44results about How to "Reduce data transfer bandwidth" patented technology

A packet switching node having a pipelined packet processing architecture processing packets received via an input port associated with the packet switching node is presented. The method performed by the apparatus includes: determining a packet frame type of the packet received; selectively extracting packet header field values specific to a packet frame type, the extracted packet header field value including packet addressing information; ascribing to the packet a preliminary action to be performed in respect of the packet; searching packet switching information tracked by the packet switching node based on extracted packet addressing information; formulating a preliminary switch response for the packet; classifying the packet into one of a plurality of packet flows; modifying the preliminary switch response in accordance with one of the preliminary action, the packet flow into which the packet was classified, and a default port action corresponding to the input port; modifying the packet header in accordance with one of the preliminary action, the packet flow, and the default port action; and processing the packet in accordance with the switch response. Advantages are derived from: pipelined processing of packets which enables short-cutting the rest of the processing for improper packets; a flexible frame type determination which is fast for well know frame types yet flexible in support of new frame types delaying obsolescence of a particular implementation; an early determination of a processing action which is successively refined by subsequent stages; a combined Layer-2 and Layer-3 network addressing search engine operating on short bit length indexed Layer-2 and Layer-3 network addresses reducing network address table storage requirements, requiring a reduced data transfer bandwidth for network address table access, a large external hashed primary network address table, and a small internal secondary network address table; an early determination of a switch response; and packet-classification-based switch response and packet header modification.

Arrangements and method for enabling and disabling cache bypass in a computer system with a cache hierarchy. Cache bypass status is identified with respect to at least one cache line. A cache line identified as cache bypass enabled is transferred to one or more higher level caches of the cache hierarchy, whereby a next higher level cache in the cache hierarchy is bypassed, while a cache line identified as cache bypass disabled is transferred to one or more higher level caches of the cache hierarchy, whereby a next higher level cache in the cache hierarchy is not bypassed. Included is an arrangement for selectively enabling or disabling cache bypass with respect to at least one cache line based on historical cache access information.

An information processing system performs a plurality of tasks within a specific time interval. The system includes a bus, a plurality of processors which transfer data via the bus, and a unit for performing a scheduling operation of determining execution start timing of each of the tasks and at least one the processors which executes the tasks, based on cost information concerning a time required to perform each of the tasks and bandwidth information concerning a data transfer bandwidth required by each of the tasks, to perform the tasks within the specific time interval without overlapping execution terms of at least two tasks of the tasks, the two tasks requiring data transfer bandwidths not less than those of the others of the tasks.

The invention provides a processing system for a binary weight convolution neural network. The system comprises at least one storage unit for storing data and an instruction, at least one control unit for obtaining the instruction stored in the storage unit and sending out a control signal, and at least one calculation unit for obtaining one layer of node value in a convolution neural network and corresponding binary weight value data from the storage unit and performing adding and subtraction operations to obtain a next layer of node value. Therefore, the data bit width during convolution neural network calculation process can be reduced; the convolution operation speed can be increased; and the storage capacity and working energy consumption can be reduced.

Techniques and tools for rendering procedural graphics are described. For example, an architecture is provided which allows evaluation of geometric, transform, texture, and shading procedures locally for a given set of procedure parameter values. This evaluation is performed in parallel for different parameter values on a single-instruction, multiple-data array to allow parallel processing of a procedure set. In another example, a sampling controller is described which selects sets of parameter points for evaluation based on information in tag maps, rate maps, and parameter maps.

The present invention introduces a device, method and computer reading media for mapping establishing text to audio time sequence. The present invention also introduces a device, method and computer reading media for playing audio text animation. A mapping agent (10) takes the text (12) and the corresponding audio record (11) as input, and makes begin and end time assign to the text unit (15). The player (50) takes text (15), audio (17) and mapping (16) as input, and makes text animation and displays text (15) in-phase along with playing the audio (17). The present invention is used to endow vitality during the process of playing audio record; to instead the traditional playback control to control audio play; to play and display the note of voice recording; to implement the characteristic of flowing audio without using basic stream protocol.

The invention provides a method and a device for compressing and storing indicator diagram data of an oil pumping unit. The method comprises the following steps: S1, acquiring the indicator diagram data, and calculating to acquire difference value data of adjacent data in the indicator diagram data; S2, selecting difference value data within a preset value range from the difference value data for serving as normal difference values; S3, using data from which the normal difference values are removed in the difference value data as special difference values; S4, storing source point data in the indicator diagram data; storing the quantity of the special difference values and the positions and data values of the special difference values in sequence; storing the data of the normal difference values in sequence according in a bitwise mode. By adopting the embodiment of the invention, the indicator diagram data can be stored in a real-time database in a non-destructive way, and the storage space and maintenance cost are saved.

