41results about How to "Reduce latency impact" patented technology

The invention discloses a rapid ambiguity determination method among network RTK reference stations of a big-dipper three-frequency signal. The method is characterized by firstly, rapidly determining a big-dipper super-wide lane ambiguity; then, using two carrier-wave-phase non-ionosphere combination observation values and the super-wide lane ambiguity to determine a wide lane ambiguity; and then, using Kalman filtering to calculate zenith troposphere delay correction between the phase non-ionosphere combination observation values and the reference stations; and finally, fixing an original ambiguity. The method comprises the following steps of (1) fixing the super-wide lane ambiguity and the wide lane ambiguity; (2) according to the Kalman filtering, estimating two-reference-station zenith troposphere delay and ambiguity floating point solutions of two groups of phase non-ionosphere observation values; and (3) according to a non-ionosphere combination ambiguity and the wide lane ambiguity, fixing the original ambiguity. In the invention, an advantage of the big-dipper three-frequency signal is fully used and the Kalman filtering is combined so as to realize rapid and accurate fixation of the ambiguity among big dipper network RTK three-frequency reference stations.

The embodiment of the invention provides a training method and device for an acceleration distributed deep neural network. The method comprises the following steps that: on the basis of parallel training, designing the training of the deep neural network into a distributed training pattern, and dividing a deep neural network model to be trained into a plurality of sub-networks; dividing a trainingsample set into a plurality of sub-sample sets; and on the basis of distributed cluster architecture and a preset scheduling method, utilizing the plurality of sub-sample sets to train the deep neural network, wherein each piece of training is carried out by a plurality of sub-networks so as to finish the distributed training of the deep neural network. Since the influence of network delay for the distributed training sub-network can be reduced through data localization on the basis of the distributed cluster architecture and the preset scheduling method, a training strategy is regulated in real time and the progresses of parallel-training sub-networks can be synchronized, the training completion time of the distributed deep neural network can be shortened, and the training of the deep neural network is accelerated.

The invention provides an improved Smith estimator-based H-infinity control method of an aero engine, and belongs to the technical field of aero engine control and simulation. The improved Smith estimator-based H-infinity control method comprises the steps of building a reasonable small deviation linear model for an aero engine non-linear model, and selecting data of a state space model in a certain working condition as a controlled object designed by a controller; selecting an appropriate performance index weighting function parameter, solving an H-infinity output feedback controller, and adjusting the parameter to basically achieve control requirement; and designing an improved-structure Smith estimation compensator according to a closed-loop feedback control system designed according toan H-infinity control law so as to form a composite controller, additionally arranging a deviation correction controller, designed according to a PID control law, in the control system according to relatively large deviation between an estimation model and the parameter of the controlled object and a true model and the parameter, employing the deviation correction controller to stabilize the controlled object, making adaptability correction according to comparison between the controlled object and a model output signal so as to further improve robustness of the system.

Disclosed herein is a data transfer circuit including, a plurality of data transfer lines, a plurality of data outputting sections, a plurality of data holding sections, a data-acquiring-clock supplying section a clock supplying section, and a column scan section.

The invention provides a bypass blocking method, device, system, and electronic device. The bypass blocking method comprises the steps that a bypass blocking device receives a mirror image report of predetermined message from a distributing device, wherein the distributing device is used for distributing the downstream data of an Internet service provider (ISP network); the bypass blocking devicedetermines whether blocking is required or not based on the mirror image report of the predetermined message, and generates a blocking report having the same destination address as the predetermined message, and sends the blocking report to a plurality of core routers. The invention is advantageous in that the disposition difficulty and cost of bypass blocking can be reduced, and the time-delay demand of the bypass blocking can be guaranteed to a certain degree.

The invention discloses an information prompting method and an information prompting device in a mobile terminal. The method comprises the following steps of: establishing a prompting box corresponding to an operation result of a current process; displaying the established prompting box to a user; recording the display time of the prompting box and detecting a closing indication of a user continuously; and when the close indication of the user is not detected and the display time reach a display time threshold, closing the prompting box. By using the method and the device, the operation burden of a user can be lightened; and delayed impact on other user operations can be reduced.

The invention discloses a subharmonic detection method based on cascade delay signal elimination method, comprising steps of obtaining a,b,c three-phase voltage signals of a power grid load terminal through a voltage sensor and a digital regulation circuit, utilizing a Clark transform to obtain a voltage signal under two-phase static coordinate system vabc=[va,vb,vc]<T>, (2) performing filtering process based on the delay signal elimination on the V alpha Beta which is obtained through transformation to obtain voltage signals which only contains odd- number harmonic component and voltage signal which only contains the even-number harmonic component, (3) performing filtering processing based on the cascade delay signal elimination on the voltage signals which only contain the odd-number harmonic wave component and the voltage signals which only contain the even-number harmonic component to obtain the components of positive sequence and negative sequence of any subharnonic, and (4) performing superposition on the harmonic components to obtain a total harmonic signal. The invention is high in detection accuracy and high a practicability.

