CN114520975B - Lightweight passive identity authentication system and method based on wireless network - Google Patents
Lightweight passive identity authentication system and method based on wireless network Download PDFInfo
- Publication number
- CN114520975B CN114520975B CN202210058405.5A CN202210058405A CN114520975B CN 114520975 B CN114520975 B CN 114520975B CN 202210058405 A CN202210058405 A CN 202210058405A CN 114520975 B CN114520975 B CN 114520975B
- Authority
- CN
- China
- Prior art keywords
- data
- csi
- cloud server
- identity authentication
- ghostnet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000010801 machine learning Methods 0.000 claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000012216 screening Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 10
- 230000005021 gait Effects 0.000 claims description 10
- 238000013480 data collection Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 239000000470 constituent Substances 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000013461 design Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 4
- 230000008447 perception Effects 0.000 abstract description 2
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/06—Authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/40—Security arrangements using identity modules
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Computer Security & Cryptography (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention belongs to the technical field of identity authentication, and provides a lightweight passive identity authentication system and method based on a wireless network. The invention finally provides theoretical basis and practical experience for the application of the machine learning technology in the fields of passive perception identity authentication and identification by combining the technologies of signal processing, machine learning and the like, and the designed system provides new design ideas and application prototypes for the application of passive identification and tracking, indoor intrusion detection, company sign-in and the like. The invention designs and develops a lightweight passive identity authentication system based on machine learning. And preprocessing the WiFi signal to extract the CSI characteristics, and adopting an algorithm based on an SNN+GhostNet network to realize high-precision real-time authentication of the identity. The developed system is hopeful to fill the blank of passive identity authentication application, and creates an application example for passive identification and positioning based on machine learning. The system can be widely applied to the fields of identification and tracking, indoor intrusion detection, company check-in and the like.
Description
Technical Field
The invention belongs to The technical field of identity authentication, and relates to a lightweight passive identity authentication system and method based on a wireless network.
Background
Identity authentication is a key technology in pervasive computing and man-machine interaction. The existing identity authentication technology is mainly divided into two methods, namely active authentication and passive authentication. The active identity authentication method requires additional special equipment such as cameras and multifunctional sensors, and the equipment cost is high. In addition, such methods often require accurate recognition by the user in proximity to the sensing device, such as iris recognition, gesture recognition, etc., not only present privacy concerns, but also do not cover large areas. Passive identity authentication methods, particularly machine learning based methods, typically require a large number of data samples to ensure high accuracy authentication. Furthermore, this approach tends to result in reduced accuracy when crossing environments. In order to solve the problems of the method, the invention takes Jiang Pushi WiFi signals as a research object, fully extracts rich characteristics from limited WiFi data by using a lightweight machine learning method, and realizes passive identity authentication in a rapid mode.
WiFi signals have ubiquitous, fine-grained characteristics. After the WiFi signal sent by the transmitting end of the COTS equipment is transmitted through paths in various modes such as signal scattering, signal direct incidence and signal reflection, signals with various paths overlapped together, namely multipath overlapped signals, are formed at the receiving end of the COTS equipment. The signal, affected by the actual physical space environment during propagation, contains information that is representative of the characteristics of the environment. The environment referred to herein is the physical space in which signals propagate, including both human factors (human position, characteristics, posture, motion, etc.) and other external factors. In this way, the corresponding environment information is modulated in the WiFi signal propagation process, the environment information can be obtained by demodulating the information at the receiving end, and the behavior characteristics of people in the perceived environment are also included, so that the scientificity of realizing high-precision person identification by utilizing the WiFi signal is ensured in principle.
Disclosure of Invention
The invention aims to solve the technical problem of how to use the WiFiCSI (Channel State Information ) of a physical layer to realize authentication of human identity in a lightweight machine learning mode. The invention finally provides theoretical basis and practical experience for the application of the machine learning technology in the fields of passive perception identity authentication and identification by combining the technologies of signal processing, machine learning and the like, and the designed system provides new design ideas and application prototypes for the application of passive identification and tracking, indoor intrusion detection, company sign-in and the like. The invention aims to provide a lightweight identity authentication method based on machine learning.
