CN107194219A - Intelligent terminal identity identifying method based on similarity - Google Patents
Intelligent terminal identity identifying method based on similarity Download PDFInfo
- Publication number
- CN107194219A CN107194219A CN201710459425.2A CN201710459425A CN107194219A CN 107194219 A CN107194219 A CN 107194219A CN 201710459425 A CN201710459425 A CN 201710459425A CN 107194219 A CN107194219 A CN 107194219A
- Authority
- CN
- China
- Prior art keywords
- intelligent terminal
- mrow
- behavioural characteristic
- similarity
- msub
- 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
Links
Classifications
-
- 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
-
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06F21/45—Structures or tools for the administration of authentication
Abstract
The present invention proposes a kind of intelligent terminal identity identifying method based on similarity, including training stage and authentication phase, the technical problem relatively low for solving the certification degree of accuracy present in existing implicit authentication method.Realize that step is：Training dataset is obtained, the training vector set obtained by pretreatment calculates the similarity in training vector set between behavioural characteristic distribution probability vector to train user behavior characteristic model, presets authentication threshold value based on comentropy；If certification password is matched completely, then pretreatment current behavior characteristic sequence obtains current behavior feature distribution probability vector, obtain current average similarity, judge that current average similarity whether in default authentication threshold range, obtains authentication result with the difference percentage of user behavior characteristic model.The present invention is safe, and Consumer's Experience is good, available for authenticating user identifications such as smart mobile phone and flat boards.
Description
Technical field
The invention belongs to identity identifying technology field, it is related to a kind of intelligent terminal identity identifying method, and in particular to a kind of
Intelligent terminal identity identifying method based on similarity, available for authenticating user identifications such as smart mobile phone, flat boards.
Background technology
Authentication be the user of computer system when entering system or accessing the system resource of different protection levels, be
System confirms the process of visitor's identity.The thinking of authentication is mainly：What if the visitor of current computer systems was provided
Authentication information is matched with the authentication information of the validated user existed, then it is assumed that current visitor be it is legal,
Allow it to access operation, otherwise refuse it and access operation.The purpose of authentication be control system to the mandate of user or
Certain right of user is limited, it is the first line of defence of guarantee system safety.
Intelligent terminal as computer system one kind, it has also become one of essential instrument in people's daily life,
The individual privacy information and sensitive data stored thereon is more and more, therefore safety problem is also more and more concerned.To improve
Mainly two major classes are occurred in that in the security of intelligent terminal, existing intelligent terminal identity identifying method：Explicit authentication and implicitly recognize
Card.
Explicit authentication is carried out by matching the authentication information that user is explicitly entered and the authentication information existed to user
Certification, realizes the purpose of user access control.It is more traditional authentication method, mainly includes PIN code certification, password code
Certification, gesture pattern authentication etc..These methods due to it is simple, easy-to-use the features such as be widely used in the authentication of intelligent terminal
In, but there is certain safety defect：Some users set better simply password to be easy to use so that user cipher is easily
Cracked by malicious user；During in public using explicit authentication method to intelligent terminal certification, it is easy to peeped, cause by shoulder
Password is revealed；Article " Smudge Attacks On Smartphone Touch in the safe forums of USENIX in 2010
Screens " points out that user's finger is streaked finger residue when screen is authenticated and can adhered on the touchscreen for a long time, opponent
As long as simple analysis instrument is that can obtain password, so that password is revealed, security is caused to reduce.
Implicit authentication by paired observation to user behavior feature and the behavioural characteristic that has existed, determined to make certification
It is fixed, it is main that user identity is authenticated using the behavioural characteristic during intelligent terminal using user.Behavior used is special
Levying mainly has user's touch screen feature, gesture feature, keystroke characteristic etc., with it is unique, carry-on, facilitate the features such as.Implicit authentication side
The proposition of method effectively improves the drawbacks described above of explicit authentication, is current study hotspot, but it is still deposited also in developing stage
In some defects.Such as Authorization Notice No. is CN 104134028 B, entitled " identity identifying method based on gesture feature and
The patent of invention of system ", discloses a kind of identity identifying method based on gesture feature applied to touch panel device, it passes through
The current gesture of the input of user on the touchscreen is contrasted roughly with multiple advance typing gestures, so as to obtain rough similarity；
If reaching, the current acceleration for inputting each timeslice of gesture, angle, distance are calculated, and judge whether one by one respectively right
Answer the acceleration of the advance typing gesture of timeslice, angle, in the range of the upper and lower bound of distance, and calculated according to judged result
Final similarity；If final similarity reaches the second predetermined threshold value, it is determined as authentication success.The invention is defeated according to user
Enter the difference of the stability of gesture, provide different cryptosecurity grades, but it simply sentences according to the bound scope of gesture
The similarity of disconnected current input gesture and advance typing gesture, the certification degree of accuracy is not high, it is impossible to ensure security.