The invention provides a calculation method and device applied to neural networks. The calculation method comprises the following steps of: obtaining a binary convolutional kernel which only comprisesa numerical value 1 and a -1 weight; decomposing the binary convolution kernel into an initial convolution kernel and a feature convolutional kernel, wherein the initial convolution kernel and the feature convolution kernel are same as the binary convolution kernel in the aspect of dimensionality, the initial convolution kernel is a matrix formed by a weight with a numerical value of 1, and the feature convolution kernel is a matrix formed by a weight with a numerical value of -1 relative to the binary convolution kernel; and carrying out convolution calculation in a neural network on the basis of the initial convolution kernel and the feature convolution kernel. By utilizing the calculation method and device provided by the invention, the convolution calculation efficiency can be improved and the storage circuit overhead can be saved.

The invention discloses a method and a system for data transmission. The method comprises the steps of: a web server receives first data application information of a user, acquires a second data corresponding to the user, generates a data document of a first data format according to the first data application information and the second data and transmits the data document to a gateway; the gateway extracts application processing server information carried in the data document and data format of the second data and converts the data document from the first data format to a second data format corresponding to the application processing server information; and the gateway transmits a data document of the second data format to an application processing server corresponding to the application processing server information by using a transmission method corresponding to the data format of the second data. The method and the system can be used to improve the response speed and efficiency of the web server.

The invention discloses a method for reducing an uplink data transmission bandwidth in a distributed base station. On the side of a radio remote unit, processing on antenna carrier data after intermediate frequency processing is carried out is divided into an A branch and a B branch. The A branch comprises the following steps: removing CPs from the antenna carrier data, obtaining time domain data and carrying out a fast Fourier transforming on the time domain data, obtaining frequency domain data, removing a virtual subcarrier from the frequency domain data, obtaining a valid subcarrier containing PRACHs, decomposing the valid subcarrier according to RBs, removing the RBs occupied by the invalid PRACHs, and finally obtaining valid base band data. The B branch comprises the following steps: obtaining complete PRACH information from the antenna carrier data, removing CPs from the PRACH information, obtaining a PRACH detection signal, carrying out smoothing extraction on the PRACH detection signal, and obtaining a valid PRACH detection signal. The valid base band data and the valid PRACH detection signal are mapped to a transmission unit and sent to a base band processing unit. The invention further discloses the radio remote unit. According to the method for reducing the uplink data transmission bandwidth in the distributed base station and the radio remote unit, the data transmission bandwidth between the radio remote unit and the base band processing unit can be further reduced.

The invention discloses a terminal system based on vector image play and a realization method thereof. The terminal system comprises a communication module, a data caching and sorting module, a service output data analysis module, a non-vector image dot-matrix output play module, a vector image output play module, a display cache module and a terminal display screen. The realization method comprises the following steps that: the terminal receives the related data of the content to be displayed; the terminal analyzes and sorts the data; the terminal identifies an update area; the terminal analyzes and extracts the vector and non-vector image dot-matrix data; the terminal displays the non-vector dot-matrix data; the terminal displays the vector image; the terminal updates the result and stores the updated result in the display cache; and the terminal displays the result on the display screen. The system and method have the functional characteristics of high efficiency, wide applicability and the like, are easy to realize and favorable for displaying a three-dimensional effect, and have positive significance to the realization of a novel computing service system terminal based on mobile computing, distributed computing, cloud computing and the like and the popularization, application and research of a novel computing service system.

The invention provides an output adjusting method of a switching networked flight control system based on alternating event triggering. An alternating event trigger mechanism switched between periodicsampling and continuous event trigger mechanisms is constructed, trigger information and modal information are transmitted to a controller, missing of effective information is avoided by continuouslydetecting trigger conditions, and data transmission and updating are reduced by introducing a sampling period. A flight control system is described as a networked switching system; asynchronous switching control based on an event trigger mechanism is designed under the condition that network induced time delay and packet loss exist. The flight control system accords with model dynamic rapid change characteristics in the flight control system and actual flight working conditions of data transmission and processing between internal components based on a real-time network, reduces data calculation load of the flight control system, saves limited data transmission bandwidth, and improves overall performance and precision requirements. According to the technical scheme, only the output error information needs to be used, state information of the sensor and the aircraft is not depended, and engineering implementation is easy.

The invention provides a data compression method and device, a data decompression method and device and a processing method and device based on data compression and decompression, and the data compression method comprises the steps: receiving to-be-compressed data which is sparse data outputted by any layer of a neural network model; and performing compression processing on the to-be-compressed data based on a sparse compression algorithm and a bit plane compression algorithm to obtain compressed data. By utilizing the method, a relatively high compression ratio can be realized, so that the data transmission bandwidth and the storage space of an external memory can be saved, the memory access efficiency is improved, and the chip computing power is improved.