The technical scheme of the invention is as follows:
a lightweight passive identity authentication system based on a wireless network comprises a COTS equipment transmitting end, a COTS equipment receiving end, a cloud server and a control end;
the COTS equipment transmitting end and the COTS equipment receiving end are both provided with wireless network cards, and the wireless network cards are arranged in a room at intervals; the COTS equipment transmitting end is used for transmitting the WiFi signal carrying the CSI information, and the COTS equipment receiving end is used for receiving the WiFi signal in the environment and uploading the WiFi signal to the cloud server;
the cloud server comprises a data collection module, a preprocessing module and an authentication module; the data collection module collects CSI data of a user and user actions by using COTS equipment; the preprocessing module uses a low-pass filter to reduce noise of the CSI data collected by the data collection module, screens the antenna and the CSI stream, and generates CSI characteristic data; the authentication module trains, verifies and tests the preprocessed CSI characteristic data by using a machine learning method, and completes identity authentication and action recognition through a trained network model;
the control end is a webpage or mobile phone user end, is communicated with the COTS equipment and the cloud server, and sends basic information of the user to the cloud server to monitor and early warn the identity of the user in real time.
A lightweight passive identity authentication method based on a wireless network comprises the following steps:
step one, a user enters a room, a transmitting end of COTS equipment transmits WiFi signals, the user walks to cause disturbance to the WiFi signals, at the moment, a receiving end of the COTS equipment collects and extracts four pieces of gait CSI data of the user, and the gait CSI data are uploaded to a cloud server; simultaneously, the control end is used for transmitting the basic information of the user to the cloud server;
step two, the cloud server pre-processes the CSI data in the step one, and then models the pre-processed data by using a machine learning algorithm;
the WiFi signal preprocessing comprises the following specific steps:
(1) Denoising data: removing high-frequency noise with the frequency exceeding 80Hz by using a Butterworth filter;
(2) And (3) antenna screening: the limited data samples are increased with the CSI stream in each piece of CSI data as the smallest data unit. The COTS equipment is provided with 3 antennas, the correlation coefficient of the CSI flow between the antennas is calculated to evaluate the correlation of the antennas, 1 antenna with the lowest correlation is removed, and 2 antennas with higher correlation are reserved;
(3) CSI stream screening: respectively calculating the amplitudes and variances of all the CSI streams on each antenna, and screening the CSI streams with low amplitudes and high variances;
the machine learning algorithm comprises the following specific steps:
(1) Snn+ghostnet network structure: in order to extract complete characteristics from limited data and rapidly realize identity recognition, the invention combines SNN (Siamese Neural Network, twin Network) and GhostNet (Ghost Network) together to design a SNN+GhostNet Network structure. The SNN+GhostNet network structure includes four modules, namely GhostNet, global average pool (Global Average Pooling) and Euclidean distance (Euclidean Distance), sigmoid regression, and Cross-entcopy function modules. Wherein the GhostNet module is a core part;
the SNN framework groups individual data in the limited data into data pairs to further increase the data sample size. The GhostNet module contains 2 CNN (Convolutional NeuralNetwork ) units and 1GhostNet unit. The first CNN unit performs data sampling, collates input data, and the second CNN unit adjusts data dimension so as to facilitate subsequent processing, and the GhostNet unit is used for extracting complete deep action characteristics;
(2) Global average pool to euclidean distance: and uniformly receiving and retaining all the extracted characteristic information by adopting a global average pooling layer, and reducing the dimension of the data into characteristic vectors. Then calculating Euclidean distance of the feature vector in the feature space;
(3) Sigmoid regression: calculating a normalized Euclidean distance by using a Sigmoid regression method, and mapping the normalized Euclidean distance to (0, 1);
(4) Cross-entropy function: minimizing normalized Euclidean distances from the same class to approach 0 and maximizing normalized Euclidean distances from different classes to approach 1 using a Cross-entopyloss function;
and step three, when the user enters the room again, the receiving end of the COTS equipment collects gait CSI data again and uploads the gait CSI data to the cloud server. And the cloud server performs character identity matching verification on the CSI data by using the trained model, displays the character identity and sends the character identity to the control end, and when the character information is not found in the database, the cloud server sends alarm information to the control end.