The invention discloses a method for carrying out high-precision relative navigation based on GNSS through employing a single receiver, and the method comprises the following steps: 1, making preparation; 2, determining the position of the receiver and a clock error at an initial moment; 3, searching a public visible satellite at a current epoch moment and a former epoch moment; 4, building a carrier epoch single-difference model; 5, setting an initial estimation value and a calculation constant at a current moment; 6, calculating the position of the receiver and the clock error at the current moment; 7, judging whether the calculation result meets the requirements or not; 8, updating a position estimation value and a clock error estimation value 9, judging whether the current moment is the last epoch or not. The carrier epoch single-difference model is built through the above flow, and a single-point positioning method is used as assistance. The method achieves the high-precision navigation through a relative positioning method, and the method is implemented through the single receiver, so the method has low requirements, is low in cost, and facilitates the popularization and application.

The invention provides a four-frequency cycle slip detection and repairing method based on linear combination optimization. Four linear combinations are optimized for cycle slip detection and repairing with the maximum fixation probability as a target on basis of four-frequency observation data. Current cycle slip detection and repairing methods are mainly based on three-frequency observation values, and there is no cycle slip detection and repairing method based on four-frequency observation values. The four-frequency cycle slip detection and repairing method fills up the blank of cycle slipdetection and repairing based on four-frequency observation data and has great significance in giving full play to information redundancy brought by the four-frequency observation data and further improving the positioning accuracy.

The invention provides a three-frequency cycle slip detection and restoration method based on optimal fixed probability, the maximum fixed probability is treated as a target, preferably, three sets oflinear combinations are optimized, and the probability of three-frequency cycle slip detection and restoration is improved. In the existing method, three conditions of long-wave length, low combinednoise and low ionized layer delay are considered respectively, and due to the fact that the three conditions cannot be met at the same time, the existing method is mainly of optimizing a plurality ofsets of linear combinations for cycle slip detection and restoration in a qualitive way, and the cycle slip fixed probability is not high. According to the method in the invention, the three constraints do not need to be considered independently, the three constraints are converted into fixed probabilities, the maximum fixed probability is treated as the target, and linear combination optimizationis carried out, so that the three-frequency cycle slip detection and restoration probability is the maximum, and great significance is brought to fully exerting the three-frequency observation data advantage and finally improving the positioning precision.

The application relates to a decryption method and a circuit, and a corresponding device. The method for data decryption comprises receiving (32), over an AXI bus (10) operating in burst mode, data access requests for data units stored in a memory (MC), subdividing the requests received into requests for encrypted data units and requests for non-encrypted data units, forwarding both requests for encrypted data units and requests for non-encrypted data units towards the memory (MC), retrieving the respective sets of data units over the AXI bus (20), and applying Advanced Encryption Standard, AES, processing (60, 70, 80) to the requests for encrypted data units by calculating (72) decryption masks for the encrypted data units and applying (80) the decryption masks calculated to the encrypted data units retrieved, and subdividing the requests into requests for encrypted data units and requests for non-encrypted data units is performed depending on data start addresses and security information conveyed by the requests.

The invention discloses a content proofreading method and device, a mobile terminal and a storage medium, and the method comprises the steps: proofreading a to-be-proofreading text through a plurality of word banks and rules, and generating a plurality of proofreading results; and sequentially arranging the plurality of proofreading results according to a preset logic and generating an arrangement result, and then revising the to-be-proofreading text according to the arrangement result to complete content proofreading. By adopting the embodiment of the invention, the proofreading accuracy can be improved.

The invention relates to a measurement method and device, computer equipment and a storage medium. The measurement method comprises the following steps: respectively transmitting measurement indication information to each piece of edge UE; the measurement indication information is used for indicating the edge UE to perform physical cell identification code PCI measurement on each adjacent cell to be measured; wherein the measurement indication information comprises downlink control information (DCI) which adopts bits to indicate the measurement purpose and the measurement sequence of each round; the measurement purpose comprises the starting frequency point and the measurement number of the round, and the measurement sequence comprises the measurement of the frequency points from the starting frequency point to the completion of the measurement number; and receiving a measurement report transmitted by the edge UE; wherein the measurement report contains the PCI measurement result. According to the method, the situation that part of frequency points cannot be measured and measurement is repeated is avoided, the power consumption of the terminal is reduced, and the influence of an ANR function on terminal time delay is reduced.

An embodiment of the invention provides an activation method and device, a control device, network equipment and an optical network system. A new system ONU is activated on a first OLT of an existingsystem, then the first OLT sends activation information of the new system ONU to a second OLT of the new system or an activation agent ONU activated by the second OLT sends the activation informationto the second OLT, and the activation of the ONU of the new system on the second OLT is realized. Therefore, the DAW channel is only used for communication between the new system ONU and the first OLT, and the problem of OLT signal attenuation of the existing system due to use of an optical splitter is prevented. Therefore, according to the method and the device disclosed in the scheme provided bythe embodiment of the invention, the delay influence on the service when the ONU is activated can be reduced, the communication effect of the existing system cannot be influenced by the existence ofthe optical splitter, and the communication performance of the PON is enhanced.