Beijing University of Posts and Telecommunications

National University of Singapore

Beijing Volcano Engine Technology Co., Ltd.

Chinese Academy of Sciences

Yifan Zhang

Ping Zhang

Meng Wang

Zhiyong Feng

Rui Shi

Zhiqing Wei

Qixun Zhang

Sai Huang

Guanghui Zhang

Yancan Mao

Yong Fang

Zhanghao Chen

Tianbai Ma

Yu Cao

Huadong Ma

Xiaolong Zheng

Huiyuan Fu

Yuhao Yang

Yunhao Xie

Desheng Wu

Rui Dai

Fan Ning

Weimin Wang

Yongle Wu

Tianchi Zhao

Dengkui Zhang

Yiwen Wei

Huan Sun

A broadband millimeter wave beam tracking method based on vehicle movement trajectory recognition, which relates to the field of wireless communication. First, a rough estimate of the beam angle is obtained according to the position and attitude of two vehicles in a communication scenario at an initial moment, and a state of the vehicle is judged according to the sudden change rate of the transmitting beam angle starting from a next moment: if it is in the sudden change mode, the observation value ŷof the received signal is calculated; if it is in the smooth change mode, the observation value I is calculated; the observation values in different modes are input into an improved particle filter algorithm for fine beam estimation, to obtain optimal values xand u; and the beam optimal values at the current moment are used to re-estimate a pair of the forming vector of the transmitting beam and the forming vector of the receiving beam to adjust the vehicle and the pilot signal at the next moment is transmitted; by continuously adjusting the beam forming vector at each moment, a beam tracking within an error range is obtained. The present application improves the accuracy and reliability of beam tracking in case of sudden angle change.

Broadband millimeter wave beam tracking method based on vehicle motion trajectory cognition

A service control method based on a service cluster, an electronic device, and a storage medium are described. The method includes: acquiring operation index information generated by a target service cluster when a task runs; calling one or more strategy models in a strategy model library to perform anomaly analysis on the operation index information, and determining target anomaly information of an anomaly occurring in the target service cluster when the task runs and a corresponding target solution strategy; and sending a configuration updating indication carrying the target solution strategy to the target service cluster.

Service control method based on service cluster, electronic device and storage medium

The present disclosure provides a method and an apparatus of resource adjustment for a service cluster, an electronic device and a storage medium. The method includes: determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed; determining a core quantity to be adjusted corresponding to central processor units and the second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and a resource utilization rate of central processor units in each computing node that has been created; and executing the resource adjustment operation according to the core quantity to be adjusted and the second quantity, and obtaining a service cluster subject to resource adjusting.

Method of resource adjustment for service cluster, electronic device and storage medium

The embodiments of the present application provide a method and apparatus for modulation recognition of signals based on cyclic residual network, the method comprises: obtaining a signal matrix of a to-be-recognized signal, and extracting real part information and imaginary part information of the signal matrix; generating, according to extracted real part information and imaginary part information, a real-and-imaginary-part feature matrix of the to-be-recognized signal; converting, according to a preset matrix conversion method, the real-and-imaginary-part feature matrix into an amplitude-and-phase feature matrix; and inputting the amplitude-and-phase feature matrix into a pre-trained cyclic residual network to obtain a modulation mode corresponding to the to-be-recognized signal. In the embodiments of the present application, the processing of the to-be-recognized signal is simple and easy to operate, in which neither complex algorithms nor manual processing is required, the flexibility of recognition is high, and the result of modulation recognition of the to-be-recognized signal can be accurately obtained.

Method and apparatus for modulation recognition of signals based on cyclic residual network

A coupled-line rat-race coupler includes four ports and four coupled-lines each composed of two metal lines. Two of the ports are respectively connected with the front ends of the two lines of the first coupled-line, and the other two ports are respectively connected with the front ends of the two lines of the second coupled-line. The two lines of the third coupled-line are short circuited and respectively connected with the back end of one line of the first coupled-line on one side and with the back end of one line of the second coupled-line on the other side. In the fourth coupled-line, one line is open at one end and short circuited at the other end with the other line, which is connected with one line of the first coupled-line and one line of the second coupled-line. The coupled-line rat-race coupler is characterized with smooth and stable output phase.

Coupled-line rat-race coupler with smooth in-phase and out-of-phase performances

Disclosed is an intelligent information parsing method based on cross-modal data fusion, including: step 1, acquiring information multi-modal data, step 2, processing the multi-modal data, step 3, fusing data features, step 4, establishing and training a parsing model, step 5, parsing fused cross-modal data, and step 6, viewing parsed data.

Intelligent information parsing method based on cross-modal data fusion

Embodiments of the present invention provide an end-to-end lightweight method and apparatus for license plate recognition. The method comprises: obtaining an image to be recognized; obtaining a number of a license plate in the image to be recognized and position coordinates of the license plate in the image to be recognized on the basis of the image to be recognized and a pre-trained target license plate recognition model, wherein the target license plate recognition model comprises a target feature extraction network, a target region candidate localization network, a target super-resolution generation network and a target recurrent neural network. Because in this solution, once an image to be recognized is input into the target license plate recognition model, the target license plate recognition model can output the license plate number and position coordinates of the license plate in the image to be recognized, one realizes an end-to-end model. The model has relatively strong robustness, and it can detect and recognize pictures taken under different camera angles. Moreover, computation variables such as image features can be reused without repeated computations, the model takes up less RAM and the speed of license plate recognition is greatly improved.

End-to-end lightweight method and apparatus for license plate recognition

Growing community of inventors

Beijing, China

Yifan Zhang