The content of the invention
It is an object of the invention to overcome defect present in above-mentioned existing implicit authentication technology, it is proposed that one kind is based on phase
Like the intelligent terminal identity identifying method of degree, the skill relatively low for solving the certification degree of accuracy present in existing implicit authentication method
Art problem.
To achieve the above object, the technical scheme that the present invention takes includes training stage and authentication phase, realizes that step is：
Training stage：
(1) intelligent terminal obtains the training dataset of user：
(1a) user preset certification password, inputs intelligent terminal；
Behavioural characteristic sequence during (1b) intelligent terminal record multiple input authentication password of user, forms the instruction of user
Practice data set；
(2) intelligent terminal is pre-processed to training dataset, obtains training vector set：
(2a) intelligent terminal concentrates each behavioural characteristic sequence to be normalized respectively training data：Intelligent terminal is to training
Each characteristic value of each behavioural characteristic sequence is normalized in data set, obtains multiple normalization behavioural characteristic sequences；
(2b) intelligent terminal carries out feature reconstruction respectively to multiple normalization behavioural characteristic sequences：
Interval [0,1] is divided into N number of equal interval, wherein N >=2 by (2b1) intelligent terminal；
(2b2) intelligent terminal is calculated respectively respectively normalizes behavioural characteristic sequence every in multiple normalization behavioural characteristic sequences
The distribution probability value of individual interval upper characteristic value, each normalization behavioural characteristic sequence obtains N number of distribution probability value, by each normalizing
The N number of distribution probability value for changing behavioural characteristic sequence is expressed as vector by interval division order, so as to obtain multiple behavioural characteristics point
Cloth probability vector, constitutes training vector set；
(3) intelligent terminal trains user behavior characteristic model using training vector set：
(3a) intelligent terminal calculates between any two similar of all behavioural characteristic distribution probabilities vector in training vector set
Degree, is obtainedIndividual Similarity value, wherein, R represents behavioural characteristic distribution probability vector in training vector set
Quantity, wherein R >=2, calculating formula of similarity is：
Wherein,And H (Xi) it is behavioural characteristic distribution probability vector Xi={ pi1,pi2,...,piNComentropy
Function, its formula is：
Wherein, N represents interval division sum, pinRepresent behavioural characteristic distribution probability vector XiIn n-th of interval distribution
Probability；
(3b) intelligent terminal pairIndividual Similarity value is averaged, and obtains user behavior characteristic model；
(4) user preset authentication threshold value；
Authentication phase：
(5) the behavioural characteristic sequence in intelligent terminal acquisition user currently inputs certification password and input process；
(6) intelligent terminal judges whether user's current authentication password matches completely with pre- setting authentication password, if so, to current
Behavioural characteristic sequence is pre-processed, and obtains current behavior feature distribution probability vector, and performs step (7), otherwise performs step
Suddenly (5), current behavior characteristic sequence is pre-processed, realizes that step is：
Current behavior characteristic sequence is normalized (6a) intelligent terminal, obtains currently normalizing behavioural characteristic sequence；
(6b) intelligent terminal carries out feature reconstruction to current normalization behavioural characteristic sequence：Intelligent terminal is by interval [0,1]
N number of equal interval is divided into, the distribution probability value of current normalization behavioural characteristic sequence characteristic value on each interval is calculated, will
Distribution probability value is expressed as vector by interval division order, obtains current behavior feature distribution probability vector；
(7) intelligent terminal obtains current average similarity：
(7a) intelligent terminal calculates current behavior feature distribution probability vector and each behavioural characteristic point in training vector set
Similarity between cloth probability vector, obtains R current Similarity values；
(7b) intelligent terminal is averaged to R current Similarity values, obtains current average similarity；
(8) intelligent terminal obtains authentication result：
Intelligent terminal judges whether the difference percentage of current average similarity and user behavior characteristic model is less than or equal to
Default authentication threshold value, if so, then authentication success, otherwise, authentication failure.
The present invention compared with prior art, has the following advantages that：
1st, the present invention is based on because the similarity of feature based trains user behavior characteristic model and carries out authentication
Similarity calculating method there is preferable stability, improve the authentication degree of accuracy, meanwhile, matched completely in certification password
When, implicit authentication is carried out as the assistant authentification factor using the similarity of behavioural characteristic sequence, double authentication is realized, it is and existing
Technology is compared, and is effectively improved the security of authentication.
