201results about How to "Reduce data transfer volume" patented technology

The invention discloses a large language model training method and system based on elastic federal low-rank adaptive fine tuning, which adopts federal aggregation based on singular value decomposition (SVD) to reconstruct a local LoRA module of a client into a local increment matrix, then aggregates the local increment matrix into a global increment matrix, and generates a LoRA module of the next round through SVD, thereby effectively eliminating aggregation errors of traditional FedLoRA. According to the method, the LoRA rank is dynamically distributed for each weight matrix, and the balance between statistics and hardware efficiency is realized. In order to further reduce communication overhead and maintain model precision, a self-adaptive matrix freezing mechanism is introduced, and whether freezing is carried out or not is judged according to training stability of a weight matrix. According to the method, challenges of traditional FedLoRA in the aspects of training efficiency, communication overhead, model precision and the like are solved.

The invention discloses an efficient real-time voice and video transmission method, which is characterized in that a proper video frame coding mode is selected according to the state of video content, so that a system can effectively reduce the data volume and save the network bandwidth by using a high-compression-rate coding mode in a static state; more video details are reserved by using a low-compression-rate coding mode in a dynamic state, so that the data transmission quantity is reduced to the greatest extent on the premise of not influencing the video quality, and the overall transmission efficiency is improved; meanwhile, due to the fact that the transmitted data size is reduced and the data transmission speed is increased, transmission delay of the audio and video data is reduced, the phenomenon of audio and video asynchronization caused by data packet disorder can be effectively reduced through reasonable buffer area management, it is ensured that the audio and video data are played according to the generated time sequence, the watching experience of a user is optimized, and the user experience is improved. High-quality and low-delay audio and video transmission service is realized, and high requirements of users on real-time performance and definition are met.

The invention belongs to the technical field of interventional robots, and discloses a method, device and equipment for automatically adjusting the collection frame rate and a medium, and the method comprises the steps: obtaining the current movement speed of an interventional robot; adjusting a corresponding collection frame rate according to the movement speed; and controlling a camera to perform data acquisition based on the acquisition frame rate. The appropriate frame rate is selected according to the real-time movement speed of the interventional robot, the high frame rate is provided during high-speed movement by dynamically adjusting the collection frame rate of the camera so as to reduce movement blur, and the frame rate is reduced in a low-speed or static state so as to save bandwidth resources. According to the method, unnecessary data transmission quantity is reduced, network congestion and processing burden are relieved, the overall performance and stability of the system are improved, real-time and accurate transmission of image information is ensured, smooth and clear visual feedback is provided, the accurate control capability in the intervention process is improved, and the safety of the intervention process is improved.

The invention relates to the technical field of intelligent interaction control, and discloses a dynamic loading method and device for an intelligent interaction control unit of a television box. The method comprises the following steps: performing data mining on a user operation log to obtain a function module classification list, a use frequency statistical table and a dependency graph; and performing modular reconstruction on the software system based on the classification list to generate a function module package. And constructing a dynamic priority queue by using the frequency statistical table and the dependency graph. And carrying out predictive module loading in combination with the priority queue and the function module packet to obtain a loaded module list and a resource allocation state. And unloading the intelligent module according to the loading condition, and updating a module list and a resource use state. And executing incremental updating based on the version information in the module packet, updating the functional module packet and reconstructing the dynamic priority queue to obtain a target priority queue. According to the method, the resources can be flexibly allocated according to actual requirements, resource waste caused by static loading is avoided, and the resource utilization efficiency is improved.

The invention relates to a coupler system based on MECHATROLINK-III and a communication method, the system comprises a MECHATROLINK-III master station, a MECHATROLINK-III special cable, a MECHATROLINK-III coupler and a plurality of plug-in IO modules, the MECHATROLINK-III coupler is used for managing all the plug-in IO modules in a slot so as to carry out ordered IO input and output, and the plug-in IO modules are used for carrying out plug-in IO input and output. The communication module is used for carrying out communication interaction between IO data of the whole coupler system and a MECHATROLINK-III master station; and the MECHATROLINK-III coupler is integrated with a plurality of control units or execution units. The MECHATROLINK-III coupler has the advantages that the communication stability and the communication efficiency are guaranteed while the cost of the MECHATROLINK-III coupler system is reduced.

