403 results about "Stream" patented technology

A distributed system and method uses a virtual management layer to classify incoming heterogeneous data streams by type, attach association flags to related streams, and route each stream to a selected compression or decompression subsystem executing on available computing devices. The virtual layer allocates tasks using live resource conditions to enable parallel processing across devices and per-type optimization (e.g., statistical, codebook, or neural techniques). Outputs returned from disparate devices are collected and deterministically regrouped using the association flags so related streams are preserved regardless of where processing occurred. The architecture supports both compression and decompression workflows, load-balanced scheduling, and scalable operation across multiple devices, improving throughput while maintaining relationships among multi-type data.

The present application relates to the technical field of satellite communication, in particular to a CSI feedback method based on MIMO low-orbit satellite communication, an electronic device and a medium.The CSI feedback method comprises: at the user end: extracting features of a channel matrix as downlink CSI by a BlockCsiNet encoder and compressing the channel matrix into a code word; the BlockCsiNet encoder comprises a first convolutional layer with a convolution kernel of 7*7 and 5*5; in the feature extraction process, the 7*7 and 5*5 convolution kernels of the first convolutional layer present a 11*11 receptive field, so as to extract the encoding features of the downlink CSI from the block sparse characteristics of the channel matrix of the downlink CSI; quantizing the code word into a bit stream; and returning the bit stream to a base station side BlockCsiNet decoder through a feedback link.The present application is beneficial to extracting the features of the block sparse channel matrix of low-orbit satellite communication, and effectively improves the reconstruction performance and reconstruction accuracy of the downlink CSI.

This invention discloses an elevator bearing safety early warning method, device, and medium based on the Internet of Things (IoT), relating to the field of control testing and monitoring technology. The method includes: acquiring raw analog signals and preprocessing them to generate synchronous data frames; decomposing fault impact sequences from the synchronous data frames and obtaining a bearing safety status index through weighted D-S evidence, while simultaneously matching a pre-diagnostic control logic library to generate a pre-diagnostic confidence identifier; based on the pre-diagnostic confidence identifier and the bearing safety status index, obtaining a dynamic health baseline through sliding retrieval and elastic tracking compensation, and dynamically extrapolating the control envelope through a stochastic control boundary evolution algorithm to generate a dynamic safety threshold; comparing the bearing safety status index with the dynamic safety threshold, evaluating link quality, and selecting a candidate route set; and obtaining the optimal route through a shortest delay tree to form an encrypted alarm stream. This invention achieves accurate generation of the dynamic safety threshold.

The present disclosure provides a transmission method, device and storage medium for VR live video, relating to the technical field of communication, wherein the method comprises: performing encoding processing on video data transmitted by a VR live source to generate fragmented encoding data; performing encoding processing on the fragmented encoding data to generate panoramic encoding data and main perspective encoding data; performing encapsulation processing on the panoramic encoding data, the main perspective encoding data, unicast fragmented data index information and audio to generate a multicast stream and add a PMT table; transmitting the multicast stream in a multicast group, and when receiving a unicast switching request transmitted by a terminal in the multicast group based on the PMT table, transmitting unicast fragmented data corresponding to the unicast switching request to the terminal. The method, device and medium of the present disclosure can be combined for transmission by using multicast and unicast, save network bandwidth, take advantage of operator multicast bearing, reduce CDN bearing cost, be compatible with multiple types of terminals, and improve user experience.

A communication method and apparatus, which relate to the technical field of communications. In the method, a first communication apparatus can acquire a first precoding matrix, wherein the first precoding matrix comprises a first sub-matrix corresponding to a first spatial stream and a second sub-matrix corresponding to a second spatial stream, the vector, which corresponds to a first time-frequency unit in the first sub-matrix, being a first vector, and the vector, which corresponds to a first time-frequency unit in the second sub-matrix, being a second vector; the first communication apparatus calibrates the first precoding matrix to obtain a second precoding matrix, wherein the second precoding matrix comprises a third sub-matrix corresponding to the first spatial stream, the vector, which corresponds to a first time-frequency unit in the third sub-matrix, being a second vector; and the first communication apparatus sends first information, wherein the first information is obtained by means of compressing the second precoding matrix. The method can improve the compression efficiency of channel information (such as a precoding matrix) and improve the feedback performance of the channel information.