The invention has the beneficial effects that: lightweight passive identity authentication systems based on machine learning are designed and developed. And preprocessing the WiFi signal to extract the CSI characteristics, and adopting an algorithm based on an SNN+GhostNet network to realize high-precision real-time authentication of the identity. The developed system is hopeful to fill the blank of passive identity authentication application, and creates an application example for passive identification and positioning based on machine learning. The system can be widely applied to the fields of identification and tracking, indoor intrusion detection, company check-in and the like.
Drawings
Fig. 1 is a schematic structural diagram of a cloud server.
In the figure: 1COTS equipment receiving end; a 2COTS device transmitting end; 3, users; 4, a data collection module; 5, a pretreatment module; 6, an authentication module; 7SNN framework; 8, denoising the data; 9, screening an antenna; 10CSI stream screening; 11GhostNet network architecture; 12 global average pool and euclidean distance; 13Sigmoid regression; 14 Cross-entopy function.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the following technical schemes (and accompanying drawings).
A lightweight passive identity authentication system based on a wireless network comprises a COTS equipment transmitting end, a COTS equipment receiving end, a control end and a cloud server;
firstly, a user enters a room for the first time, a WiFi signal is sent by a COTS equipment transmitting end, the WiFi signal is obviously disturbed by the user walking, and at the moment, four pieces of gait CSI data of the user are collected and extracted by a COTS equipment receiving end and uploaded to a cloud server; simultaneously, the control end is used for transmitting the basic information of the user to the cloud server;
step two, the cloud server pre-processes the CSI data in the step one, and then models the pre-processed data by using a machine learning algorithm;
the WiFi signal preprocessing comprises the following specific steps:
(1) Denoising data: the linear interpolation algorithm is applied to the passive sensing field, small gaps in the CSI data are filled, the small gaps are uniformly distributed in a time domain, and a Butterworth filter is utilized to remove high-frequency noise with the frequency exceeding 80 Hz;
(2) And (3) antenna screening: the limited data samples are increased with the CSI stream in each piece of CSI data as the smallest data unit. The COTS equipment is provided with 3 antennas, the correlation coefficient of the CSI flow between the antennas is calculated to evaluate the correlation of the antennas, 1 antenna with the lowest correlation is removed, and 2 antennas with higher correlation are reserved;
wherein ,is the correlation coefficient between the ith and jth CSI streams,/>For covariance between ith and jth CSI streams, +.> and />Respectively representing standard deviations of an ith CSI stream and a jth CSI stream;
(3) CSI stream screening: respectively calculating the amplitudes and variances of all the CSI streams on each antenna, and screening the CSI streams with low amplitudes and high variances;
the amplitude of the CSI stream may be replaced by the maximum value of the set of discrete data. And taking the maximum value of each group of CSI stream data as an abscissa and the variance as an ordinate, and making a scatter diagram in a two-dimensional coordinate system. Then defining an optimal center point P as follows:
wherein Maximum i And Variance j Representing the maximum value and variance of the corresponding CSI streams;
the center point P is a point with the minimum value of the maximum values of all CSI stream data on each antenna as the abscissa and the maximum value of the variance as the ordinate. Calculating Euclidean distances between other points and a central point P, and reserving CSI streams corresponding to the first 6 points with the minimum Euclidean distances according to experience;
the machine learning algorithm comprises the following specific steps:
(1) Snn+ghostnet network structure: in order to extract complete characteristics from limited data and rapidly realize identification, the invention combines an SNN framework and a GhostNet together and designs an SNN+GhostNet network structure. The SNN+GhostNet network structure comprises four modules, namely GhostNet, global average pool and Euclidean distance, sigmoid regression and Cross-entopy function modules. Wherein the GhostNet module is a core part;
the SNN framework groups individual data in the limited data into data pairs to further increase the data sample size, and defines the label Y of the data pairs as:
wherein ya and yb For constituent data pairs X a and Xb Is a label of (a).