The invention discloses a distributed collaborative modularized rapid detection method and device for a box-type substation, and the method comprises the following steps: enabling each distributed module to judge whether the distributed module has a fault possibility or not based on collected operation data; the distributed module with the fault possibility initiates an abnormal signaling; after the other distributed modules receive the abnormal signaling, whether correction characteristic parameters of the other distributed modules are filled in corresponding positions in the abnormal signaling or not is confirmed, and the abnormal signaling is sequentially transmitted among all the other distributed modules; and summarizing all the obtained correction characteristic parameters, calculating a posterior fault probability by the distributed module which initiates the abnormal signaling, and judging that a fault exists when the posterior fault probability exceeds a set threshold value. According to the distributed architecture, the bottleneck of a centralized system is avoided, the modules are processed in parallel, and the detection efficiency is improved. Through the feature correlation model, interference of redundant information is avoided, and the accuracy of fault judgment is improved.

The invention relates to an integrated circuit, a processor and a computing device for data processing. The integrated circuit comprises a data storage circuit and a matrix operation circuit. The data storage circuit is configured to store matrix data. The matrix operation circuit includes: a control logic circuit configured to: read a first matrix and a second matrix from a data storage circuit based on a specified data format indicating that elements in the first matrix are floating point type data and elements in the second matrix are integer type data; and the arithmetic logic circuit is configured to convert elements in the first matrix and the second matrix into elements of a target floating point type according to the data types of the elements indicated by the specified data format, and determine result matrixes corresponding to the first matrix and the second matrix based on the converted elements of the first matrix and the second matrix. The integrated circuit is helpful for improving the matrix operation efficiency and reducing the occupation of storage space and the data transmission quantity.

The invention discloses a remote monitoring method and device for an explosion-proof distribution box, and relates to the field of explosion-proof distribution boxes. The method comprises the following steps: acquiring monitoring data of a plurality of explosion-proof distribution boxes in a target area; acquiring a priority coefficient of each piece of monitoring data; and compressing the monitoring data based on the priority coefficient, and transmitting the compressed monitoring data to a central server, so that the central server performs remote monitoring on the monitoring data. By implementing the technical scheme provided by the invention, efficient remote monitoring of the explosion-proof distribution box can be realized.

According to the model training-oriented gradient data transmission method and the related equipment provided by the invention, the compression ratio is dynamically determined according to the change of the transmission rate by monitoring the transmission rate in real time, so that the gradient data can be intelligently compressed according to the compression ratio, the data transmission quantity is remarkably reduced, and the network burden is relieved; and meanwhile, the transmission rate of each working end is adjusted through congestion control according to the bandwidth of a transmission link between the switch and the parameter server so as to relieve congestion. Under the combined action of the two, the transmission time of the gradient data is remarkably shortened, and the data transmission efficiency is improved, so that the training process of the distributed artificial intelligence model is accelerated, and the overall training efficiency is improved.

The invention provides a mine video semantic communication transmission method and system based on deep learning, which can realize the recovery of observable data and the execution of pragmatic tasks. According to the method, a semantic communication system model under a mining working face is constructed, the semantic communication system model comprises a semantic encoder and a semantic decoder which are realized based on a deep neural network, and the semantic encoder is located at a transmitting end and used for extracting data containing observable data and semantic information; the semantic decoder is located at the receiving end and used for reconstructing data related to empirical data and pragmatic tasks. According to the method, a semantic encoder and a semantic decoder are trained based on a DL-SC algorithm, image restoration and pragmatic task execution are considered in the aspect of design of a loss function, appropriate distortion measure functions are selected for the image restoration and pragmatic task execution respectively, and then the weighted sum of the distortion measure functions is solved to form a semantic distortion measure function to serve as the loss function. According to the invention, more efficient information interaction can be realized with lower data transmission quantity.

