is constructed, and performance indexes such as link quality maximization, interference power minimization and beam switching frequency minimization are effectively balanced in a dynamic environment, so that an optimal beam direction is obtained. Finally, the real-time switching of the antenna array beams is realized through a low-overhead and low-delay control protocol. According to the method, track perception and a self-adaptive feedback mechanism are fused, and good real-time performance, stability and expandability are achieved.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a visual token generation method, device, equipment and medium based on a shared index, and the method comprises the steps: obtaining an input image, and extracting semantic features and pixel features through a semantic encoder and a pixel encoder; calculating the distance between each feature and a codebook thereof, and carrying out weighted summation to determine a shared index; retrieving quantitative features from the codebook using a shared index; respectively generating a reconstructed image and a reconstructed semantic feature by using a pixel decoder and a semantic decoder; jointly optimizing an encoder, a codebook and a decoder based on a reconstruction result; and generating a unified visual token sequence for the target task image by using the optimized component. Through double-flow feature extraction, shared mapping quantization and joint loss optimization, global semantic information and local pixel details can be reserved in the visual token at the same time, so that the model has accurate understanding ability, high-fidelity images can be generated, and the performance of understanding and generating tasks is improved.

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive, from a network node, an indication of an uplink resource associated with reporting channel state information (CSI). The UE may transmit, to the network node and using the uplink resource, uplink control information (UCI) including a CSI report that indicates precoding matrix indicator (PMI) values, based at least in part on prioritizing groups of information associated with the CSI report, at least one group, of the groups, being associated with time domain basis index values and non-zero coefficients of a coefficient matrix of a codebook associated with the CSI report. Numerous other aspects are provided.

A terminal according to one aspect of the present disclosure includes: a receiving section that receives, when the terminal includes eight antenna ports, downlink control information (DCI) including a plurality of fields of at least one of a precoding information and number of layers field and a transmission precoding matrix indicator (TPMI) field for codebook-based physical uplink shared channel (PUSCH) transmission; and a control section that controls uplink transmission, based on the DCI. According to one aspect of the present disclosure, UL transmission using more than four antenna ports can be appropriately controlled.

Methods and apparatuses for PT-RS enhancement are disclosed. In one embodiment, a UE comprises a transceiver; and a processor coupled to the transceiver, wherein the processor is configured to receive, via the transceiver, a configuration to transmit 2 or 4 PT-RS ports when 4 antenna ports codebook based PUSCH transmission is configured; and determine the mapping between PUSCH antenna ports and the PT-RS ports.

The invention provides a training method and device for a robot control model, a control method and device and electronic equipment, and relates to the technical field of equipment control, and the method comprises the steps: obtaining a preset motion data set, and employing a preset large language model in the robot control model to obtain a preset motion data set; generating an action sequence text and a prediction target direction corresponding to the sample natural language command based on the sample natural language command; and encoding the action sequence text by adopting an initial text encoder in the robot control model, determining a target action code corresponding to the action code from a preset action codebook, and performing action planning according to the target action code and a predicted target direction by adopting an initial action planning model in the robot control model. Generating action strategy information corresponding to the sample natural language command; and parameter adjustment training is carried out on the initial motion planning model in the robot control model. The generalization ability of the robot control model is improved, and control over the robot is more accurate.

The invention discloses a directional broadcast multi-band interference suppression method based on dynamic beam forming, and relates to the technical field of computer processing, and the method comprises the steps: S01, synchronously capturing electromagnetic signals of at least two working bands through a heterogeneous band receiving array synchronized by a hardware-level clock; s02, analyzing space parameters of a target signal and an interference source in the electromagnetic signal in real time; s03, constructing a layered adaptive codebook based on the interference dynamic parameters; s04, fusing the target signal parameter and the interference dynamic parameter by adopting a block matrix acceleration algorithm and an orthogonal decomposition technology, and calculating a beam forming weight; s05, executing operation through the multi-band coprocessor; and S06, establishing an interference source azimuth drift prediction model, and setting a trigger threshold as the coverage tolerance angle theta. According to the method, the anti-interference real-time performance is improved by utilizing layered adaptive wave beams, frequency band isolation, super-resolution multi-path identification and block decomposition.

A terminal according to an aspect of the present disclosure includes: a reception section that receives downlink control information (DCI) including a priority indicator field for scheduling a physical downlink shared channel (PDSCH); and a control section that controls transmission of hybrid automatic repeat request acknowledgement (HARQ-ACK) information for the PDSCH on the basis of the DCI, in which the DCI includes a field for at least one of one-shot HARQ-ACK feedback for the PDSCH or an enhanced dynamic HARQ-ACK codebook for the PDSCH. According to one aspect of the present disclosure, radio communication in an NR-Unlicensed (U) system can be appropriately controlled.

