CN110062378A - Identity identifying method based on channel state information under a kind of gesture scene - Google Patents

Identity identifying method based on channel state information under a kind of gesture scene Download PDF

Info

Publication number
CN110062378A
CN110062378A CN201910300116.XA CN201910300116A CN110062378A CN 110062378 A CN110062378 A CN 110062378A CN 201910300116 A CN201910300116 A CN 201910300116A CN 110062378 A CN110062378 A CN 110062378A
Authority
CN
China
Prior art keywords
data
gesture
movement
channel state
state information
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.)
Pending
Application number
CN201910300116.XA
Other languages
Chinese (zh)
Inventor
苘大鹏
杨武
王巍
玄世昌
吕继光
赵晓宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN201910300116.XA priority Critical patent/CN110062378A/en
Publication of CN110062378A publication Critical patent/CN110062378A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/60—Context-dependent security
    • H04W12/68—Gesture-dependent or behaviour-dependent

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Security & Cryptography (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Software Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Human Computer Interaction (AREA)
  • Social Psychology (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Collating Specific Patterns (AREA)

Abstract

The present invention provides the identity identifying methods based on channel state information under a kind of gesture scene, belong to channel state information application field.The present invention analyzes the gesture motion of people by channel state information and realizes authentication, and carries out feature extraction to movement section set, completes the identification process to gesture motion using SVM classifier.In the feature extraction phases of certification, by adjacent gesture motion section and movement interval section component movement sequence and corresponding characteristic data set is obtained.Finally, the verification process of the identity for different people under various gestures is completed using BP neural network.The present invention can complete efficient and convenient verification process according to the exercise habit of the gesture of different people under gesture scene, dimensionality reduction is carried out to data in the way of Principal Component Analysis combination wavelet threshold function and is gone hot-tempered, reduce the data complexity of overall calculation process, and residual noise information is further eliminated, the identity identifying method for various gestures can be provided in application process.