The invention discloses a ClickHouse-based fused CDN (Content Delivery Network) business data processing method, device and equipment, and the method comprises the steps: obtaining business data of a corresponding fused CDN manufacturer through calling a data interface of each fused CDN manufacturer in real time; performing preprocessing including data format standardization, field expansion and invalid data filtering on the service data by using Spark, packaging the preprocessed service data and sending the packaged service data to a Kafka cluster; the method comprises the following steps: creating a distributed data table based on a ReplacingMergeTree engine in a ClickHouse database, consuming service data in a Kafka cluster in real time by utilizing Spark, and writing the service data into the distributed data table of the ClickHouse database in batches; and retrieving target business data from the ClickHouse database according to a data query request of a user, and converting the target business data into a JSON format to serve as a query result to be returned. The query performance, the storage efficiency, the expandability and the write-in performance can be improved.

The embodiment of the invention provides a data processing method of a display device and the display device, for each backlight driver, if the dimming data of the current frame of the backlight driver is different from the dimming data of the previous frame, a backlight controller sends the dimming data of the current frame to the backlight driver; the backlight driver drives the light emitting unit to emit light according to the dimming data of the current frame, if the dimming data of the current frame of the backlight driver is the same as the dimming data of the previous frame, the backlight controller does not send the dimming data of the current frame to the backlight driver, and the backlight driver drives the light emitting unit to emit light according to the dimming data of the previous frame. Compared with sending the dimming data of all the backlight drivers, the dimming data of the current frame is only sent to the backlight drivers with changed data, so that the data transmission quantity, the electromagnetic interference, the power loss and the data transmission time can be reduced, and the response rate of the display equipment is improved.

The invention discloses a data monitoring system based on power telecontrol, which comprises a sensor module, a preprocessing module, a transmission analysis module, a data processing module, a data transmission module, a data receiving module, a data verification module, a reacquisition module and a monitoring display module, and relates to the technical field of data monitoring. According to the data monitoring system based on power telecontrol, through an intelligent transmission analysis mode, unnecessary data transmission is avoided, network resources are saved, only operation data exceeding the standard fluctuation difference are transmitted, the data transmission amount is greatly reduced, the communication cost and the system burden are reduced, and the transmission efficiency is improved; different attribute operation parameters of a plurality of time nodes in a specified period are preprocessed, an accurate fluctuation interval and a standard fluctuation difference are calculated, the operation condition of a controlled station can be reflected more accurately, and the accuracy and reliability of data processing are ensured based on scientific variance calculation and comparison.

The invention discloses an analysis query method and device for a database, and the method comprises the steps: obtaining data corresponding to a query instruction from a pre-constructed preset storage structure through a preset index algorithm when the query instruction is received in a data query process of a database system, the preset storage structure is obtained by optimizing a storage structure design layer of a multi-version concurrency control mechanism, the preset storage structure is used for avoiding volume expansion caused by multi-version control and improving the performance of data analysis and query, and data is compressed through a data compression algorithm. And storing the compressed data in a preset storage structure in a column storage mode.

The invention provides a UDP report system and method based on a DPU, electronic equipment and a medium. The system comprises an initialization module, a software report data receiving module, a hardware report data receiving module, an arbitration module, a UDP packing module and a sending module. The initialization module is used for performing initialization configuration on the received static data; the static data is used for representing header information of a current transaction report; the arbitration module is used for sequentially transmitting the report data received by the software report data receiving module and the report data received by the hardware report data receiving module to the UDP packaging module according to the transmission priority; the report data is business data transmitted by each transaction report; the UDP packaging module is used for packaging the static data and the report data to obtain a UDP data packet when the report data is received; and the sending module is used for sending the UDP data packet to a transaction network port.