A correlation-aware adaptive codebook compaction system for multi-modal data compression that preserves cross-modal relationships while providing enhanced reconstruction quality. The system analyzes temporal and spatial relationships between different data modalities to generate correlation maps that guide compression decisions. A virtual management layer performs stream characterization and adaptive routing, while a processing pipeline implements primary codebook compression with mismatch handling for novel data blocks. High-entropy data segments receive pre-compression processing before codebook compression. Sequential registration data is processed through matrix factorization and dedicated matrix codebooks. The system continuously monitors data distribution characteristics and automatically retrains codebooks when drift thresholds are exceeded. A neural upsampling subsystem uses correlation information to guide cross-modal enhancement processes through modality-specific networks and attention mechanisms. The unified output includes compressed data streams, correlation maps, synchronization metadata, neural model parameters, and updated codebooks, enabling synchronized reconstruction with preserved cross-modal relationships and enhanced quality through correlation-guided neural upsampling.

The invention provides a multi-granularity open vocabulary query method based on an object-level lossless Gaussian field, and the method comprises the steps: introducing an object-level Gaussian field with a global consistency codebook, rendering a learnable semantic tag vector in the Gaussian field back to a corresponding object tag; the direct mapping between the label and the corresponding uncompressed high-dimensional feature is established through the code book, so that the semantic feature of any dimension is supported, additional compression is not needed, and the understanding capability on an object is remarkably improved; according to the method, wide quantitative and qualitative evaluation is carried out in a plurality of scenes, excellent performance in the aspects of object level zero sample segmentation and open vocabulary understanding is shown, the highest precision is particularly achieved in object-component hierarchical retrieval, and meanwhile multi-granularity scene editing is supported.

A system and method for deep learning using a large codeword model with hierarchical caching is disclosed. The system processes input prompts into tokens, maps them to codewords using a codebook, and processes these through a machine learning core to generate responses. A sophisticated caching architecture stores and retrieves responses across both local and global cache tiers. The local cache maintains frequently accessed responses on edge devices through short-term and persistent storage components, while the global cache enables knowledge sharing across multiple devices. A context aggregator identifies relationships between cached responses to form comprehensive contextual representations. This hierarchical caching system significantly reduces computational requirements by reusing previously generated responses for similar prompts, while continuously optimizing cache contents based on relevance scoring and usage patterns. The approach enables efficient scaling across distributed environments while maintaining response quality.

A communication device for providing a channel state information, CSI, feedback in a wireless communication system includes a transceiver to receive, from a transmitter a radio signal via a time-variant, frequency-selective MIMO channel, the radio signal including downlink reference signals according to a reference signal configuration including a number of antenna ports, and downlink signals including the reference signal configuration; and a processor. The processor estimates an explicit CSI in the frequency domain using measurements on the downlink reference signals on the radio channel, selects a Doppler-delay precoder matrix (W) for a composite Doppler-delay-beam three-stage precoder, calculates either one or more of a channel quality indicator, CQI, and / or a precoder matrix indicator, PMI, and / or a rank indicator, RI, and reports to the transmitter the CSI feedback including either one or more of the CQI, and / or the PMI and / or the RI.

An adaptive random access system and method with learned query optimization for compacted data files that enhances random access performance through machine learning and pattern recognition. The system incorporates a query pattern learning module that analyzes historical access patterns and user behavior to build statistical models of data usage. An adaptive estimator module improves location estimation accuracy by incorporating learned patterns rather than relying solely on mathematical calculations. A predictive boundary detector uses learned codeword patterns to more accurately identify boundaries in compacted data, reducing misalignment errors. An intelligent search engine coordinates optimization strategies including context-aware search string parsing and encoding strategy selection based on learned performance data. A dynamic codebook optimizer reorganizes sourceblock layout based on access frequencies and co-occurrence patterns to improve retrieval speed. An enhanced search cache implements predictive caching algorithms that anticipate user queries and proactively load relevant data.

A computer system for compacting video data. The system acquires a video stream, reduces redundancy through pre-processing, and analyzes the stream to identify patterns and irregularities. It detects spatial or temporal anomalies in the video and produces three outputs: a conditioned video stream based on statistical analysis, an error stream reflecting adjustments made during conditioning, and an anomaly meta-stream containing metadata about detected anomalies. The system communicates with one or more remote systems to synchronize and negotiate a compatible compression codebook, optionally exchanging compact updates that represent differences between local and remote codebooks. The conditioned video stream is then compressed using the agreed codebook. The system outputs a compacted representation of the video that includes the compressed stream, the error stream, and the anomaly metadata, supporting efficient storage or transmission while maintaining the ability to detect, trace, and reconstruct key information within the video.

