Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021042639 - DATA PACKET VALIDITY CONFIRMATION METHOD BASED ON EDGE COMPUTING AND DISCRETE RANDOM CONVOLUTION

Publication Number WO/2021/042639
Publication Date 11.03.2021
International Application No. PCT/CN2019/129458
International Filing Date 27.12.2019
IPC
H04W 12/08 2009.01
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
WWIRELESS COMMUNICATION NETWORKS
12Security arrangements, e.g. access security or fraud detection; Authentication, e.g. verifying user identity or authorisation; Protecting privacy or anonymity
08Access security
H04W 12/06 2009.01
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
WWIRELESS COMMUNICATION NETWORKS
12Security arrangements, e.g. access security or fraud detection; Authentication, e.g. verifying user identity or authorisation; Protecting privacy or anonymity
06Authentication
H04B 7/0413 2017.01
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
BTRANSMISSION
7Radio transmission systems, i.e. using radiation field
02Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas
04using two or more spaced independent antennas
0413MIMO systems
G06K 9/62 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
CPC
G06K 9/6268
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
H04B 7/0413
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
BTRANSMISSION
7Radio transmission systems, i.e. using radiation field
02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
04using two or more spaced independent antennas
0413MIMO systems
H04W 12/08
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
WWIRELESS COMMUNICATION NETWORKS
12Security arrangements; Authentication; Protecting privacy or anonymity
08Access security
Applicants
  • 电子科技大学 UNIVERSITY OF ELECTRONIC SCIENCE AND TECHNOLOGY OF CHINA [CN]/[CN]
  • 南方电网科学研究院有限责任公司 LECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO., LTD. [CN]/[CN]
Inventors
  • 谢非佚 XIE, Feiyi
  • 许爱东 XU, Aidong
  • 文红 WEN, Hong
  • 蒋屹新 JIANG, Yixin
  • 张宇南 ZHANG, Yunan
  • 徐鑫辰 XU, Xinchen
Agents
  • 成都巾帼知识产权代理有限公司 CHENGDU JINGUO INTELLECTUAL PROPERTY AGENCY CO., LTD.
Priority Data
201910832461.804.09.2019CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) DATA PACKET VALIDITY CONFIRMATION METHOD BASED ON EDGE COMPUTING AND DISCRETE RANDOM CONVOLUTION
(FR) PROCÉDÉ DE CONFIRMATION DE VALIDITÉ DE PAQUET DE DONNÉES BASÉ SUR UN CALCUL INFORMATISÉ EN PÉRIPHÉRIE ET UNE CONVOLUTION ALÉATOIRE DISCRÈTE
(ZH) 基于边缘计算和离散随机卷积的数据包合法性确认方法
Abstract
(EN)
Disclosed is a data packet validity confirmation method based on edge computing and discrete random convolution. The method comprises the following steps: pre-storing an original pilot signal in an edge server and a known terminal device; the known terminal device inserting a known original pilot signal into a sent signal and sending same to the edge server; the edge server performing pilot separation on the received signal to obtain a received pilot signal matrix; calculating an estimated value of a channel matrix; for the known terminal device, measuring a set of estimated values of a plurality of channel matrices; for different known terminal devices, measuring a corresponding channel matrix estimation set to construct a training set; establishing a convolution kernel and a convolution kernel moving rule, and performing training to obtain a mature neural network classifier; measuring a channel matrix estimation set of a terminal device to be verified; and classifying channel matrices of the terminal device to be verified. The present invention improves a recognition effect of a classifier constructed by a convolutional neural network in a MIMO channel matrix, and improves the accuracy of recognition.
(FR)
Un procédé de confirmation de validité de paquet de données basé sur un calcul informatisé en périphérie et une convolution aléatoire discrète est divulgué. Le procédé fait appel aux étapes suivantes : le pré-stockage d'un signal pilote d'origine dans un serveur périphérique et un dispositif terminal connu; le dispositif terminal connu insérant un signal pilote d'origine connu dans un signal envoyé et l'envoyant au serveur périphérique; le serveur périphérique effectuant une séparation pilote sur le signal reçu afin d'obtenir une matrice de signal pilote reçue; calculant une valeur estimée d'une matrice de canal; pour le dispositif terminal connu, la mesure d'un ensemble de valeurs estimées d'une pluralité de matrices de canaux; pour différents dispositifs terminaux connus, la mesure d'un ensemble d'estimation de matrice de canal correspondant afin de construire un ensemble d'apprentissage; l'établissement d'un noyau de convolution et une règle de déplacement de noyau de convolution, et la réalisation d'un apprentissage afin d'obtenir un classificateur de réseau neuronal mature; la mesure d'un ensemble d'estimation de matrice de canal d'un dispositif de terminal à vérifier; et la classification de matrices de canal du dispositif de terminal à vérifier. La présente invention améliore l'effet de reconnaissance d'un classificateur construit par un réseau neuronal convolutif dans une matrice de canal MIMO, et améliore la précision de la reconnaissance.
(ZH)
本发明公开了一种基于边缘计算和离散随机卷积的数据包合法性确认方法,包括以下步骤:在边缘服务器与已知终端设备中预存原始导频信号;已知终端设备在发送信号中插入已知的原始导频信号发送给边缘服务器;边缘服务器对接收信号进行导频分离,得到接收到的导频信号矩阵;计算信道矩阵的估计值;对于已知终端设备,测得多个信道矩阵的估计值的集合;对于不同的已知终端设备,测得对应的信道矩阵估计的集合,构建训练集合;建立卷积核与卷积核移动规则,训练得到成熟的神经网络分类器;测得待验证终端设备信道矩阵估计的集合;对待验证终端设备的信道矩阵进行分类。本发明提高了卷积神经网络构建的分类器在MIMO信道矩阵中识别效果,提高识别的准确率。
Also published as
Latest bibliographic data on file with the International Bureau