The invention relates to the technical field of tunnel deformation monitoring, in particular to a tunnel primary support surrounding rock deformation monitoring system and method, and the method comprises the steps: data collection, data transmission, data preprocessing, model training and verification, real-time analysis, feature extraction and early warning and response. Data acquisition equipment and an edge server are installed and deployed on the inner side wall of tunnel primary support surrounding rock, the data acquisition equipment acquires original data, it is ensured that data coverage is comprehensive and not excessive redundancy exists, the original data are timely transmitted to the edge server in a wireless mode of the low-delay Internet of Things, the data transmission process is combined with an MQTT protocol, and the data transmission efficiency is improved. The method comprises the following steps: preprocessing data, selecting a trained convolutional neural network model CNN, carrying out online analysis on the data transmitted in real time, predicting the deformation trend of the tunnel primary support surrounding rock, and immediately sending an early warning to a monitoring and early warning module when the deformation trend exceeds a set deformation threshold value, so that the system can monitor the deformation of the tunnel primary support surrounding rock in real time and ensure the accuracy and reliability of the data.

The invention provides a device for limiting windage yaw galloping of a high-altitude power transmission line and a use method, and relates to the technical field of high-altitude power transmission lines, the device architecture comprises a line suppression module, a data monitoring and processing module, a data co-processing module, a line repair module, a pulse type air injection device and a plasma flow generator, a dynamic change magnetic field can be generated to counteract the windage yaw acting force by regulating and controlling the magnetic field of the superconducting coil and intelligently distributing the magnetic material; meanwhile, the miniature crawling robot and the flying micro-robot cluster can quickly respond when the sensor detects that the line or the windage yaw preventing device is damaged, the miniature crawling robot can crawl on the surface of the power transmission line to check and repair local damage, and the flying micro-robot cluster can deal with large-area damage or damage of multiple parts; the possibility and amplitude of windage yaw galloping are reduced from the source, the reliability of power transmission is improved, and windage yaw galloping is effectively inhibited.

The invention relates to the technical field of communication, in particular to a multiple access method based on model semantic orthogonality and a related device, which are applied to a scene in which a sending end transmits a first quantity of service information to a receiving end, and comprises the following steps: the sending end converts the first quantity of first service information into a first quantity of semantic codes; the sending end quantizes the semantic code to obtain an orthogonal code, and the orthogonal code is used as a to-be-transmitted signal; the sending end carries out superposition transmission on the to-be-transmitted signal at an air interface; and the receiving end decodes the received signal to obtain second service information corresponding to the first number of first service information.

The invention discloses an experimental equipment management system based on LoRaWAN, and aims to solve the problems of resource allocation conflicts, low data transmission efficiency and the like in laboratory equipment management. The system comprises an equipment control unit, a LoRaWAN communication unit, an electric energy monitoring unit, a user interaction display unit, a remote reservation management platform, a gateway and server unit and a firmware management and upgrade unit. A differential packet is generated by using a BSDIFF algorithm, and an LZ77 compression algorithm is combined, so that the transmission data volume of the system is effectively reduced; meanwhile, the gateway dynamically adjusts the size of a data block according to the network quality, and adopts a frame insertion strategy to embed upgrade data in a service data transmission gap; and after the terminal node receives and verifies the differential packet, new firmware is generated to replace old firmware, so that remote efficient updating is realized under the condition of ensuring normal use of equipment. The laboratory equipment management system greatly improves the management efficiency and stability of laboratory equipment, and is suitable for equipment management scenes of various scientific research laboratories.

The embodiment of the invention provides a communication method, device and equipment, and relates to the technical field of AIs, and the method comprises the steps: obtaining the respective AI calculation capability of a plurality of second network function equipment; dividing the first AI task into a plurality of first AI sub-tasks based on the respective AI computing capabilities of the plurality of second network function devices, and selecting a plurality of third network function devices for executing the plurality of first AI sub-tasks from the plurality of second network function devices; and sending respective corresponding first AI subtasks to the plurality of third network function devices, and receiving first AI subtask execution results sent by the plurality of third network function devices, thereby achieving the purpose of executing the AI task in the mobile communication network.