A federated system and method for data compression optimization in distributed device networks. The system comprises multiple edge devices that analyze local data patterns to generate device characteristic profiles while performing local compression optimization and maintaining data privacy. Edge devices contribute to collaborative learning by generating privacy-preserved updates without transmitting raw data. A central coordination system aggregates encrypted contributions using secure multi-party computation protocols, identifies device groups based on data pattern similarities, and generates optimized compression parameters for each group. The system coordinates collaborative training of data reconstruction models across device groups and deploys group-optimized reconstruction capabilities. Device grouping is performed by calculating similarity scores between device characteristic profiles and clustering devices with scores above predetermined thresholds. The system dynamically adapts compression and reconstruction parameters through federated learning while preserving individual device data privacy, enabling efficient data compression and near-lossless recovery across heterogeneous Internet-of-Things networks.

Systems and methods for detecting and diagnosing machine faults are discussed. An exemplary system includes at least one sensor node to sense a signal indicative of an operation status of a machine part, and a machine health analyzer circuit to generate a computational machine fault model comprising an autoencoder (AE) network and an associative module. The AE network encodes the sensed signal into signal features in a latent feature space, and decodes the signal features to produce a reconstructed signal. The associative module transforms the encoded signal features into an associative output using a dynamically updatable codebook. The machine health analyzer circuit detects a presence or absence of fault in the machine part based on reconstruction losses determined respectively from the reconstructed signal and the associative output. The detected fault can be presented to a user or to a process such as fault diagnosis or fault correction.

Devices and methods for providing channel state information reporting for multi-transmission-reception-point coherent joint transmission in a wireless communication system. Channel state information configuration information is provided by a base station to a wireless device. The channel state information configuration information indicates one or more restrictions for the UE in reporting a precoding matrix indicator (PMI) for each of multiple transmission reception points. The wireless device provides a PMI to the base station based on the channel state information configuration information.

Provided in the present application are a method for determining a feedback information transmission location and a device, used when one piece of DCI simultaneously indicates resource release with respect to SPS PDSCHs on a plurality of carrier waves, to implement determining a location of an HARQ-ACK for the DCI in an HARQ-ACK codebook. A method of the present application comprises: Determining that a DCI indicates performing resource release with respect to SPS PDSCHs on a plurality of carrier waves; determining a target location of an HARQ-ACK corresponding to the DCI in an HARQ-ACK codebook according to an SLIV for a first SPS PDSCH among released SPS PDSCHs, wherein the first SPS PDSCH is an SPS PDSCH on a target carrier wave among the plurality of carrier waves. Normal feedback of the HARQ-ACK for the DCI is ensured, and uniform understanding for HARQ-ACK transmission by a network-side device and a terminal ensures normal transmission of the HARQ-ACK.

Modality agnostic Large Codeword Model (“LCM”) is an advanced deep learning architecture that processes discrete, compressed data representations called codewords across multiple modalities. Unlike traditional models using raw tokens and dense embeddings, LCMs efficiently handle diverse input types including text, images, audio, and video. The system employs a modality agnostic encoder, unified codebook, and multimodal machine learning core to capture inherent data structures and patterns. This approach enables more generalizable and interpretable feature learning, facilitating transfer learning across domains. The LCM's scalable and flexible architecture includes components for modality-specific processing, cross-modal attention, and joint representation learning. With its computational efficiency and versatility, the Modality Agnostic LCM offers significant potential for various AI applications, including natural language processing, computer vision, and multimodal reasoning.

Provided is a method performed by a user equipment (UE). The method comprises: receiving, from a base station (BS), one or more messages comprising precoding information, wherein the precoding information is used by the UE to determine a precoder used in a codebook-based Physical Uplink Shared Channel (PUSCH) transmission; performing, by the UE, the codebook-based PUSCH transmission using eight antenna ports based on the precoder.

A method and apparatus generate a 2-dimensional (2D) image. A method for generating a 2-dimensional (2D) image includes obtaining a time index and a view. The method further includes obtaining first coding indices and a first codebook for canonical 3D Gaussians, wherein the canonical 3D Gaussians are 3D Gaussians corresponding to a reference time index, and represent a 3D space corresponding to the reference time index. The method also includes obtaining second encoding indices and a second codebook for a parameter offset, wherein the parameter offset indicates a difference between the canonical 3D Gaussians and 3D Gaussians for the time index. The method further includes reconstructing the canonical 3D Gaussians based on the first coding indices and the first codebook. The method also includes reconstructing parameter offsets of the 3D Gaussians for the time index based on the second coding indices and the second codebook. The method further includes adding the reconstructed canonical 3D Gaussians and the reconstructed parameter offset to reconstruct the 3D Gaussians for the time index. The method also includes generating a second image for the view based on the reconstructed 3D Gaussians.