Description

Identity identifying method based on channel state information under a kind of gesture scene
Technical field
The invention belongs to the application fields of channel state information, mainly propose one kind under gesture scene by channel shape State information realization identity authentication method.
Background technique
With the fast development and prosperity of science and technology building, WiFi signal is widely spread and applies in public life. The authentication of the WiFi signal research hotspot emerging as one, is obtaining such as general fit calculation, human-computer interaction, intrusion detection Etc. multiple fields extensive concern.WiFi signal research is with deployment way is simple, deployed position is hidden, application scenarios range is wide The advantages that general, low in cost.Meanwhile the channel state information in WiFi signal is analyzed for the behavior posture and body of people Part feature, to provide strong data basis for identity identifying technology.
Authentication generally requires through the strategy of high precision the authentication result obtained to the identity of people.In practical life In work, authentication can provide technology to many aspects such as the protection of the privacy of user, personal safety protection and safeguarding of assets It ensures.In recent years, a kind of efficient technological means as unaccommodated action recognition and field of identity authentication correlative study, nothing Channel state information in line local area network is capable of providing the identification method without user's perception, and this mode can be obviously improved use Family experience simultaneously provides the service for life of light and fast to people.Ali et al. proposes a kind of Wikey method, and this method being capable of benefit With miniature gesture motion of channel state information identification people under the conditions of tapping keyboard；Tan et al. proposes a kind of WiFinger Method, this method can successfully identify 8 kinds of gesture motions using channel state information under 93% accuracy rate；Wang and Liu Et al. propose a kind of E-eye method, this method has been successfully completed daily to the families such as wash the dishes, take a shower by channel state information The identification of behavior；Wang and Zou et al. propose a kind of WiHear method, and this method identifies saying for people by channel state information Words behavior and the conversation content of people is analyzed and translated；Xin et al. proposes a kind of Freesense method, this method Realize the authentication procedures of Behavior-based control in activity using k nearest neighbor algorithm indoors.
The present invention propose it is a kind of by channel state information analyze people gesture motion and realize identity authentication method, should Method splits data into movement section set and movement interval section set, and carries out feature extraction to movement section set, answers The identification process to gesture motion is completed with SVM classifier.In the feature extraction phases of certification, by adjacent gesture motion section Interval section component movement sequence and corresponding characteristic data set is obtained with movement.Finally, using BP neural network complete for The verification process of identity of the different people under various gestures.
Summary of the invention
The present invention provides the identity identifying methods based on channel state information under a kind of gesture scene, it is therefore intended that utilizes Channel state information realizes authentication procedures under gesture scene, and promotes entirety by dense convolutional neural networks model The accuracy rate of verification process simultaneously reduces calculating time-consuming.
Method flow provided by the invention as shown in Figure 1, specifically includes the following steps:
1, data acquire: carrying out in radio local network environment, under the los path between receiving end and transmitting terminal more The data acquisition of kind gesture motion passes through the corresponding data of data packet acquisition for capturing the channel state information that receiving end receives.
2, data prediction: since indoor environment can generate a large amount of high frequency environment noise for channel state information, because This subcarrier information directly analyzed in channel state information cannot obtain the relevant data information of very intuitive human body behavior. In order to eliminate to the noise information in initial data, preliminary denoising process is carried out to data by low-pass filter, and Further Data Dimensionality Reduction and denoising are carried out to data using principal component analysis combination wavelet threshold function.
3, it acts Concourse Division: since gesture motion is carried out with periodic formation, movement interval censored data is included in data With non-action interval censored data.For the ease of the progress of characteristic extraction procedure, need by movement section fragmentation procedure to data into Action makees section and acts the segment processing of interval section, and forms corresponding two kinds of data acquisition systems.
4, feature extraction: the data information segmentation of gesture part is particularly significant, because movement section can provide movement Specific features information simultaneously provides corresponding basic data for the classification of motion, and act section be capable of providing gesture motion speed, The motion informations such as gesture motion duration can provide the features such as gesture motion type, movement speed, duration And data set is provided for subsequent authentication.Simultaneously as movement interval of the different people when carrying out same gesture movement Difference, therefore the duration that can use movement interval section is further analyzed as the characteristic information of supplement.
5, classify and authenticate: being divided using characteristic data set of the BP neural network to the gesture motion of representative capacity information Class simultaneously obtains authentication result.
Compared with prior art, present invention has an advantage that
1, the present invention proposes the identity identifying method based on channel state information under a kind of gesture scene.Currently, being based on The correlative study of the authentication of channel state information is largely concentrated and is studied with the gait information to people, to have ignored Identity identifying method under gesture scene.The exercise habit that the present invention is capable of gesture under gesture scene according to different people is complete At efficient and convenient verification process.