The invention discloses a livestock farm harmful gas monitoring system and method. The system comprises a distributed gas monitoring module, a sensor control module, a data analysis module and an alarm module. The distributed gas monitoring module is used for acquiring gas concentration data in the monitoring area in real time; the sensor control module is used for acquiring a key monitoring area based on animal gathering behaviors and controlling the work of a gas sensor arranged in the key monitoring area; the data analysis module is used for constructing a harmful gas evaluation model and analyzing the gas concentration data by using the harmful gas evaluation model to obtain a harmful gas evaluation result; and the alarm module gives an alarm when the harmful gas evaluation result is abnormal. According to the invention, distributed anomaly detection is realized, the accuracy and the real-time performance of gas monitoring are improved, the power consumption and the data transmission quantity are obviously reduced, and intelligent and efficient management of farm environment monitoring is realized.

The invention discloses a luggage falling identification system and method for an independent carrying unit of a luggage processing system, and the system is characterized in that the system comprises an image collection unit, an edge calculation unit, an information processing unit, and an alarm prompt unit; the image acquisition unit is used for acquiring image data of a luggage loading node and a plurality of nodes in a luggage conveying process; the edge calculation unit comprises a preprocessing module, a luggage feature recognition module, a character recognition module and a network module; the luggage feature recognition module comprises an input layer, a luggage feature enhancement layer, a backbone network, a feature fusion and attention mechanism layer and a luggage feature prediction head which are connected in sequence; the information processing unit is used for receiving the feature data sent by the edge device, performing deep analysis and comparison, and detecting the luggage falling condition; and the alarm prompt unit is used for detecting the falling of the luggage and immediately notifying related personnel when the luggage falls. According to the invention, through cooperative work of the edge calculation and the background server, a luggage falling event is accurately identified, and an alarm is triggered in time, so that the security and efficiency of the airport luggage processing system are improved.

The invention relates to a configuration data synchronization method and device, equipment, a storage medium and a program product. The method comprises the steps of determining a changed target sub-Merkle tree according to a hash value of each node of a top-layer Merkle tree in response to a data synchronization request sent by a client, determining a changed target leaf node according to a hash value of each node in the target sub-Merkle tree, and synchronizing the changed target leaf node with the target leaf node according to the hash value of each node in the target leaf node. And sending first target information comprising resource configuration data corresponding to the target leaf node, a hash value of the target leaf node, a hash value of an upper-layer node of the target leaf node and a hash value of a root node of a top-layer Merkle tree to the client, so that the client performs configuration data synchronization. Therefore, the changed target leaf node can be determined layer by layer based on the hierarchical Merkle tree, the pertinence of configuration data synchronization is improved, only the first target information related to the changed target leaf node needs to be transmitted, the speed of configuration data synchronization is improved, and the delay of configuration data synchronization is reduced.

The invention discloses a smart city system, a data transmission method and a program product, the smart city system at least comprises a smart lamp post, an edge server and a cloud server, the smart lamp post is provided with a binocular camera and a data transmission device, the binocular camera is configured to acquire an image in a field of view, calculate a disparity map based on the image, and transmit the disparity map to the cloud server; the data transmission device is configured to send the image and the disparity map to the edge server; the edge server is configured to cut the image based on a parallax value in the parallax image to obtain an image block, generate an encryption key based on the parallax value corresponding to the image block, encrypt the image block by using the encryption key, generate encrypted data, and send the encrypted data to the cloud server; and the cloud server is configured to decrypt the encrypted data and construct a smart city. According to the technical scheme of the invention, the data transmission amount in the construction process of the smart city can be reduced, and the safety of data transmission can be improved.

The invention belongs to the field of vehicle-mounted tire pressure monitoring, and discloses a vehicle-mounted tire pressure monitoring platform and method based on the Internet of Things. Comprising the following steps: collecting comprehensive state data of each tire, and constructing a joint structure chart of the tires; calculating the difference data of the joint structure chart at the current time point and the previous time point to form a difference packet, and transmitting the difference packet by using an improved difference technology; building and training a monitoring and early warning model, and importing the difference packet into the trained monitoring and early warning model to obtain early warning information; and the early warning information is displayed through a vehicle-mounted interface, and the vehicle is adjusted, so that intelligent, high-efficiency and high-quality monitoring of the vehicle-mounted tire pressure based on the Internet of Things is realized.