The invention relates to the technical field of wireless semantic communication, and discloses a semantic information transmission method, device and equipment and a storage medium, a sending end extracts features of to-be-transmitted data, and matches the features with vectors in a preset discrete vector codebook to obtain discrete feature vectors; and determining an orthogonal frequency division multiplexing time-frequency resource distribution grid according to the feature importance weights of the discrete feature vector and the initial feature vector, generating a sending signal, and transmitting the sending signal to a receiving end. And the receiving end inputs the semantic feature vector obtained after processing the received signal into the vector quantization variational self-decoder, and the semantic feature vector is matched with the vector in the discrete vector codebook again to correct the transmission error, thereby obtaining the reconstructed data to be transmitted. Discretization processing of the features is achieved through a discrete vector codebook, important semantic feature bits are quantized according to the importance weight of each feature and then distributed to orthogonal frequency division multiplexing for transmission, so that key semantic information is protected, and anti-interference transmission of the semantic information is achieved.

An all-binary neural network system and method for processing and analyzing multi-source time series data is disclosed. The system employs a shared codebook to encode input streams into binary codewords, which are then processed through a series of binary convolutional layers, binary LSTM layers, and binary fully connected layers. The system maintains binary representations throughout, enabling efficient computation and reduced memory requirements while effectively capturing temporal and inter-source relationships in the data.

A method for acknowledgement information enhancement includes receiving first information for first cells; second information for second cells; first and second DCI formats; and N PDSCHs on N cells, from the second cells, which are scheduled by a DCI format from the second DCI formats. The method further includes determining first HARQ-ACK information bits associated with the first DCI formats and second HARQ-ACK information bits associated with the second DCI formats. The second HARQ-ACK information bits include, for the DCI format, M=Σi∈N Mi HARQ-ACK information bits corresponding to receptions of the N PDSCHs on N cells, in ascending order of cell indexes for the N cells and (Mmax−M) HARQ-ACK information bits with NACK values that are appended to the M HARQ-ACK information bits. The method further transmitting a channel that includes a HARQ-ACK codebook generated by appending the second HARQ-ACK information bits to the first HARQ-ACK information bits.

A system and method for handling acknowledgments in a wireless system. In some embodiments, the method includes, receiving, by a User Equipment (UE), one or more first transmissions; decoding, by the UE, the first transmissions, to generate a first sub-codebook including one or more corresponding Hybrid Automatic Repeat Request Acknowledgement (HARQ ACK) bit strings each having a first length; receiving, by the UE, one or more second transmissions; decoding, by the UE, the second transmissions, to generate a second sub-codebook including one or more corresponding HARQ ACK bit strings each having a second length, greater than the first length; receiving, by the UE, one or more third transmissions; and decoding, by the UE, the third transmissions, to generate a third sub-codebook including one or more corresponding HARQ ACK bit strings each having a third length, greater than the second length.

A system and method for data compaction with adaptive codebook statistical estimates. Training data sets determine sourceblock frequencies while a dynamic mismatch probability system continuously refines estimates based on observed patterns. Context-aware handling selects appropriate secondary encoding methods for different data types (text, binary, image, executable). Machine learning models predict optimal mismatch probabilities from extracted features. Edge-optimized training enables codebook development on resource-constrained devices with intelligent resource management. Differential updates transmit only changes between codebook versions, minimizing bandwidth usage. Federated learning enables multiple devices to contribute to shared codebooks while maintaining data privacy. A secure synchronization protocol with authentication and verification ensures codebook consistency. The distributed maintenance method provides systematic validation, conflict resolution, and optimization across device networks, enabling efficient encoding across heterogeneous systems.

Methods, systems, and devices for wireless communications are described. Feedback based codebooks for reducing or eliminating interference between orbital angular momentum (OAM) based backhaul communications between a first network node and a second network node and access communications between the first network node and a user equipment (UE) are described. The first network node may transmit a message to a UE indicating a feedback codebook for a set of reference signals associated with OAM modes. The first network node may transmit the reference signals, where each reference signals may be associated with a respective OAM mode. The UE may generate and indicate in a feedback report to the first network node codewords corresponding to UE feedback for the received reference signals. The first network node may communicate with the UE via an access link and with the second network node via a backhaul link based on the feedback report.

A system and method for real-time time series forecasting using a compound large codeword model with integrated supervisory neurons. The system processes diverse inputs through adaptive codebook generation and codeword allocation. A projection network fuses different data types for a latent transformer-based machine learning core. A hierarchical supervisory network, comprising low-level, mid-level, and high-level nodes, monitors local neural network regions, performing real-time statistical analysis and implementing structural modifications. The system efficiently handles multi-modal data, capturing complex relationships between input types. An adaptive codebook generation method, coupled with the supervisory architecture, ensures responsiveness to evolving data patterns and task requirements. This approach provides accurate and timely forecasts by leveraging diverse data types in a sophisticated, integrated manner, while continuously adapting its structure during operation to maintain optimal performance.