The invention relates to a discrete waveform data compression method based on a slope distribution. The method comprises the following steps: first of all, converting original discrete waveform data into a slope sequence, and enabling these slope values to form a set; next, according to a distribution of the slope sequence, selecting a plurality of category centers and taking the plurality of category centers as representative slope values; once again, in an error allowed range, enabling each of the representative slope values to approximate a line segment for each of adjacent points in an original discrete data waveform and connecting each line segment together, thereby implementing an approximation to the original discrete waveform and thus implementing a sparse representation of the original discrete waveform data; and then, performing compressed encoding on the sparse representation of the original discrete waveform data, thereby implementing compression processing on a waveform signal. A result shows that the method provided by the invention can effectively implement the compression of the waveform data under a condition in which useful components are retained.

The invention provides a random access method. The method comprises the steps of sequentially selecting and randomly accessing modules independent of the antenna number to a BBU, adding selected modules to an RRU to construct a newly built RRU, deleting selected modules in the BBU to construct a newly built BBU, sending the random access detection information from the newly built RRU to the newly built BBU through an IR interface module upon receiving the construction information sent by the newly built BBU, transmitting the construction information required for RRU treatment to the newly built RRU from the newly built BBU through the IR interface module, and reporting the random access detection information upon receiving the random access detection information. The invention also discloses a random access method applied to a randomly accessed radio remote unit and a base station processing unit. According to the embodiments of the present invention, the interface data transmission amount during the randomly accessing process is reduced. Furthermore, the data transmission bandwidth required for a randomly accessed transmission interface is reduced.

The invention relates to a data transmission method and device based on a remote exhibition and marketing system. The method comprises the following steps: A, collecting images of a display stand at atime interval that can be dynamically changed; B, comparing a front-frame image collected at the current time interval with a back-frame image collected at the previous time interval, and comparing the front-frame image collected at the current time interval with a back-frame image collected at the current time interval to detect whether image changes occur; C, adding video data composed of the image of each frame collected at the current time interval into a transmission queue once image changes are detected, otherwise, adding the front-frame image collected at the current time interval intothe transmission queue; D, carrying out transmission after conducting compression or interpolation on the data in the transmission queue. The device includes corresponding modules for implementing the above method. The technical scheme of the invention improves a software system of a remote exhibition and marketing system, makes data transmission more efficient, saves the data transmission bandwidth, and does not affect the real-time property of on-site interactive transmission.

The invention relates to a data processing method and device, electronic equipment and a storage medium, relates to the technical field of networks, and can improve the message data processing efficiency of the electronic equipment and reduce the storage space occupied by data, the data transmission bandwidth and the data transmission time consumption. Comprising the following steps: acquiring current to-be-transmitted target signaling data, and determining a target data type corresponding to the target signaling data; determining a target data dictionary corresponding to the target signaling data according to the target data type and a plurality of pre-trained data dictionaries; the target data dictionary is used for indicating the corresponding relation between the target data type and the target compression algorithm, and the target compression algorithm indicated by the target data dictionary is the compression algorithm with the highest compression ratio in multiple compression algorithms corresponding to the target data type; and performing compression processing on the target signaling data through a target compression algorithm indicated by the target data dictionary to obtain compressed target signaling data, and transmitting the compressed target signaling data.

The application relates to an image reconstruction method and device, computer equipment and a storage medium. The method comprises the following steps of based on the flying time of each photon, obtaining the flying time difference value of coincidence photons; and transmitting the flying time difference value to an image reconstruction machine, and performing image reconstruction. According to the image reconstruction method and the device, the computer equipment and the storage medium disclosed by the invention, through the method comprising the following steps of based on the flying time of each photon, obtaining the flying time difference value of the coincidence photons, transmitting the flying time difference value to the image reconstruction machine, and performing image reconstruction, the calculation of the flying time difference value is completed before data is transmitted to the image reconstruction machine, and then transmitting the flying time difference value to the image reconstruction machine is only needed, so that the data transmission band width is effectively reduced, the transmission time consumption is shortened, and the image reconstruction speed is accelerated.

The invention discloses an image event detection marking method. The method is used for outputting image event detection data after processing an original image data flow, and comprises the following steps of: obtaining the first frame of image data in the original image data flow; executing the following operations sequentially for each frame of image data after the first frame of image data: replacing dark pixel row data of the frame image by a system alignment code; calculating the residual error among all pixel points of effective pixel rows of the frame image and the corresponding pixel rows of the previous-frame image; according to the range which the residual error belongs to, outputting event detection data of all the pixel points in the current pixel row till all the effective pixel rows of the current frame are processed; and, after all image data of the original image data flow is processed, outputting the first frame of original image data, the system alignment code of other frame images and the event detection data.