Analytics equipment (14) for a communication network (10) stores a stream (18) of data records (20) from the communication network (10). The analytics equipment (14) samples the stored stream (18T) to obtain a sampled stream (18S) that includes fewer data records (20) than the stored stream (18T). The analytics equipment (14) explores the sampled stream (18S) to identify characteristics (28) of data records (20) to be used for insight creation. Based on the identified characteristics (28), the analytics equipment (14) filters the stored stream (18T) to obtain a filtered stream (18E) that includes data records (20) with the identified characteristics (28). The analytics equipment (14) then creates one or more insights (16) about the communication network (10) using the filtered stream (18E).

This invention discloses a bypass transmission architecture for encrypted network interface cards (NICs) based on parallel feature extraction. The architecture includes: a packet stream input interface, a control path unit, a data path unit, a decision window merging unit, a fragmentation-aware incremental checksum correction engine unit, and a security control unit. This invention can reduce forwarding latency, improve transmission determinism and resource utilization, and enhance packet forwarding efficiency in high-speed network scenarios. By using a deterministic delay storage module that matches worst-case latency, timing synchronization between control flow and data flow is achieved, reducing the risk of timing races and rule misjudgments. Simultaneously, the fragmentation-aware incremental checksum correction engine reduces the risks of insufficient fragmentation packet compatibility, checksum miscalculations, and resulting packet loss in traditional solutions.

The application relates to a factory data management method, system, terminal and storage medium, and relates to the field of data management.The method comprises the following steps: receiving an original data stream; performing timestamp alignment processing on the original data stream to obtain a standardized data sequence; extracting a production batch identifier and a process identifier in the standardized data sequence; generating a process sorting sequence corresponding to the production batch identifier based on the process identifier and a process sequence; grouping data items belonging to the same production batch identifier in the standardized data sequence according to the process sequence, and outputting process-level data records corresponding to the production batch identifier; extracting process parameters having a transmission relationship between adjacent processes, and calculating deviation values between the process parameters; when the deviation value exceeds a deviation threshold value, determining a parameter correction amount from a rule library according to a process position and a parameter deviation direction, and correcting the process parameters of an abnormal process node. The application has the effects of improving production operation efficiency and guaranteeing product quality.

Approaches presented herein relate to the distribution and processing of stream data in a networked computing environment. A set of streaming data sources can be identified for a given cluster of nodes and the data from those data sources can be tracked, managed, and / or routed to a number of different microservices. Stream distribution and routing (SDR) agents may be used to manage a routing table that includes an identifier for each of the data sources. The individual agents may distribute information about the sensors to individual microservices, which may determine which data streams are relevant and then initiate individual processes and functions using the relevant data streams. Output results from the individual microservices may also be assigned an identifier, which may further be used by the individual agents to route outputs toward their particular microservices, which may then use the outputs from other microservices within their own workstreams.

The embodiment of the application discloses a data conversion method, system and device, the method comprises the following steps: when a trigger condition corresponding to a scheduling task is met, triggering the scheduling task, and transmitting task parameters to a push task main program; through the push task main program, target data in a database is queried based on the task parameters in a stream mode, the target data is parsed according to a template file, and the target data is written into a landing shard data file; after the target data in the database is queried and written into the landing shard data file, the total number of the target data under each data mode is determined, a text file corresponding to each data mode is generated according to the total number of the target data under each data mode and each landing shard data file under each data mode; and the text file corresponding to each data mode is compressed, and the compressed text file is uploaded to a target server. The method can ensure the timeliness and accuracy of data pushing, and realizes decoupling of the database and a data consumer.

A queue application programming interface (QAPI) in a first data center receives incident messages, each with an incident identifier. The first data center is in communication with additional data centers. The messages are either ordered or unordered. Ordered messages must be processed in the order received, while unordered messages do not. The QAPI transmits the messages to a message stream data structure. A first worker sub-system retrieves unordered messages from the message stream for processing. In parallel, the QAPI transmits the messages to a group stream data structure which arranges the messages into groups. A second worker sub-system retrieves groups of ordered messages for processing. At least one of the first or second worker sub-systems is associated with the additional data centers.

The present invention relates to an apparatus for analyzing peak data generated from a mass signal acquired by a mass spectrometry system, comprising: a streaming device including a first network interface and configured to (a) generate peak data from the acquired mass signal or receive peak data generated externally from the acquired mass signal, (b) group the peak data into a plurality of independently processable data packets, the data packets relating to a processing task, and (c) distribute the data packets as a task-specific stream onto a network via the first network interface; and at least one analysis device including a second network interface and configured to (a) retrieve data packets of a task-specific stream from the network via the second network interface, (b) perform a processing task on the retrieved data packets to generate result data, (c) encapsulate the result data in a result data packet, and (d) distribute the result data packet onto the network via the first network interface.