2nd, the present invention is because the similarity by the use of behavioural characteristic sequence is as the assistant authentification factor, in the behavior of record user
During characteristic sequence, user is without deliberately coordinating, compared with prior art, and Consumer's Experience is good on the premise of authentication security is ensured
It is good.
Brief description of the drawings
Fig. 1 is the implementation process figure of the present invention.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment, the present invention is described in further detail.
The present embodiment is with the touch-screen input certification password progress authentication on the Typical Representative smart mobile phone of intelligent terminal
Exemplified by：
Reference picture 1, the intelligent terminal identity identifying method based on similarity, including training stage and authentication phase, are realized
Step is：
Training stage：
Step 1, smart mobile phone obtain the training dataset of user：
Step 1a) user preset certification password, input smart mobile phone；
Step 1b) the multiple touch-screen input certification password of user collects training dataset, big due to training dataset scale
Small to influence authentication result, scale is too small, and the certification degree of accuracy is not high, and scale is excessive, then computing cost increase, and data acquisition fiber crops
It is tired, therefore the present embodiment takes training dataset scale to contain 5 behavioural characteristic sequences for 5, i.e. training dataset, user needs 5
Secondary touch-screen input certification password, if certification password contains k password character, a behavioural characteristic sequence of the present embodiment is by k-
Interval time, k password character duration, k-1 password character incoming frequency, k between 1 adjacent password character is individual
Password character touch screen pressure these numeric type behavioural characteristics composition, therefore each behavioural characteristic sequence has 4k-2 characteristic value, form
For：(Interval1,Hold1,Freq1,Press1,Interval2,Hold2,Freq2,Press2,...,Intervalk,
Holdk,Freqk,Pressk), smart mobile phone records the behavioural characteristic sequence during 5 touch-screen input certification passwords of user, shape
Into the training dataset of user；
Step 2, smart mobile phone are pre-processed to training dataset, obtain training vector set：
Step 2a) smart mobile phone is to step 1b) in 5 behavioural characteristic sequences concentrating of obtained training data carry out respectively
Normalization, for a behavioural characteristic sequence (Interval1,Hold1,Freq1,Press1,Interval2,Hold2,Freq2,
Press2,...,Intervalk,Holdk,Freqk,Pressk), its normalized implementation method is：Seek its 4k-2 characteristic value
In maximum and minimum value, each characteristic value for normalizing formula pair using most value is normalized, arranged by former order
To sequence be a normalization behavioural characteristic sequence, similarly, concentrate other 4 behavioural characteristic sequences to use training data
The normalized implementation method of identical, is obtained 5 normalization behavioural characteristic sequences；
Step 2b) smart mobile phone is to step 2a) in 5 normalization behavioural characteristic sequences obtaining carry out feature weight respectively
Structure：
Step 2b1) interval [0,1] is divided into N number of equal interval by smart mobile phone, and this interval quantity N can basis
The characteristic value sum 4k-2 of one behavioural characteristic sequence determines that the present embodiment takes N values to be 2k, and same user is to multiple
When normalizing behavioural characteristic sequence progress feature reconstruction, interval quantity N is identical；
Step 2b2) smart mobile phone is respectively to each normalization behavioural characteristic sequence statistic in 5 normalization behavioural characteristic sequences
Its distribution number of 4k-2 characteristic value on each interval, because interval quantity is 2k, so respectively normalization behavioural characteristic sequence
It can count and obtain 2k distribution number, to obtained distribution number difference divided by the characteristic value total number 4k- of a behavioural characteristic sequence
2, obtain distribution probability value prn, r=1,2 .., 5, n=1,2 .., 2k, prnRepresent behavioural characteristic sequence XrIt is interval at n-th
Distribution probability, by N number of distribution probability value p of each normalization behavioural characteristic sequencern, n=1,2 .., 2k is by interval division order
It is expressed as vectorial Xr={ pr1,pr2,...,pr2k, that is, obtain 5 behavioural characteristic distribution probability vectors, composition training vector set P
={ X5, wherein, 5 be the quantity of behavioural characteristic distribution probability vector in training vector set；
Step 3, smart mobile phone train user behavior characteristic model using training vector set：
Step 3a) for degree behavioural characteristic distribution probability vector XrBetween Similarity value, its calculation formula is：
Wherein,And H (Xi) it is behavioural characteristic distribution probability vector Xi={ pi1,pi2,...,piNComentropy
Function, its formula is：
Further, I=2 is worked as, andThere was only two behavioural characteristic distribution probability vectors, respectively X1=
{p11,p12,...,p1N, X2={ p21,p22,...,p2NWhen, above formula can be converted into：
Wherein, N represents the total 2k, p of interval division1nRepresent behavioural characteristic sequence X1In n-th of interval distribution probability,
p2nRepresent behavioural characteristic sequence X2In n-th of interval distribution probability；