The invention discloses a Doppler spectrum imaging method, device and equipment and a storage medium, ultrasonic signals are transmitted at multiple deflection angles, so that each time of ultrasonic scanning can cover the whole area, and sampling windows at any different positions can be selected; the method comprises the steps of receiving an ultrasonic signal, receiving a plurality of echo signals corresponding to the ultrasonic signal, and generating Doppler spectrums corresponding to a plurality of deflection angles in a target region of interest according to the plurality of echo signals, and because the echo signals also correspond to a plurality of deflection angles, a user does not need to manually detect an included angle between a probe and a wave beam; therefore, the measurement accuracy of the hemodynamic characteristics is improved; and finally, determining a target Doppler spectrum corresponding to the optimal deflection angle, and displaying the target Doppler spectrum with the best signal to a user, so that the user can check a clearer Doppler spectrum, and the diagnosis difficulty of the user is reduced.

The invention relates to a solid image sensor and an image sensing method. According to one embodiment, the solid image sensor can comprise a pixel array, an analog-digital converter and an output unit, wherein the pixel array is provided with a plurality of pixel units which are arrayed into rows and lines; each pixel unit is provided with a photo-electricity transforming device for transformingan optical signal into an electrical signal; the analog-digital converter is configured for transforming an analog electrical signal generated by each pixel into a digital signal, thereby obtaining apixel value of each pixel unit; the output unit is configured for outputting pixel values, changing compared with previous-frame pixel values, of the pixel units. According to the solid image sensor and the image sensing method disclosed by the invention, the changed pixel values of the pixels are only transmitted, thereby saving data transmission bandwidth; the frame rate can be improved under the condition of not sacrificing resolution; in addition, the subsequent image processing operation can be simplified, therefore, the solid image sensor and the image sensing method are quietly suitablefor being used in video surveillance, automotive advanced driving assistance systems and the like.

The invention relates to a satellite navigation baseband signal playback device, and an automatic verification system and method. The device is characterized in that one end of a peripheral interfacecontrol chip is connected to an upper computer, and the other end of the peripheral interface control chip is connected to a microprocessor through a peripheral interface; an input end of a USB control interface is connected to the upper computer, and an output end of the USB control interface is connected to the input end of a multi-channel data selector through an IO data interface; the input end of the multi-channel data selector is also connected to an antenna through a radio frequency front end; the output end of the multi-channel data selector is connected to a baseband processor, and the baseband processor is connected to the microprocessor; and a selection control signal of the multi-channel data selector is from the microprocessor. In the invention, through simplifying a function,a satellite navigation baseband signal playback function is only realized, a low-cost satellite navigation baseband signal playback device is realized through a mature, stable and low-cost interfacedata playback board and a chip verification board, and a hardware condition is provided for an automatic test of playback data.

The invention discloses a logistics equipment fault prediction terminal based on edge computing. The prediction terminal comprises a power supply circuit, an ADC detection circuit, a microprocessor circuit, a fault indication circuit and a serial port circuit. The input end of the ADC detection circuit is connected with a sensor arranged on the logistics equipment; the ADC detection circuit is used for collecting operation state data of logistics equipment. The output end of the ADC detection circuit is connected with the data input end of the microprocessor circuit; the fault indication circuit is connected with the data output end of the microprocessor circuit through an I / O interface and is used for displaying fault signals of logistics equipment. The serial port circuit is connected with the corresponding end of the microprocessor circuit and used for transmitting the data processed by the microprocessor circuit to the cloud server; the power supply circuit provides working power supply for all circuits; according to the invention, part or all of computing tasks of a cloud computing center are migrated to the vicinity of a data source of logistics equipment to be executed, so that the data transmission bandwidth is reduced, private data is protected, and the risk of privacy leakage of terminal sensitive data is reduced.

The present application relates to a video playing method, system, electronic equipment and storage medium. The method includes: when playing the first original video picture, collecting local sensory data; according to the local sensory data, determining whether the first device is unmanned; when it is determined that the first device is unmanned, generating an unattended Notification; sending the notification to the second device, so that when the second device receives the notification, the second original video image is degraded to obtain a degraded video image, and Return to the degraded video picture; the degraded picture includes lowering the resolution and / or lowering the frame rate; playing the degraded video picture. By adopting this method, when no one is watching on the first device side, the second device can degrade the image quality of the subsequent video, reducing the video data that needs to be transmitted, so as to better improve the network transmission resources during the video conferencing process occupancy, improve the utilization of network transmission resources.

The invention provides a processing system applied to a binary weight convolutional neural network. The system includes: at least one storage unit for storing data and instructions; at least one control unit for obtaining instructions stored in the storage unit and sending control signals; at least one computing unit for obtaining from the storage unit The node value of one layer in the convolutional neural network and the corresponding binary weight value data are obtained by performing addition and subtraction operations to obtain the node value of the next layer. The system of the invention reduces the data bit width in the calculation process of the convolutional neural network, improves the convolution operation speed, and reduces the storage capacity and work energy consumption.