2, the present invention by Principal Component Analysis combination wavelet threshold function in the way of to data progress dimensionality reduction and go it is hot-tempered, this Help to reduce the data complexity of overall calculation process, and noise information further remaining in elimination data.
3, the present invention can carry out gesture identification to various gestures, and be carried out further using the classification results of gesture identification Verification process.Therefore, the present invention can provide the identity identifying method for various gestures in application process.
Detailed description of the invention
Fig. 1 is whole processing flow schematic diagram
Fig. 2 is the SVM classifier structural schematic diagram of binary tree
Fig. 3 is the flow chart of BP neural network training process
Specific embodiment
It is for a more detailed description to the present invention below with reference to specific implementation:
1, data acquire: in the collection process of data, if transmitter has m antenna (Tx) and receiver there is n Antenna (Rx), then m × n transmissions links will be formed in communication process, wherein carrying containing 30 sons in every transmissions links Wave information then each includes the data that matrix form is m × n × 30 in the data packet containing channel state information.In this hair In bright, transmitting terminal has 3 transmitting antennas, and receiving end has 1 and receives antenna, therefore channel state information is 1 in communication process × 3 × 30 complex matrix form, and 90 subcarriers are all had in each data packet.
2, denoising: since indoor environment can generate a large amount of high frequency environment noise for channel state information, It is related that the amplitude information of a large amount of subcarrier directly in analysis channel state information cannot obtain very intuitive human body behavior Data information.In the present invention, original channel state information is denoised using 5 rank Butterworth low-pass filters Processing, and the action message with the gesture existing for low frequency form including gesture is retained.
3, principal component analysis: in order to extract the strongest subcarrier data of gesture respond to people, the present invention is used The mode of Principal Component Analysis is analyzed by pretreated channel state information data.If by the channel status of denoising The complex matrix form of information data are as follows:
(1) the column average value of matrix H is sought
(2) average value is subtracted in original matrix, seeks normalized matrix R:
(3) according to normalized matrix, the covariance S of data characteristics is sought:
(4) the characteristic value V and feature vector P for seeking covariance S, characteristic value is arranged from big to small, and by corresponding feature Vector is rearranged in the form of column vector to be formed new matrix and can obtain:
V=[v1,v2,...,v90] (5)
P=[p1,p2,...,p90] (6)
(5) Data Dimensionality Reduction: preceding the 3 of selection matrix P arrange and form new data matrix D:
D=[p1,p2,p3] (7)
4,5 layers of wavelet threshold function ω wavelet threshold function: are applied in the present inventionλEach column data of D is carried out hot-tempered Processing, if ω1Former wavelet coefficient before being carried out for wavelet threshold function, ω2The new wavelet systems obtained after representative function processing Number, sign function is sign function, and n is adjustable adaptability parameter, and the standard deviation of noise is indicated that L represents number to be processed by δ According to signal length, d represents the wavelet transformation number of plies chosen, and λ represents threshold value and λ and ωλForm be respectively as follows:
5, act Concourse Division: the working frequency applied in the present invention is 200Hz, therefore it is 0.05s that window k, which is arranged,.
(1) windowed segments: to each column main component p in matrix Di(i=1,2,3), total duration is set as t, according to Window value k is divided, siRepresent the information of i-th of window, window sum N=t/k.
(2) it setsFor piConceptual data average value, calculate the variance of the data in each window:
(3) average variance of all windows is calculated:
(4) movement segmentation: by the variance E of each windowiWith whole average variance EavgIt is compared.Work as EiCompare EavgHour, By window data SiIt is added in movement interval section set Q.Otherwise, by SiIt is added in movement section set A.
6, feature extraction: in order to which the periodic feature to gesture motion is comprehensively analyzed, in movement section set Q Each element qiThe motion feature for carrying out gesture extracts, and counts the duration of each element in movement interval section set A.
(1) feature extraction of gesture: movement interval censored data q is calculated separatelyiStandard deviation stdi, average value meaniMaximum value maxi, and form characteristic Ui。
(2) gesture identification: as shown in Fig. 2, characteristic data set is identified using the SVM classifier of binary tree structure And obtain the recognition result X of gesture motion.
(3) time-frequency convert: to Article 2 main component p2Using Hilbert transform and obtain corresponding phase data.IfIt represents and passes through transformed band value function, and x (t) isBy the original tape value function that inverse transformation obtains, as convolution Distinctive signal h (t) is 1/ π t.
(4) it calculates moving distance: setting δ and be adjustable scale parameter, f is the frequency of current radio signal, according to p2 Movement section distribution situation and corresponding phase data changing valueCalculate each movement section qiStart/stop time hand Gesture moving distance d:
(5) according to the moving distance d and duration Δ t in movement section, the speed V of gesture is calculated:
(6) component movement sequence: the adjacent set section in time upper front and back is incorporated as with movement interval section One group of continuous motion sequence.If the duration for acting interval section is Δ k, the gesture motion period of the motion sequence is calculated H and gesture motion duration accounting S:
H=Δ k+ Δ t (16)
(7) feature extraction of motion sequence: FiIt represents by i-th group of movement section and 1 adjacent movement spacer region thereafter Between extract feature in the action sequence that forms and the data matrix that is formed, form are as follows:
Fi=[d, V Δ t, S, H] (18)
7, classification and certification: carrying out assorting process using BP neural network, applies ten to the characteristic data set of motion sequence Cross validation is rolled over, and obtains authentication result.