A method includes recording a first video stream characterized by a first value of a first quality characteristic. The method includes determining that the first video stream satisfies a trigger criterion. The trigger criterion characterizes a threshold amount of video content change information. The method includes, in response to determining that the first video stream satisfies the trigger criterion, obtaining a second video stream characterized by a second value of a second quality characteristic. The second video stream includes scene information also included in the first video stream. The second value of the second quality characteristic is indicative of a higher quality video stream than the first value of the first quality characteristic. The method includes generating a third video stream by adding information from the second video stream to the first video stream. The third video stream corresponds to a higher quality version of the first video stream.

The present application discloses an encoding method, a decoding method, a code stream, an encoder, a decoder, and a storage medium. The method comprises: determining one or more vector parameters of a current block; determining at least one second reference region on the basis of a first reference region indicated by the one or more vector parameters, wherein the second reference region is obtained by merging first reference regions indicated by at least some vector parameters among the one or more vector parameters; determining one or more prediction modes of the current block on the basis of the at least one second reference region, wherein the one or more prediction modes are determined on the basis of gradient information derived from the at least one second reference region; and determining a prediction sample of the current block on the basis of the one or more prediction modes.

This application provides an LDPC multi-level coded transmission method, apparatus, and electronic device, introducing a multi-level coded modulation scheme to achieve inter-stream and inter-bit code rate allocation based on channel quality differences, and implementing joint iterative optimization of detection and decoding at the receiver. At the transmitter, multi-level coded modulation is performed according to a pre-constructed parameter table, which allocates code rates at the sub-stream level and between bits for single-stream or multi-stream data. High-reliability streams / bits use higher code rates, while low-reliability streams / bits use lower code rates, thus achieving fine-grained code rate allocation. At the receiver, MIMO detection, demodulation, and decoding are jointly iteratively processed according to the corresponding multi-level coded modulation structure based on the same parameter table. Joint optimization of detection and decoding performance is achieved through hard or soft information feedback mechanisms, providing a high-spectral-efficiency and high-reliability transmission scheme for future mobile communication systems.

The application discloses a heterogeneous device log processing method and device of an industrial Internet of Things, an apparatus and a medium, relates to the technical field of Internet of Things logs, and comprises the following steps: creating a runtime instance and initializing a log module; collecting log data of an application processor and a communication processor to form a unified log stream; selecting a log analysis strategy according to a source identifier to generate structured log analysis data; and receiving configuration update data to perform dynamic hot updating. The heterogeneous log processing is decoupled from the AP firmware, all logs are isolated and processed in a WASM sandbox, the stability of a host program and an AP system is not affected by an exception, the log module and the strategy configuration can be dynamically and hotly updated, log levels and filtering rules can be adjusted in real time in response to requirements, atomic switching and automatic fallback ensure business continuity, log collection and processing are not interrupted, differential configuration updating effectively reduces network transmission overhead, and the application is particularly suitable for an industrial Internet of Things edge computing scene in which network bandwidth and computing resources are limited.

Disclosed is a system (104) and a method (500) for stream-based packet processing in a communication network. The method comprises receiving values corresponding to a plurality of configuration parameters including a parsing unit count. The method further comprises receiving, from a network interface, a plurality of packets corresponding to one or more data streams based upon the values of the plurality of configuration parameters and forming one or more batches of the plurality of packets based on the values corresponding to the plurality of configuration parameters. The method further comprises retrieving the one or more batches stored in a shared memory based on the value of the parsing unit count. The method further comprises parsing the plurality of packets in the retrieved batches to extract protocol header fields and classifying, based on the parsing, the plurality of packets into parsed packets and unparsed packets to enable the stream-based packet processing.

Replicating data to the cloud includes a first dataset that is replicated from a first storage subsystem to the cloud, with only the changes made to the first dataset since a previous replication event being replicated to the cloud. Additionally, the data may be deduplicated and compressed before being replicated from the first storage subsystem to the cloud. In some cases, the first dataset may be replicated to the cloud in a stream format which is not directly usable and then, at a later point in time, loaded onto a second storage subsystem from the cloud.