Then pair X data a and Xb Input to the GhostNet module. It contains two CNN units and 1GhostNet unit. The first CNN unit performs data sampling, collates input data, the second CNN unit adjusts data dimension so as to facilitate subsequent processing, and the GhostNet unit is used for rapidly extracting deep action features;
wherein G(Xa) and G(Xb ) To obtain a characteristic diagram;
(2) Global average pool to euclidean distance: uniformly receiving and retaining all extracted characteristic information by adopting a global average pooling layer, and reducing the dimension of data into characteristic vectors;
calculated at A (X a) and A(Xb ) Euclidean distance of feature vectors in feature space;
Euc(X a ,X b )=||A(X a )-A(X b )||. (6)
(3) Sigmoid regression: calculation of normalized Euclidean distance using Sigmoid regression method, will Euc (X a ,X b ) Mapping to (0, 1);
(4) Cross-entropy function: minimizing S (X) from the same class using Cross-entopyloss function a ,X b ) Approaching it to 0, maximizing S (X) from different classes a ,X b ) Making the temperature approach to 1;
L(X a ,X b ,Y)=(1-Y)log(1-S(X a ,X b ))+Ylog S(X a ,X b )+λω, (8)
wherein ω is L 2 A regulater, λ representing a regularization coefficient;
empirically, a threshold of 0.2 is set to determine whether the input data is from the same category or from a different category.
And thirdly, when the user enters the room again, the receiving end of the COTS equipment collects gait CSI data again and uploads the gait CSI data to the cloud server, the cloud server verifies the physical identity of the person with the trained model, the physical identity is displayed and is sent to the control end, and if the physical information does not exist in the database, the cloud server sends alarm information to the control end.
Claims (1)
1. The lightweight passive identity authentication method based on the wireless network is characterized by being realized through a lightweight passive identity authentication system based on the wireless network, wherein the lightweight passive identity authentication system comprises a COTS device transmitting end, a COTS device receiving end, a cloud server and a control end;
the COTS equipment transmitting end and the COTS equipment receiving end are both provided with wireless network cards, and the wireless network cards are arranged in a room at intervals; the COTS equipment transmitting end is used for transmitting the WiFi signal carrying the CSI information, and the COTS equipment receiving end is used for receiving the WiFi signal in the environment and uploading the WiFi signal to the cloud server;
the cloud server comprises a data collection module, a preprocessing module and an authentication module; the data collection module collects CSI data of a user and user actions by using COTS equipment; the preprocessing module uses a low-pass filter to reduce noise of the CSI data collected by the data collection module, screens the antenna and the CSI stream, and generates CSI characteristic data; the authentication module trains, verifies and tests the preprocessed CSI characteristic data by using a machine learning method, and completes identity authentication and action recognition through a trained network model;
the control terminal is a webpage or mobile phone user terminal, is communicated with the COTS equipment and the cloud server, and transmits basic information of a user to the cloud server to monitor and early warn the identity of the user in real time;
the authentication method comprises the following steps:
firstly, a user enters a room for the first time, a WiFi signal is sent by a COTS equipment transmitting end, the WiFi signal is obviously disturbed by the user walking, and at the moment, the COTS equipment receiving end collects and extracts CSI data of four gait of the user and uploads the CSI data to a cloud server; simultaneously, the control end is used for transmitting the basic information of the user to the cloud server;
step two, the cloud server pre-processes the WiFi signals in the step one, and then models the pre-processed data by using a machine learning module;
(1) Denoising data: the linear interpolation algorithm is applied to the passive sensing field, gaps in the CSI data are filled, the gaps are uniformly distributed in a time domain, and a Butterworth filter is utilized to remove high-frequency noise with the frequency exceeding 80 Hz;