Claims (7)

1. the identity identifying method based on channel state information under a kind of gesture scene, it is characterised in that: the following steps are included:
Step 1: data acquire: being carried out in radio local network environment, under the los path between receiving end and transmitting terminal more The data acquisition of kind gesture motion passes through the corresponding data of data packet acquisition for capturing the channel state information that receiving end receives；
Step 2: data prediction: eliminating to the noise information in initial data, carried out by low-pass filter to data Preliminary denoising process, and application principal component analysis combination wavelet threshold function carries out further Data Dimensionality Reduction to data and goes It makes an uproar processing；
Step 3: movement Concourse Division: gesture motion is carried out with periodic formation, and movement interval censored data and non-action are included in data Interval censored data carries out movement section to data by movement section fragmentation procedure and acts the segment processing of interval section, and Form corresponding two kinds of data acquisition systems；
Step 4: feature extraction: the specific features information of movement section offer movement simultaneously provides corresponding basis for the classification of motion Data, and act section and be capable of providing the features such as gesture motion type, movement speed, duration and be subsequent body Part certification provides data set；Since movement interval of the different people when carrying out same gesture movement is different, it is spaced using movement The duration in section is further analyzed as the characteristic information of supplement；
Step 5: classification and certification: being carried out using characteristic data set of the BP neural network to the gesture motion of representative capacity information Classify and obtains authentication result.
2. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: the acquisition of data described in step 1 is specific as follows: transmitting terminal has 3 transmitting antennas, and receiving end has 1 and receives antenna, Channel state information is 1 × 3 × 30 complex matrix form in communication process, and all has 90 sons in each data packet and carry Wave.
3. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: denoising process described in step 2 is specific as follows: using 5 rank Butterworth low-pass filters to original channel Status information carries out denoising, and protects to the action message with the gesture existing for low frequency form including gesture It stays.
4. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: principal component analysis described in step 2 is specific as follows: setting the complex matrix of the channel state information data by denoising Form are as follows:
(4.1) the column average value of matrix H is sought
(4.2) average value is subtracted in original matrix, seeks normalized matrix R:
(4.3) according to normalized matrix, the covariance S of data characteristics is sought:
(4.4) the characteristic value V and feature vector P for seeking covariance S, characteristic value is arranged from big to small, and by corresponding feature to Amount is rearranged in the form of column vector to be formed new matrix and can obtain:
V=[v1,v2,...,v90] (5)
P=[p1,p2,...,p90] (6)
(4.5) Data Dimensionality Reduction: preceding the 3 of selection matrix P arrange and form new data matrix D:
D=[p1,p2,p3] (7)。
5. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: wavelet threshold function described in step 2 is specific as follows: using 5 layers of wavelet threshold function ωλTo each column data of D Hot-tempered processing is carried out, if ω1Former wavelet coefficient before being carried out for wavelet threshold function, ω2It is obtained after representative function processing new Wavelet coefficient, sign function are sign functions, and n is adjustable adaptability parameter, and the standard deviation of noise is indicated by δ, L represent to The signal length of data is handled, d represents the wavelet transformation number of plies chosen, and λ represents threshold value and λ and ωλForm be respectively as follows:
6. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: movement Concourse Division described in step 3 is specific as follows: the working frequency of application is 200Hz, and setting window k is 0.05s；
(6.1) windowed segments: to each column main component p in matrix Di(i=1,2,3), total duration is set as t, according to window Value k is divided, siRepresent the information of i-th of window, window sum N=t/k；
(6.2) it setsFor piConceptual data average value, calculate the variance of the data in each window:
(6.3) average variance of all windows is calculated:
(6.4) movement segmentation: by the variance E of each windowiWith whole average variance EavgIt is compared, works as EiCompare EavgHour, will Window data SiIt is added in movement interval section set Q, otherwise, by SiIt is added in movement section set A.
7. the identity identifying method based on channel state information under a kind of gesture scene according to claim 1, feature Be: feature extraction described in step 4 is specific as follows: to each element q in movement section set QiCarry out the motion feature of gesture It extracts, and counts the duration of each element in movement interval section set A；
(7.1) feature extraction of gesture: movement interval censored data q is calculated separatelyiStandard deviation stdi, average value meaniMaximum value maxi, and form characteristic Ui；
(7.2) gesture identification: characteristic data set is identified using the SVM classifier of binary tree structure and obtains gesture motion Recognition result X；
(7.3) time-frequency convert: to Article 2 main component p2Using Hilbert transform and corresponding phase data is obtained, ifIt represents and passes through transformed band value function, and x (t) isBy the original tape value function that inverse transformation obtains, as convolution Distinctive signal h (t) is 1/ π t
(7.4) it calculates moving distance: setting δ and be adjustable scale parameter, f is the frequency of current radio signal, according to p2It is dynamic Make section distribution situation and corresponding phase data changing valueCalculate each movement section qiStart/stop time gesture move Dynamic distance d:
(7.5) according to the moving distance d and duration Δ t in movement section, the speed V of gesture is calculated:
(7.6) the adjacent set section in time upper front and back and movement interval section component movement sequence: are incorporated as one The continuous motion sequence of group calculates the gesture motion cycle H of the motion sequence if the duration of movement interval section is Δ k With gesture motion duration accounting S:
H=Δ k+ Δ t (16)
(7.7) feature extraction of motion sequence: FiIt represents by i-th group of movement section and 1 adjacent movement interval section group thereafter At action sequence in extract feature and the data matrix that is formed, form are as follows:
Fi=[d, V Δ t, S, H] (18).
CN201910300116.XA 2019-04-15 2019-04-15 Identity identifying method based on channel state information under a kind of gesture scene Pending CN110062378A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910300116.XA CN110062378A (en) 2019-04-15 2019-04-15 Identity identifying method based on channel state information under a kind of gesture scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910300116.XA CN110062378A (en) 2019-04-15 2019-04-15 Identity identifying method based on channel state information under a kind of gesture scene