This invention discloses an adaptive frame abstraction interface management method and system for multiple encoding formats, relating to the field of data processing technology. The method includes: constructing a format configuration state diagram and a page transition time sequence; identifying the page transition time interval for corresponding format jumps; extracting structural difference parameters of the preceding and following formats and making them dimensionless to obtain fragmentation expectation parameters; making the actual time dimensionless and adding it to the expectation parameters to obtain reassembly resistance parameters; calculating the pre-allocated memory capacity based on the resistance parameters and the real-time bit rate; and performing physical isolation and decoding context preloading. This invention solves the problem of latency accumulation and screen stuttering caused by insufficient underlying memory reassembly and state machine switching during dynamic switching of multiple encoded video streams, achieving smooth transitions and low-latency decoding of video streams.

The application discloses a data storage method and device, electronic equipment and storage medium, comprising: obtaining the original data stream of the vehicle, extracting data features from the original data stream, generating the hash value of the original data stream, obtaining the current load state of the vehicle and the data priority of the original data stream, determining the data shunting operation to be performed on the hash value, wherein the data shunting operation includes one of data chaining, local caching and discarding, and the target hash value of the data shunting operation as data chaining is stored to the preset blockchain, wherein the blockchain includes an edge chain and a core chain, the edge chain is used to build the target hash value as a state tree storage, and the core chain is used to store the root hash value of the state tree in the edge chain. The application uses the edge layer to perform hash calculation on the data to be stored, retains the core information of the data, dynamically allocates the edge layer storage data shunting, and realizes real-time storage of decentralized tamper-proof data in combination with the layered blockchain.

The application provides a log storage method, device, electronic equipment and storage medium, wherein the log storage method comprises: obtaining a file instance used for processing a to-be-stored log; encoding an attribute field in the to-be-stored log to obtain a first byte stream corresponding to a digital number and storing the first byte stream in a block cache area; compressing the byte stream in the block cache area based on the file instance to obtain a second byte stream; encrypting the second byte stream in the compressed cache area based on the file instance to obtain a third byte stream; and writing the third byte stream into a byte stream file based on the file instance, wherein the byte stream file comprises a file block and a file header, and the third byte stream is stored in the file block. The application can write the to-be-stored log into the byte stream file without the aid of a third-party data storage tool. On the other hand, the application can obtain a byte stream file with smaller storage space by encoding and compressing the to-be-stored log.

This invention discloses a multi-source data intelligent connection management method, belonging to the field of data processing and information fusion technology. The method includes: deploying an intelligent agent program to listen to window message streams and reconstruct operation paths; synchronously capturing and parsing network requests to obtain structured business data; establishing a mapping relationship between operations and data through multimodal fusion parsing; filtering target data based on a no-code rule base; standardizing formats using a conversion engine; generating RPA instructions with the assistance of a large model; embedding a digital credential mechanism throughout the entire chain to ensure data integrity; and dynamically enhancing business logic through extension point injection. This invention aims to solve the problems of difficult data collaboration, high adaptation costs, and insufficient security caused by the lack of standard interfaces in heterogeneous business systems. It achieves high-precision, low-latency, and auditable automatic data collection, conversion, and entry across closed systems, significantly shortening the supplier access cycle and improving the efficiency of supply chain four-flow collaboration and system scalability.

This application provides an in-vehicle data communication method and system. The in-vehicle data communication method includes: running an ADMA ROS driver node on an industrial control computer to receive raw data streams sent by ADMA devices and publish the raw data streams as a first ROS topic; running a data parsing and encapsulation node to subscribe to the first ROS topic to parse the raw data streams to obtain truth data, encapsulating the truth data into a ROS message format, and publishing it as a second ROS topic; and running at least one algorithm application node to subscribe to the second ROS topic to obtain the truth data in the ROS message format. This application completely bypasses the traditional intermediary layer of MCU and CAN / DBC, proposing to directly connect ADMA to an industrial control computer equipped with ROS via Ethernet, and constructing a standardized, software-defined, and highly decoupled data pipeline through well-defined ROS nodes and a publish-subscribe message mechanism.

This invention provides a method and system for aggregating and storing water management control data based on edge acquisition, relating to the field of data processing technology. The method includes: deploying edge acquisition terminals to collect multi-source water management control data; performing quality cleaning to obtain standardized water management control data; performing real-time event detection and feature value extraction on the standardized water management control data at the edge side to generate target water management control data; based on the target water management control data and network perception, executing local reliable caching, online sending, and breakpoint resumption control according to preset priorities to obtain a data reporting stream; establishing a hierarchical message buffer and asynchronous decoupling channel on the central platform to persistently access and aggregate the data reporting stream to obtain a data aggregation stream; performing fine-grained classification and hierarchical storage processing on the data aggregation stream to obtain traceable hierarchical storage results; and providing a unified data service interface for the traceable hierarchical storage results to support on-demand subscription and distribution, completing the data value closed loop.