(2) And (3) antenna screening: increasing limited data samples by taking the CSI flow in each piece of CSI data as the smallest data unit; the COTS equipment is provided with 3 antennas, the correlation coefficient of the CSI flow between the antennas is calculated to evaluate the correlation of the antennas, 1 antenna with the lowest correlation is removed, and 2 antennas with higher correlation are reserved;
wherein ,is the correlation coefficient between the ith and jth CSI streams,/>For covariance between ith and jth CSI streams, +.> and />Respectively representing standard deviations of an ith CSI stream and a jth CSI stream;
(3) CSI stream screening: respectively calculating the amplitudes and variances of all the CSI streams on each antenna, and screening the CSI streams with low amplitudes and high variances;
the amplitudes of the CSI streams are replaced by the maximum value of the CSI stream data of the corresponding group; taking the maximum value of each group of CSI stream data as an abscissa and the variance as an ordinate, and making a scatter diagram in a two-dimensional coordinate system; then defining an optimal center point P as follows:
wherein, maximum i And Variance j Representing the maximum value and variance of the corresponding CSI streams;
the center point P is a point with the minimum value of the maximum values of all CSI stream data on each antenna as the abscissa and the maximum value of the variance as the ordinate; calculating Euclidean distances between other points and a central point P, and reserving CSI streams corresponding to the first 6 points with the minimum Euclidean distance;
the machine learning algorithm comprises the following specific steps:
(1) Snn+ghostnet network structure: combining the SNN framework and the GhostNet together to form an SNN+GhostNet network structure; the SNN+GhostNet network structure comprises four modules, namely a GhostNet module, a global average pool and Euclidean distance, sigmoid regression and a Cross-entopy function, wherein the GhostNet module is a core part;
the SNN framework groups individual data in the limited data into data pairs to further increase the data sample size, and defines the label Y of the data pairs as:
wherein ,ya and yb For constituent data pairs X a and Xb Is a label of (2);
then pair X data a and Xb The input module is input into a GhostNet module and comprises two CNN units and 1GhostNet unit; the first CNN unit performs data sampling, collates input data, the second CNN unit adjusts data dimension so as to facilitate subsequent processing, and the GhostNet unit is used for rapidly extracting deep action features;
wherein ,G(Xa) and G(Xb ) To obtain a characteristic diagram;
(2) Global average pool to euclidean distance: uniformly receiving and retaining all extracted characteristic information by adopting a global average pooling layer, and reducing the dimension of data into characteristic vectors;
calculated at A (X a) and A(Xb ) Euclidean distance of feature vectors in feature space;
Euc(X a ,X b )=||A(X a )-A(X b )|| (6)
(3) Sigmoid regression: calculation of normalized Euclidean distance using Sigmoid regression method, will Euc (X a ,X b ) Mapping to (0, 1);
(4) Cross-entropy function: minimizing S (X) from the same class using Cross-entopyloss function a ,X b ) Approaching it to 0, maximizing S (X) from different classes a ,X b ) Making the temperature approach to 1;
L(X a ,X b ,Y)=(1-Y)log(1-S(X a ,X b ))+Y logS(X a ,X b )+λω (8)
wherein ω is L 2 A regulater, λ representing a regularization coefficient; setting a threshold value of 0.2 to determine whether the input data is from the same category or different categories;
and thirdly, when the user enters the room again, the receiving end of the COTS equipment collects gait CSI data again and uploads the gait CSI data to the cloud server, the cloud server performs person identity matching verification on the CSI data by using the trained model, the person identity is displayed and sent to the control end, and if the person information does not exist in the database, the cloud server sends alarm information to the control end.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210058405.5A CN114520975B (en) | 2022-01-19 | 2022-01-19 | Lightweight passive identity authentication system and method based on wireless network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210058405.5A CN114520975B (en) | 2022-01-19 | 2022-01-19 | Lightweight passive identity authentication system and method based on wireless network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114520975A CN114520975A (en) | 2022-05-20 |
CN114520975B true CN114520975B (en) | 2023-08-08 |
Family