Publications (1)

Publication Number Publication Date
CN110062378A true CN110062378A (en) 2019-07-26

Family

ID=67319091

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910300116.XA Pending CN110062378A (en) 2019-04-15 2019-04-15 Identity identifying method based on channel state information under a kind of gesture scene

Country Status (1)

Country Link
CN (1) CN110062378A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110738130A (en) * 2019-09-21 2020-01-31 天津大学 Gait recognition method with independent path based on Wi-Fi
CN111596564A (en) * 2020-05-19 2020-08-28 哈尔滨工程大学 Smart home management system based on WiFi gesture recognition
CN112350748A (en) * 2020-10-20 2021-02-09 华北电力大学 PLC channel impulse noise detection method and system by utilizing chi-square distribution
CN112765550A (en) * 2021-01-20 2021-05-07 重庆邮电大学 Target behavior segmentation method based on Wi-Fi channel state information
CN113300750A (en) * 2021-05-24 2021-08-24 南京邮电大学 Personnel identity authentication and handwritten letter identification method based on WIFI signal

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120172682A1 (en) * 2005-12-21 2012-07-05 Norconnect Inc. Method and apparatus for biometric analysis using eeg and emg signals
CN106899968A (en) * 2016-12-29 2017-06-27 南京航空航天大学 A kind of active noncontact identity identifying method based on WiFi channel condition informations

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120172682A1 (en) * 2005-12-21 2012-07-05 Norconnect Inc. Method and apparatus for biometric analysis using eeg and emg signals
CN106899968A (en) * 2016-12-29 2017-06-27 南京航空航天大学 A kind of active noncontact identity identifying method based on WiFi channel condition informations

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
DAPENG MAN ET-AL: "PWiG:A Phase-based Wireless Gesture", 《2018 INTERNATIONAL CONFERENCE ON COMPUTING, NETWORKING AND COMMUNICATIONS (ICNC)》 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110738130A (en) * 2019-09-21 2020-01-31 天津大学 Gait recognition method with independent path based on Wi-Fi
CN111596564A (en) * 2020-05-19 2020-08-28 哈尔滨工程大学 Smart home management system based on WiFi gesture recognition
CN111596564B (en) * 2020-05-19 2022-08-02 哈尔滨工程大学 Smart home management system based on WiFi gesture recognition
CN112350748A (en) * 2020-10-20 2021-02-09 华北电力大学 PLC channel impulse noise detection method and system by utilizing chi-square distribution
CN112350748B (en) * 2020-10-20 2021-06-18 华北电力大学 PLC channel impulse noise detection method and system by utilizing chi-square distribution
CN112765550A (en) * 2021-01-20 2021-05-07 重庆邮电大学 Target behavior segmentation method based on Wi-Fi channel state information
CN112765550B (en) * 2021-01-20 2024-05-07 济南杰睿信息科技有限公司 Target behavior segmentation method based on Wi-Fi channel state information
CN113300750A (en) * 2021-05-24 2021-08-24 南京邮电大学 Personnel identity authentication and handwritten letter identification method based on WIFI signal

Similar Documents

Publication Publication Date Title
CN110062378A (en) Identity identifying method based on channel state information under a kind of gesture scene
CN106658590B (en) Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information
CN106899968A (en) A kind of active noncontact identity identifying method based on WiFi channel condition informations
CN110062379B (en) Identity authentication method based on channel state information under human behavior scene
US20180131554A1 (en) Method, apparatus, server, and systems of time-reversal technology
CN110730473B (en) WiFi activity recognition-oriented signal feature extraction method
CN106599866A (en) Multidimensional user identity identification method
CN107016346A (en) gait identification method and system
CN103093133B (en) Biological identity authentication method facing institute of electrical and electronic engineers (IEEE) 802.15.6
Guo et al. WiReader: Adaptive air handwriting recognition based on commercial WiFi signal
WO2019192253A1 (en) Mobile device-based user identity authentication method and system
EP3623973B1 (en) Unlocking control method and related product
CN104135327A (en) Spectrum sensing method based on support vector machine
CN107947874B (en) Indoor map semantic identification method based on WiFi channel state information
CN109947238B (en) Non-cooperative gesture recognition method based on WIFI
Ding et al. Wihi: WiFi based human identity identification using deep learning
CN113609976A (en) Direction-sensitive multi-gesture recognition system and method based on WiFi (Wireless Fidelity) equipment
CN111142668B (en) Interaction method based on Wi-Fi fingerprint positioning and activity gesture joint recognition
CN112347450A (en) Identity verification method based on blink sound signal
CN114222202A (en) Environment self-adaptive activity detection method and system based on WiFi CSI
Zhu et al. Wi-ATCN: Attentional temporal convolutional network for human action prediction using WiFi channel state information
Man et al. PWiG: A phase-based wireless gesture recognition system
CN111652132B (en) Non-line-of-sight identity recognition method and device based on deep learning and storage medium
Chen et al. WiTT: Modeling and the evaluation of table tennis actions based on WIFI signals
WO2024041053A1 (en) Indoor passive human behavior recognition method and apparatus

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190726