The present invention discloses an industrial automatic control data processing system and method based on digital twin, which belongs to the field of industrial automation control technology. The present invention establishes a connection between the edge side and the cloud center through the edge side management module, automatically collects industrial data through the edge collection node, and stores the collected industrial data in the edge device; sends a data request to the cloud center through the data mapping module, and the cloud center receives the industrial data uploaded by the edge side and maps the received industrial data in the digital space based on the digital twin technology; analyzes and feedbacks the data collection status of the edge side through the cloud processing module; updates the data collection strategy of the edge side according to the analysis results fed back by the cloud center through the collaborative control module, and transmits the updated data collection strategy to the edge device, and adjusts the data collection business of the edge collection node. The data in the industrial production process is processed more flexibly, which reduces the resource consumption of operation and maintenance.

The invention relates to the technical field of over-the-horizon remote control mobile robots, in particular to an over-the-horizon remote control mobile robot low-delay double-view-angle control system and method, and the system comprises a mobile robot end and a remote control end, and the mobile robot end comprises a fisheye lens group and an image fusion processing module; the fisheye lens group captures a 360-degree panoramic image, the image fusion processing module preprocesses image data, generates a panoramic video stream in real time through an image fusion algorithm, and transmits the panoramic video stream to the remote control end through the communication module; and the remote control end performs selective double-view-angle display, and adjusts the observation view angle in real time based on the head action of the user or a manual instruction. The data transmission quantity is reduced, the video quality is ensured, the transmission efficiency is improved, the stable transmission of the panoramic video stream is ensured, the delay and packet loss phenomena are reduced, the time delay problem is solved, and the real-time performance and comfort of control are better.

The invention discloses a photovoltaic module defect lightweight detection method for an unmanned aerial vehicle, which performs lightweight improvement based on a YOLOv5s model, so that the method can be deployed on an embedded GPU unmanned aerial vehicle with limited computing resources to perform real-time detection on photovoltaic module defects. Comprising the following steps: establishing a photovoltaic module surface defect data set, and preprocessing an image; a YOLOv5s model is improved, a ShuffleNetV2 network is adopted to replace an original backbone network CSP structure, the parameter quantity and the calculation quantity of the model are greatly reduced to realize light weight, an effective channel attention module ECA is introduced, and the recognition capability of the model on defect characteristics of a photovoltaic module is enhanced; training parameters are set, and the improved model is trained; and inputting the defect image of the photovoltaic module to be detected into the trained model, and outputting the position and category information of the surface defect of the photovoltaic module to be detected. According to the method, the size and complexity of the model are effectively reduced, and the method is particularly suitable for efficient photovoltaic module defect detection on an unmanned aerial vehicle system with limited resources.

The invention relates to a video call method and system. The method comprises the following steps: acquiring a video stream of a video call, and extracting all video frames from the video stream; determining a key video frame and a non-key video frame corresponding to the key video frame from all the video frames based on a dynamic vector detection algorithm; carrying out downsampling processing on the non-key video frame; and sending the non-key video frames after downsampling processing and the corresponding key video frames to an intelligent terminal. According to the invention, the problem that resource finiteness of equipment cannot be considered while video call quality is improved in the prior art is improved.

The present invention discloses an XR information transmission control method, device, system and storage medium, the method comprising: controlling a cloud server to parse and virtually render received XR information to generate a virtual 3D image; the XR information is transmitted by an XR application; the XR information comprises: user posture information and scene information; compressing the virtual 3D image and the original audio signal frame by frame respectively to generate image compression data and audio compression data, and transmitting the image compression data and the audio compression data as compressed data to an edge end; controlling the edge end to decompress and detect changes in the compressed data to generate a detection result, and sending the detection result to a terminal; the detection result comprises the changed virtual 3D image or audio information and the change type thereof; controlling the terminal to update the corresponding data in the client according to the detection result, so as to reduce the data transmission volume and bandwidth occupancy, thereby improving the efficiency and stability of XR information transmission.