A terminal including a transceiver, an analog-to-digital converter and control circuitry. The transceiver receives a stream of bursts from a plurality of units. One of the units is an electronic device. The analog-to-digital converter receives an upstream burst from the transceiver during an active time slot. The upstream burst is one of the bursts. In response to sampling the upstream burst at a sample rate, the analog-to-digital converter converts the upstream burst from an analog waveform into a sequence of digital values. In response to a coefficient for the electronic device being a default coefficient, the control circuitry stores the digital values into memory during the active time slot.

The application discloses a large language model weight compression method, an inference operation method, a system and a medium, and belongs to the technical field of large language model weight compression and inference operation optimization. The method comprises the following steps: acquiring a weight matrix of a large language model and constructing an exponential high-frequency window and a benchmark value, classifying and encoding each weight data into a fixed-length code word and splitting the fixed-length code word into an independent bitmap, synchronously generating a compact value stream and a rollback value stream, and packing the compact value stream and the rollback value stream into weight compression data; loading the weight compression data, generating a state mask and a channel prefix mask in a register based on the bitmap, calculating a reading offset by using a population count instruction, reading a compressed data segment from a corresponding data stream, reconstructing weight data in the register in combination with the code word, and finally directly inputting the weight data into a matrix multiplication unit for operation; therefore, by implementing the application, the problem of efficiency reduction caused by control flow divergence and redundant memory access in GPU inference of existing variable-length encoding can be solved, and the model inference efficiency under lossless compression can be improved.

This invention relates to the field of distributed streaming data processing and dynamic reconfiguration of cross-domain processing logic, and particularly to a real-time data stream analysis method and system for the international automotive market. The method includes: acquiring pure physical kinematic state data of the vehicle and pre-authorized original environmental data from a pre-defined system; calculating and generating a transient symmetric key; encrypting the original environmental data into ciphertext data and writing it into a memory buffer; inputting the kinematic data into a gradient descent optimization model, fine-tuning the feature transformation computation graph weight matrix and bias terms, and outputting a new rule domain matching operator path; reading the hardware clock cycle value and injecting it into the header of the ciphertext data tensor to generate a routing marker, guiding the data flow to the new operator path; after confirming the path switch, sending an overwrite command to replace the transient symmetric key with an all-zero physical state, blocking the decryption and restoration path of the remaining ciphertext data. This invention solves the problem of feature data leakage or computational interruption caused by differences in feature transformation standards during cross-regional driving.

This application provides a communication method and apparatus. The method can be applied to protocols supporting IEEE protocols, such as 802.11be / Wi-Fi 7 / Wi-Fi 8, IEEE 802.11Integrated mmWave, IEEE 802.11bf / Sensing, UWB, or StarFlash. The method includes: sending an SCS request frame containing an SCS identifier, a first TID, and a first UP. The SCS identifier identifies a first SCS stream, the first TID is the TID to which the first SCS stream is requested to be mapped, and the first UP is the UP to which the first TID is requested to be mapped. The values ​​of the first TID and the first UP are different. The method also includes receiving an SCS response frame in response to the SCS request frame. This reduces the time required for the receiving end to parse data packets of the SCS stream.

This disclosure provides a method, apparatus, electronic device, and storage medium for synchronizing video inference results, relating to the field of video processing technology. The method includes: acquiring the inference result of each frame in a video stream task, associating each frame's inference result with its corresponding frame number, and storing it in a database; upon receiving a frame acquisition instruction from a service scheduling terminal, extracting new inference results whose frame numbers follow those recorded in the real-time frame pointer from the database, starting from the frame number recorded in the real-time frame pointer; wherein the real-time frame pointer is used to record the frame number corresponding to the last frame's inference result sent to the service scheduling terminal; and sending the new inference results to the service scheduling terminal so that the service scheduling terminal parses the target frame number corresponding to the last frame's inference result in the new inference results and updates the frame number recorded in the real-time frame pointer to the target frame number. This disclosure avoids redundant data transmission and scheduling through incremental data synchronization.