ID=81597326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210058405.5A Active CN114520975B (en) | 2022-01-19 | 2022-01-19 | Lightweight passive identity authentication system and method based on wireless network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114520975B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108901021A (en) * | 2018-05-31 | 2018-11-27 | 大连理工大学 | A kind of deep learning identification system and method based on channel state information of wireless network |
CN109784212A (en) * | 2018-12-27 | 2019-05-21 | 东莞理工学院 | A kind of biology typewriting recognition methods based on WiFi signal |
CN109891436A (en) * | 2016-10-24 | 2019-06-14 | Lg 电子株式会社 | Security system and its control method based on deep learning neural network |
CN113360874A (en) * | 2021-06-04 | 2021-09-07 | 湖南大学 | User biological characteristic information authentication method and system |
WO2021225191A1 (en) * | 2020-05-07 | 2021-11-11 | 엘지전자 주식회사 | Method and device for generating user identification model on basis of wireless sensing |
-
2022
- 2022-01-19 CN CN202210058405.5A patent/CN114520975B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109891436A (en) * | 2016-10-24 | 2019-06-14 | Lg 电子株式会社 | Security system and its control method based on deep learning neural network |
CN108901021A (en) * | 2018-05-31 | 2018-11-27 | 大连理工大学 | A kind of deep learning identification system and method based on channel state information of wireless network |
CN109784212A (en) * | 2018-12-27 | 2019-05-21 | 东莞理工学院 | A kind of biology typewriting recognition methods based on WiFi signal |
WO2021225191A1 (en) * | 2020-05-07 | 2021-11-11 | 엘지전자 주식회사 | Method and device for generating user identification model on basis of wireless sensing |
CN113360874A (en) * | 2021-06-04 | 2021-09-07 | 湖南大学 | User biological characteristic information authentication method and system |
Non-Patent Citations (1)
Title |
---|
吴习芊.基于WiFi的目标反射信号提取及定位算法.无线互联科技.2020,(第09期),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN114520975A (en) | 2022-05-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gu et al. | Paws: Passive human activity recognition based on wifi ambient signals | |
CN108901021B (en) | Deep learning identity recognition system and method based on wireless network channel state information | |
Yu et al. | Noninvasive human activity recognition using millimeter-wave radar | |
AU2016200905B2 (en) | A system and method for identifying and analyzing personal context of a user | |
CN114423034B (en) | Indoor personnel action recognition method, system, medium, equipment and terminal | |
Ding et al. | Device-free location-independent human activity recognition using transfer learning based on CNN | |
Cai et al. | GBDT‐Based Fall Detection with Comprehensive Data from Posture Sensor and Human Skeleton Extraction | |
Li et al. | Collaborative fall detection using smart phone and Kinect | |
Shang et al. | LSTM-CNN network for human activity recognition using WiFi CSI data | |
Kim et al. | Deep neural network-based indoor emergency awareness using contextual information from sound, human activity, and indoor position on mobile device | |
CN114818788A (en) | Tracking target state identification method and device based on millimeter wave perception | |
Zhu et al. | Wi-ATCN: Attentional temporal convolutional network for human action prediction using WiFi channel state information | |
Moshiri et al. | CSI-based human activity recognition using convolutional neural networks | |
CN113300750A (en) | Personnel identity authentication and handwritten letter identification method based on WIFI signal | |
CN114520975B (en) | Lightweight passive identity authentication system and method based on wireless network | |
Wu et al. | Are you in the line? RSSI-based queue detection in crowds | |
Wang | Data feature extraction method of wearable sensor based on convolutional neural network | |
CN114970640A (en) | Knowledge distillation-based gesture recognition method and system | |
CN108123765A (en) | A kind of personnel's real-time detection method and system | |
CN115393955A (en) | Gesture recognition method and system based on BVP and WiTransformer by utilizing WiFi | |
CN112086165B (en) | Upper limb rehabilitation monitoring method and system based on deep learning | |
CN114764580A (en) | Real-time human body gesture recognition method based on no-wearing equipment | |
CN210899633U (en) | Indoor positioning system based on deep neural network | |
Kaudki et al. | IOT enabled human fall detection using accelerometer and RFID technology | |
Zhang et al. | An indoor human action recognition method based on spatial location information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |