CN109302532A - A kind of identity identifying method based on smart phone acceleration transducer - Google Patents
A kind of identity identifying method based on smart phone acceleration transducer Download PDFInfo
- Publication number
- CN109302532A CN109302532A CN201811301498.XA CN201811301498A CN109302532A CN 109302532 A CN109302532 A CN 109302532A CN 201811301498 A CN201811301498 A CN 201811301498A CN 109302532 A CN109302532 A CN 109302532A
- Authority
- CN
- China
- Prior art keywords
- gait
- vector
- data
- axis
- acceleration transducer
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
- H04M1/72463—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions to restrict the functionality of the device
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
- H04M1/72454—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to context-related or environment-related conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2250/00—Details of telephonic subscriber devices
- H04M2250/12—Details of telephonic subscriber devices including a sensor for measuring a physical value, e.g. temperature or motion
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Environmental & Geological Engineering (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of identity identifying methods based on smart phone acceleration transducer, belong to field of identity authentication.The data that mobile phone acceleration sensor generates are collected first and are pre-processed, obtain Euclidean distance curve, gait cycle division is carried out as the periodical of rule is presented in gait cycle according to it, and the corresponding gait vector of each gait cycle is calculated, each gait vector is compared to the Euclidean distance for calculating the two vectors with a upper gait vector.If Euclidean distance is recorded as continuous similar gait vector less than 2.5, by the two gait vectors, and number is added one；It takes the last one gait vector as the gait vector template generated when number reaches 6, and is compared with current gait vector template, authentication is carried out according to comparing result.If authenticating successful gait vector number reaches threshold value set by user, the authentication success of user.Computation complexity of the present invention is low, has outstanding anti-attack ability, user-friendly.
Description
Technical field
The invention belongs to field of identity authentication, are related to mobile phone acceleration sensor, especially a kind of to be added based on smart phone
The identity identifying method of velocity sensor.
Background technique
With smart phone in people's lives universal and it is widely used, safety of the people than more focusing on mobile phone in the past
Property.However traditional identification scheme can not meet various needs well, such as numerical ciphers and picture password, by
It is frequently necessary to continually use mobile phone in user and most of times used is not long, continually carrying out authentication can make
People is sick of, while this kind of password is easily stolen to peep or guess and cracks or attack by social engineering, and correlative study shows super
The user for crossing 40% does not use this kind of identification scheme in their smart phone.
Compared to traditional identification scheme, based on the authentication method of user biological feature, such as finger print identifying, iris is recognized
It demonstrate,proves with high security, without advantages such as memories.However authentication is carried out based on this kind of biological characteristic, user is required every time
Explicit operation carries out act of authentication, brings inconvenience to user to a certain extent, therefore occurs based on gait, using habit
The identification research of the biological characteristics such as used, key features.Multiple sensors built in smart phone, as acceleration transducer,
Direction sensor etc. provides possibility for this identity authentication scheme.Identification such as is carried out using gait, by sensor
The analysis of data can complete identification procedure in the case where leaving user alone, shorten the process of authentication.
Gait feature is the feature that people generates in walking movement, and the pioneer of scientific gait analysis is Aristotle
Research: the gait of animal.As everyone possesses a secondary unique face, everyone also possesses a kind of distinguished
Gait.From anatomical angle analysis, the physical basis of gait uniqueness is the otherness of everyone physiological structure, different legs
Bone length, different muscle strengths etc. have codetermined the uniqueness of gait.
The existing research that authentication is carried out using acceleration transducer data, is divided into two classes according to authentication method, the
One kind is to divide continuous acceleration transducer data with the period of walking, and uses machine learning algorithm after extracting feature, such as
Support Vector Machine (SVM), K-Nearest-Neighbours (KNN) etc. carry out authentication.Second class is
Continuous acceleration transducer data are compared using dynamic time warping (DTW) to carry out body with trained template
Part certification.
During carrying out identification using biological characteristic, need to carry out feature extraction to the initial data of acquisition,
User biological feature is portrayed with the feature extracted, the result of feature extraction will directly affect the most termination of identification
Fruit.However existing Feature Selection Algorithms are often based upon experience or experiment is attempted to get, it not only can not be right from biological meaning
Feature explains, and the feature of redundancy brings computation burden to real-time identity authentication.This makes these methods in mobile phone
Upper completion real-time identity authentication becomes difficult.
Summary of the invention
In view of the above-mentioned problems, the invention proposes a kind of identity identifying method based on smart phone acceleration transducer,
Authentication is implicitly carried out in the case where leaving user alone.
It mainly comprises the steps that
Step 1: collecting the mobile phone acceleration sensor in certain time period t when certain user's carrying mobile phone is walked and producing
Raw data are simultaneously pre-processed, and Euclidean distance curve is obtained；
Specifically:
Firstly, being directed to certain time period t, mobile phone acceleration sensor generates continuous three number of axle evidence of several groups, by every group of number
According to being respectively synthesized as sub- resultant acceleration；
For data A group, the sub- resultant acceleration calculation formula of this group of data is as follows:
rAFor the sub- resultant acceleration of data A group；xAIndicate the numerical value in the acceleration transducer X-axis of data A group；yAIt indicates
Numerical value in the acceleration transducer Y-axis of data A group；zAIndicate the numerical value on the acceleration transducer Z axis of data A group；
The respective sub- resultant acceleration of the several groups data of time period t is synthesized together, conjunction continuous in time is formed and adds
Speed rt；
rt={ r1,r2,....rA,...}；
Then, to continuous resultant acceleration rtIt is pre-processed, obtains Euclidean distance curve；
Specifically: from continuous resultant acceleration rtAll groups of sub- resultant acceleration data of middle interception the 1st second, will as benchmark
Reference data is continuously slided to the right, the Euclidean distance of the resultant acceleration data of calculating benchmark data and corresponding position, to obtain
One Euclidean distance curve.
Originating corresponding position is 1, and Euclidean distance is the distance between 1 and reference data；Be followed successively by 2 with reference data+1 it
Between distance, and so on.
Step 2: gait cycle division is carried out according to Euclidean distance curve as the periodicity of rule is presented in gait cycle,
And calculate the corresponding gait vector g of each gait cyclev；
Each correspondence of each gait cycle one indicates the gait vector of behavior on foot, includes several in each gait cycle
Continuous three number of axle evidence of group；
Gait vector gvCalculation formula is as follows:
gv=(x1,x2,x3,x4,x5,x6,x7)
x1~x7For all groups in current gait cycle of acceleration transducer value xt, ytAnd ztStatistics value；
Specifically: x1For in three axis data groups all in current gait cycle, the upper quartile of acceleration transducer Z axis
Number；x2For in three axis data groups all in current gait cycle, the degree of bias of acceleration transducer X-axis；x3For in current gait cycle
In all three axis data groups, the kurtosis of acceleration transducer X-axis；x4To add in three axis data groups all in current gait cycle
The kurtosis of velocity sensor Y-axis；x5It is three number of axle all in current gait cycle according to the resultant acceleration being combined into；x6Currently to walk
In the state period in all three axis data groups, the upper quartile of acceleration transducer X-axis；x7It is in current gait cycle all three
In axis data group, the average value of acceleration transducer X-axis.
Step 3 successively selects each gait vector in time period t, by each gait vector and a upper gait vector
It is compared the Euclidean distance for calculating the two vectors.
Step 4: judging Euclidean distance whether less than 2.5, if it is, the two gait vectors are recorded as continuous phase
Add one like gait vector, and by number；Otherwise, recording continuous similar gait vector number is 0.
Continuous similar gait vector number initial value is 0.
Step 5: then taking the last one gait vector as the step generated when the number of continuous similar gait vector reaches 6
State vector template.
Step 6: newly-generated gait vector and current gait vector template are compared, carried out according to comparing result
Authentication.
When current gait vector template initial value is that the number of continuous similar gait vector reaches 6, the last one gait to
Amount.
Specifically: the Euclidean distance for calculating newly-generated gait vector and the gait vector template obtained judges this
Whether distance if it is, newly-generated gait vector authenticate successfully, enters step seven less than 2.5；Otherwise authentification failure.
Step 7: being updated using newly-generated gait vector to gait vector template, and return step six；
Update gait vector template formula are as follows:
tn=0.8*t0+0.2*v
t0For the gait vector template before update；tnFor the gait vector template after update；V is for updating gait
The gait vector of vector template.
Step 8: judging whether the successful gait vector number of certification reaches threshold value set by user, if it is, user
Authentication success, otherwise, the authentication of user fails.
The advantages and positive effects of the present invention are:
(1) a kind of identity identifying method based on smart phone acceleration transducer, authentication calculation amount are small；Carry out
During authentication, complicated calculating is not introduced, so that authentication can be with real-time perfoming simultaneously without increasing significantly
Big mobile phone power consumption amount.
(2) a kind of identity identifying method based on smart phone acceleration transducer, implicit authentication；It can be in user
Authentication is completed when on foot explicitly to operate without necessarily user, it is convenient for users.
(3) a kind of identity identifying method based on smart phone acceleration transducer, it is highly-safe；It ensure that user makes
Under the premise of convenient, there is outstanding anti-attack ability, it can not be effective attacker is trained professional person
Pass through imitate user complete attack.
Detailed description of the invention
Fig. 1 is a kind of identity identifying method flow chart based on smart phone acceleration transducer of the present invention；
Fig. 2 is the gait cycle schematic diagram that the present invention divides；
Fig. 3 is 7 feature ranking results of the generation gait vector that the present invention selects；
Fig. 4 is influence of the Characteristic Number to accuracy rate in gait vector of the present invention；
Fig. 5 is that the present invention updates influence schematic diagram of two coefficient difference values to result in gait vector template；
Fig. 6 is whether gait feature vector template of the present invention is updated Contrast on effect；
Fig. 7 is that walking detection sensor of the present invention triggers schematic diagram；
Specific embodiment
Below in conjunction with attached drawing, the present invention is described in further detail.
In the prior art, the authentication procedures main flow based on user behavior is as follows: requiring participant to carry first
Mobile phone completes movement as defined in certain, such as walks, running, collects several mobile phone sensor data in the process；Then will
The sensing data being collected into is divided into several segments, chooses several features, for each segment, calculates its corresponding feature
Characteristic value, then using these characteristic values come the characteristics of portraying participant.It will receive carry out authentication when
New data are equally divided into several fragment computations characteristic values, judge new data by comparing the difference condition of new and old characteristic value
Whether from the same user to completing authentication.
In the process, crucial step is how to portray user, that is, which feature is selected to be calculated.Sorry
It is that existing method is similar in selected characteristic, what is often chosen is that these are common statistically for average value, variance etc.
Index, often got according to the observation in the experience of researcher or daily life, it is each during actual authentication
Influence of a feature for last authentication result is also unknown, it is common practice to choose several features and authenticate then sight
It examines as a result, simply adjusting feature by result.
This choosing method has the disadvantage that.1) inanimate object meaning.Lack specific biological meaning and also mean that and does not know
Why road will select this feature, only be aware of the specific biological meaning of the feature of selection, just can know that it is which movement
By different people to distinguishing, why these features can be distinguished people.2) computation burden.In the mistake for carrying out identification
Cheng Zhong, calculation amount are the factors having to take into account that.Identification system is realized on mobile phone, is often required that and is recognized in real time
Card.And it is limited to the computing capability of mobile phone local, excessively complicated feature will will affect calculating speed.In addition, more existing
Occur the feature of redundancy in method, it is not necessary to feature lead to unnecessary calculating.3) lack safety.For authentication
For, the attack tolerant of feature has to consider, if some features are easy to be imitated, in identification
Cheng Zhong, attacker can complete attack by imitating the movement of user to increase the success attack rate of oneself.For such spy
Sign, should be prudent when choosing.
Based on considerations above, through ten people of analysis when on foot, the right hand holds the acceleration that mobile phone generates and passes the present invention
Sensor data are evaluated various features from many aspects and are chosen validity feature generation gait vector, propose one kind and be based on
The identity identifying method of smart phone acceleration transducer mainly includes that gait vector generates, gait vector template generates, gait
Several processes such as the certification of vector and the update of gait vector template.
As shown in Figure 1, mainly comprising the steps that
Step 1: collecting the mobile phone acceleration sensor in certain time period t when certain user's carrying mobile phone is walked and producing
Raw data are simultaneously pre-processed, and Euclidean distance curve is obtained；
The walking of user's carrying mobile phone, acceleration transducer are generated data in real time, are then authenticated using real-time data,
The speed that acceleration transducer data generate is not fixed, and is probably that can produce within 1 second tens groups, each group of data all include three
It counts (x, y, z), each group of three numbers is calculated, so that it may obtain a sub- resultant acceleration, resultant acceleration is that a string of son conjunctions add
Speed is the data of Time Continuous.Specifically:
Firstly, being directed to certain time period t, mobile phone acceleration sensor generates continuous three number of axle evidence of several groups, by every group of number
According to being respectively synthesized as sub- resultant acceleration；
For data A group, the sub- resultant acceleration calculation formula of this group of data is as follows:
rAFor the sub- resultant acceleration of data A group；xAIndicate the numerical value in the acceleration transducer X-axis of data A group；yAIt indicates
Numerical value in the acceleration transducer Y-axis of data A group；zAIndicate the numerical value on the acceleration transducer Z axis of data A group；
The respective sub- resultant acceleration of the several groups data of time period t is synthesized together, conjunction continuous in time is formed and adds
Speed rt；
rt={ r1,r2,....rA,...}；
Then, to continuous resultant acceleration rtIt is pre-processed, obtains Euclidean distance curve；
Specifically: from continuous resultant acceleration rtMiddle interception time is all groups of sub- resultant acceleration data of 1S as base
Standard continuously slides to the right reference data, the Euclidean distance of the resultant acceleration data of calculating benchmark data and corresponding position, thus
Obtain an Euclidean distance curve.
Originating corresponding position is 1, and Euclidean distance is the distance between 1 and reference data；Be followed successively by 2 with reference data+1 it
Between distance, and so on.For example 1000 groups of sub- resultant acceleration data are shared in certain time period t, 200 groups are produced in first second
Data, then just using this 200 groups data as benchmark, then by this reference data and 1-200,2-201,3-
202 ... 799-1000 are compared, and are compared 799 times, this process is known as sliding.
Step 2: gait cycle division is carried out according to Euclidean distance curve as the periodicity of rule is presented in gait cycle,
And calculate the corresponding gait vector g of each gait cyclev；
The present invention uses existing period division methods, when dividing the period on foot, uses the synthesis of three axis of sensor
Value come measure a people walk gait.Periodicity extraction schematic diagram is as shown in Figure 2, it can be seen that, a people accelerates when on foot
Spend three axis of sensor composite value present it is apparent periodically, for people in the case where stablizing and walking line state, the time of a step is 0.5-
0.8S.To guarantee that the information in gait vector template is complete, the data conduct that a segment length is 1S is had chosen in the centre of data
Template.By template respectively forwardly, move backward then calculation template and data Euclidean distance, it can be seen that obtained knot
Fruit is presented periodically, to carry out period division, then carries out the disposal of gentle filter to result, Filtering Formula sees below formula, and M is filter
Wavelength degree is selected as 4 in the present embodiment.
The wave crest and trough of calculated result waveform diagram obtain the period on foot, to calculate gait vector.
In gait vector generation phase, each gait cycle can generate corresponding one gait for indicating behavior on foot
Vector, interior each gait cycle includes continuous three number of axle evidence of several groups；
For the present embodiment through ten people of analysis when on foot, the right hand holds the acceleration transducer data that mobile phone generates, right
Various features are evaluated, as shown in figure 3,7 features has been selected to generate gait vector:
gv=(x1,x2,x3,x4,x5,x6,x7)
x1~x7For all groups in current gait cycle of acceleration transducer value xt, ytAnd ztStatistics value；Select this 7
Influence of a feature to accuracy rate, as shown in Figure 4.
Specifically: x1For in three axis data groups all in current gait cycle, the upper quartile of acceleration transducer Z axis
Number；x2For in three axis data groups all in current gait cycle, the degree of bias of acceleration transducer X-axis；x3For in current gait cycle
In all three axis data groups, the kurtosis of acceleration transducer X-axis；x4To add in three axis data groups all in current gait cycle
The kurtosis of velocity sensor Y-axis；x5It is three number of axle all in current gait cycle according to the resultant acceleration being combined into；x6Currently to walk
In the state period in all three axis data groups, the upper quartile of acceleration transducer X-axis；x7It is in current gait cycle all three
In axis data group, the average value of acceleration transducer X-axis.
Step 3 successively selects each gait vector in time period t, by each gait vector and a upper gait vector
It is compared the Euclidean distance for calculating the two vectors.
Step 4: judging Euclidean distance whether less than 2.5, if it is, the two gait vectors are recorded as continuous phase
Add one like gait vector, and by number；Otherwise, recording continuous similar gait vector number is 0.
Continuous similar gait vector number initial value is 0.
Step 5: then taking the last one gait vector as the step generated when the number of continuous similar gait vector reaches 6
State vector template.
When carrying out authentication, need to obtain a gait feature vector in advance as template, new data are come
When be compared with template data to obtaining authentication result.When calculating gait feature template, it is desirable that user carries out
On foot to generate template.After obtaining data, continuous data are subjected to gait cycle division, calculate its corresponding gait
Feature vector, therefore a series of gait feature vectors can be obtained.When the value of gait feature vector gradually settles out, with regard to obtaining
Gait feature template: by calculating the distance between each gait vector and a upper gait vector, if continuous 6
As soon as gait vector is less than given threshold value at a distance from upper gait vector, using the last one gait vector as gait vector
Template.
Step 6: newly-generated gait vector and current gait vector template are compared, carried out according to comparing result
Authentication.
When current gait vector template initial value is that the number of continuous similar gait vector reaches 6, the last one gait to
Amount.
Specifically: the Euclidean distance for calculating newly-generated gait vector and the gait vector template obtained judges this
Whether distance if it is, newly-generated gait vector authenticate successfully, enters step seven less than 2.5；Otherwise authentification failure.
During generating gait feature template, a thresholding variables Threshold is introduced to measure two steps
Whether state feature vector is alike enough.When the Euclidean distance of two gait feature vectors be less than this threshold value when think the two to
Amount is probably from the same user.
The present invention determines this threshold value by considering the variation degree of the gait feature vector of the same user.In walking plus
The middle section data of velocity sensor data are relatively stable, therefore take out among everyone all gait feature vectors
100 gait feature vectors calculate the average distance between vector two-by-two.For the success for guaranteeing identification, threshold value should be slightly larger
In user itself gait feature vector average distance and be less than different user between gait feature vector distance.It sets accordingly
Threshold value Threshold is 2.5.
Step 7: being updated using newly-generated gait vector to gait vector template, and return step six；
People's gait when on foot can tend towards stability, however from the perspective of for a long time, these gait features may
As the time gradually changes.Therefore it in order to adapt to this variation preferably so as to chronically carry out identification, needs
A kind of update mechanism is wanted to guarantee that algorithm can adaptive change at any time.It is proposed to this end that a kind of more new algorithm is special for gait
Levy the update of vector.Core concept is if one-time identity authentication behavior passes through, i.e. the user of mobile phone generates new gait
Feature vector, then this new gait feature vector just represents the newest gait situation of user to a certain extent and becomes
Gesture, therefore can use newest gait feature vector and gait feature is updated so as to original gait feature vector template
Vector template more adapts to newest state.
Update gait vector template formula are as follows:
tn=0.8*t0+0.2*v
t0For the gait vector template before update；tnFor the gait vector template after update；V is for updating gait
The gait vector of vector template.Wherein 0.8,0.2 selection reason is as shown in Figure 5.
Step 8: judging whether the successful gait vector number of certification reaches threshold value set by user, if it is, user
Authentication success, otherwise, the authentication of user fails.
In order to show the necessity of update, by displaying authentication twice as a result, as shown in fig. 6, abscissa indicates
Data, ordinate indicate the distance of gait feature vector and gait feature vector template to be tested, and solid line is to use more new algorithm
It is updated, dotted line is without using more new algorithm.It can as seen from the figure, after having used more new algorithm, due to gait feature
The updating ability of vector template, when constantly receiving new data, if meeting the condition updated, gait feature vector mould
Plate can independently be updated, so that next new data are more in line with, so as to preferably carry out identification.
It is realized in Android mobile phone using test authentication power consumption condition.As it is desirable that the case where leaving user alone
Lower carry out authentication, so program needs always in running background, however the operation of lasting monitoring sensor will increase consumption
Electricity, therefore used walking detection sensor as Trigger Function.As shown in fig. 7, walking detection sensor is by Android
The sensor that can be triggered when each user walks provided, it is less compared to ordinary sensors power consumption.When walking detects
After sensor is triggered, start the data for collecting acceleration transducer, if no longer through walking detection sensor after a period of time
It is triggered, then program just restores dormant state, continues etc. to be triggered.
Claims (3)
1. a kind of identity identifying method based on smart phone acceleration transducer, which is characterized in that mainly comprise the steps that
Step 1: collecting what the mobile phone acceleration sensor in certain time period t generated when certain user's carrying mobile phone is walked
Data are simultaneously pre-processed, and Euclidean distance curve is obtained；
Specifically:
Firstly, being directed to certain time period t, mobile phone acceleration sensor generates continuous three number of axle evidence of several groups, by every group of data point
Sub- resultant acceleration is not synthesized；
For data A group, the sub- resultant acceleration calculation formula of this group of data is as follows:
rAFor the sub- resultant acceleration of data A group；xAIndicate the numerical value in the acceleration transducer X-axis of data A group；yAIndicate data A
Numerical value in the acceleration transducer Y-axis of group；zAIndicate the numerical value on the acceleration transducer Z axis of data A group；
The respective sub- resultant acceleration of the several groups data of time period t is synthesized together, resultant acceleration continuous in time is formed
rt；
rt={ r1,r2,....rA,...}；
Then, to continuous resultant acceleration rtIt is pre-processed, obtains Euclidean distance curve；
Step 2: carrying out gait cycle division, and count according to Euclidean distance curve as the periodicity of rule is presented in gait cycle
Calculate the corresponding gait vector g of each gait cyclev；
Each correspondence of each gait cycle one indicates the gait vector of behavior on foot, includes that several groups connect in each gait cycle
Three continuous number of axle evidences；
Gait vector gvCalculation formula is as follows:
gv=(x1,x2,x3,x4,x5,x6,x7)
x1~x7For all groups in current gait cycle of acceleration transducer value xt, ytAnd ztStatistics value；
Step 3 successively selects each gait vector in time period t, each gait vector and a upper gait vector is carried out
Compare the Euclidean distance for calculating the two vectors；
Step 4: judging Euclidean distance whether less than 2.5, if it is, the two gait vectors are recorded as continuous similar step
State vector, and number is added one；Otherwise, recording continuous similar gait vector number is 0；
Step 5: when the number of continuous similar gait vector reaches 6, then take the last one gait vector as generation gait to
Measure template；
Step 6: newly-generated gait vector and current gait vector template are compared, identity is carried out according to comparing result
Certification；
Specifically: the Euclidean distance for calculating newly-generated gait vector and the gait vector template obtained judges this distance
Whether less than 2.5, if it is, newly-generated gait vector authenticates successfully, otherwise authentification failure；
Step 7: being updated using newly-generated gait vector to gait vector template, and return step six；
Update gait vector template formula are as follows:
tn=0.8*t0+0.2*v
t0For the gait vector template before update；tnFor the gait vector template after update；V is for updating gait vector
The gait vector of template；
Step 8: judging whether the successful gait vector number of certification reaches threshold value set by user, if it is, the body of user
Part authenticates successfully, and otherwise, the authentication of user fails.
2. a kind of identity identifying method based on smart phone acceleration transducer as described in claim 1, which is characterized in that
The method of Euclidean distance curve is obtained in the step 1 specifically: from continuous resultant acceleration rtMiddle interception the 1st second all
The sub- resultant acceleration data of group are continuously slided to the right reference data, the conjunction of calculating benchmark data and corresponding position adds as benchmark
The Euclidean distance of speed data, to obtain an Euclidean distance curve；
Originating corresponding position is 1, and Euclidean distance is the distance between 1 and reference data；It is followed successively by between 2 and reference data+1
Distance, and so on.
3. a kind of identity identifying method based on smart phone acceleration transducer as described in claim 1, which is characterized in that
The step two specifically: x1For in three axis data groups all in current gait cycle, upper four points of acceleration transducer Z axis
Digit；x2For in three axis data groups all in current gait cycle, the degree of bias of acceleration transducer X-axis；x3For current gait cycle
In interior all three axis data groups, the kurtosis of acceleration transducer X-axis；x4For in three axis data groups all in current gait cycle,
The kurtosis of acceleration transducer Y-axis；x5It is three number of axle all in current gait cycle according to the resultant acceleration being combined into；x6It is current
In gait cycle in all three axis data groups, the upper quartile of acceleration transducer X-axis；x7To own in current gait cycle
In three axis data groups, the average value of acceleration transducer X-axis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811301498.XA CN109302532B (en) | 2018-11-02 | 2018-11-02 | Identity authentication method based on smart phone acceleration sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811301498.XA CN109302532B (en) | 2018-11-02 | 2018-11-02 | Identity authentication method based on smart phone acceleration sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109302532A true CN109302532A (en) | 2019-02-01 |
| CN109302532B CN109302532B (en) | 2020-02-07 |
Family
ID=65145758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811301498.XA Active CN109302532B (en) | 2018-11-02 | 2018-11-02 | Identity authentication method based on smart phone acceleration sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109302532B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110968857A (en) * | 2019-12-03 | 2020-04-07 | 南京航空航天大学 | Smart watch identity authentication method based on arm lifting action |
| CN112966248A (en) * | 2021-03-23 | 2021-06-15 | 西安电子科技大学 | Continuous identity authentication method of mobile equipment in uncontrolled walking scene |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103442114A (en) * | 2013-08-16 | 2013-12-11 | 中南大学 | Identity authentication method based on dynamic gesture |
| CN106850955A (en) * | 2016-12-20 | 2017-06-13 | 陕西尚品信息科技有限公司 | A kind of mobile phone identity verification method based on Gait Recognition |
| CN107103297A (en) * | 2017-04-20 | 2017-08-29 | 武汉理工大学 | Gait identification method and system based on mobile phone acceleration sensor |
| JP6193539B2 (en) * | 2012-01-17 | 2017-09-06 | 日本信号株式会社 | Human body communication card holder and RFID tag authentication system |
| CN107273726A (en) * | 2017-06-02 | 2017-10-20 | 中国人民解放军信息工程大学 | Equipment owner's identity real-time identification method and its device based on acceleration cycle variation law |
| CN107831907A (en) * | 2017-12-07 | 2018-03-23 | 深圳先进技术研究院 | Identity identifying method and device based on Gait Recognition |
-
2018
- 2018-11-02 CN CN201811301498.XA patent/CN109302532B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6193539B2 (en) * | 2012-01-17 | 2017-09-06 | 日本信号株式会社 | Human body communication card holder and RFID tag authentication system |
| CN103442114A (en) * | 2013-08-16 | 2013-12-11 | 中南大学 | Identity authentication method based on dynamic gesture |
| CN106850955A (en) * | 2016-12-20 | 2017-06-13 | 陕西尚品信息科技有限公司 | A kind of mobile phone identity verification method based on Gait Recognition |
| CN107103297A (en) * | 2017-04-20 | 2017-08-29 | 武汉理工大学 | Gait identification method and system based on mobile phone acceleration sensor |
| CN107273726A (en) * | 2017-06-02 | 2017-10-20 | 中国人民解放军信息工程大学 | Equipment owner's identity real-time identification method and its device based on acceleration cycle variation law |
| CN107831907A (en) * | 2017-12-07 | 2018-03-23 | 深圳先进技术研究院 | Identity identifying method and device based on Gait Recognition |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110968857A (en) * | 2019-12-03 | 2020-04-07 | 南京航空航天大学 | Smart watch identity authentication method based on arm lifting action |
| CN112966248A (en) * | 2021-03-23 | 2021-06-15 | 西安电子科技大学 | Continuous identity authentication method of mobile equipment in uncontrolled walking scene |
| CN112966248B (en) * | 2021-03-23 | 2022-10-21 | 西安电子科技大学 | Continuous identity authentication method of mobile equipment in uncontrolled walking scene |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109302532B (en) | 2020-02-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | Accelerometer-based speed-adaptive gait authentication method for wearable IoT devices | |
| Muaaz et al. | Smartphone-based gait recognition: From authentication to imitation | |
| Wu et al. | Liveness is not enough: Enhancing fingerprint authentication with behavioral biometrics to defeat puppet attacks | |
| CA2835460C (en) | System and method for enabling continuous or instantaneous identity recognition based on physiological biometric signals | |
| CN107831907A (en) | Identity identifying method and device based on Gait Recognition | |
| CN107103297A (en) | Gait identification method and system based on mobile phone acceleration sensor | |
| Qin et al. | A fuzzy authentication system based on neural network learning and extreme value statistics | |
| CN105184254B (en) | A kind of identity identifying method and system | |
| JP2003248828A (en) | Identification method, program, and device | |
| WO2018152711A1 (en) | Electrocardiographic authentication-based door control system and authentication method therefor | |
| CN108629167A (en) | A kind of more smart machine identity identifying methods of combination wearable device | |
| Xu et al. | Challenge-response authentication using in-air handwriting style verification | |
| CN106411952A (en) | Telekinetic-dynamic-gesture-based user identity authentication method and apparatus | |
| CN105530095B (en) | A kind of method and device of user identity authentication | |
| CN109302532A (en) | A kind of identity identifying method based on smart phone acceleration transducer | |
| Karnan et al. | A model to secure mobile devices using keystroke dynamics through soft computing techniques | |
| Derawi | Smartphones and biometrics: Gait and activity recognition | |
| Sim et al. | Identity recognition based on bioacoustics of human body | |
| CN107273728B (en) | Smart watch unlocking and authentication method based on motion sensing behavior characteristics | |
| Fan et al. | Cancelable HD-SEMG biometric identification via deep feature learning | |
| Wu et al. | Toward robust detection of puppet attacks via characterizing fingertip-touch behaviors | |
| Meharia et al. | The human key: Identification and authentication in wearable devices using gait | |
| Lu et al. | Fmcode: A 3d in-the-air finger motion based user login framework for gesture interface | |
| CN109145548A (en) | User authentication device and method and computer readable recording medium based on multi-mode | |
| Derawi et al. | Fusion of gait and ECG for biometric user authentication |
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 |