In the present embodiment, that due to calculating is training vector set P={ X5In 5 behavioural characteristic distribution probability vector Xr,
R=1,2 .., 5 Similarity value between any two, therefore use formula DJS(X1||X2) calculate, obtainIndividual similarity
Value；
Step 3b) smart mobile phone is to step 3a) obtainIndividual Similarity value is averaged, that is, obtains user behavior
Characteristic modelThe model is a similarity average value, is the foundation of authenticating user identification；
Step 4, user preset authentication threshold value σ, its big I are set by user oneself；
Authentication phase：
Behavioural characteristic sequence in certification password and input process that step 5, smart mobile phone acquisition user currently input, its
In, form and the form in step 1 of current behavior characteristic sequence are consistent, be will not be repeated here；
Step 6, smart mobile phone judge whether user's current authentication password matches completely with pre- setting authentication password, if so, right
Current behavior characteristic sequence is pre-processed, and obtains current behavior feature distribution probability vector, and performs step 7, is otherwise performed
Step 5, current behavior characteristic sequence is pre-processed, implementation step is：
Step 6a) current behavior characteristic sequence is normalized first intelligent hand mobile phone, be specially：Ask its 4k-2 spy
Maximum and minimum value in value indicative, each characteristic value for normalizing formula pair using most value are normalized, by original order
It is current normalization behavioural characteristic sequence to arrange obtained sequence；
Step 6b) smart mobile phone carries out feature reconstruction to current normalization behavioural characteristic sequence：Due in training stage area
Between division numbers be 2k, and the interval quantity N of same user is identical, so to current normalization behavioural characteristic sequence statistic
When being distributed number, interval [0,1] is also divided into 2k equal interval, point of its 4k-2 characteristic value on each interval is counted
Cloth number, because interval quantity is 2k, 2k distribution number is obtained so can count, to obtained 2k distribution number difference divided by one
The characteristic value total number 4k-2 of individual behavioural characteristic sequence, obtains 2k distribution probability value pi, wherein, i=1,2 .., 2k, by area
Between stripe sequence be expressed as vector, that is, obtain current behavior feature distribution probability vector C=(p1,p2,...,p2k)；
Step 7, smart mobile phone obtain current average similarity：
Step 7a) smart mobile phone calculating current behavior feature distribution probability vector C=(p1,p2,...,p2k) with train to
Duration set P={ X5In each behavioural characteristic distribution probability vector Xr={ pr1,pr2,...,pr2k, r=1,2 ..., it is similar between 5
Degree, it is with calculating the similarity of all behavioural characteristic distribution probability vectors between any two in training vector set in step (3a)
Calculation formula is identical, will not be repeated here, finally, obtains 5 current Similarity values；
Step 7b) smart mobile phone averages to 5 current Similarity values, obtains current average similarity
Step 8, smart mobile phone obtain authentication result：
Smart mobile phone judges current average similarityWith user behavior characteristic modelDifference percentage it is whether small
In equal to default authentication threshold value, if so, then authentication success, otherwise, authentication fails, specifically：IfThen authentication success；IfThen current user identities authentification failure.
Because the behavioural characteristic of user can be continually changing with the time, to adapt to changing behavioural characteristic, further carry
The high certification degree of accuracy, the present embodiment can be also updated by retraining user behavior characteristic model to it, and implementation method is：Delete
Except training vector set P={ X5In the behavioural characteristic distribution probability vector X that obtains earliest1, by current behavior feature distribution probability
Vectorial C=(p1,p2,...,p2k), to ensure the maximum quantity of behavioural characteristic distribution probability vector in training vector set as 5；
User behavior characteristic model is trained again, it is consistent with the training user behavior characteristic model method of step (3), no longer go to live in the household of one's in-laws on getting married herein
State；
Particular embodiments described above, the present invention is further described, and should be understood that above institute
Specific embodiment only of the invention is stated, the protection domain being not intended to limit the present invention comes for those skilled in the art
Say that various corresponding changes and deformation are made in technical scheme that can be more than and design, and all these change and become
Shape should be construed as being included within the protection domain of the claims in the present invention.
Claims (5)
1. a kind of intelligent terminal identity identifying method based on similarity, it is characterised in that including training stage and authentication phase,
Realize that step is：
Training stage：
(1) intelligent terminal obtains the training dataset of user：
(1a) user preset certification password, inputs intelligent terminal；
Behavioural characteristic sequence during (1b) intelligent terminal record multiple input authentication password of user, forms the training number of user
According to collection；
(2) intelligent terminal is pre-processed to training dataset, obtains training vector set：
(2a) intelligent terminal concentrates each behavioural characteristic sequence to be normalized respectively training data：Intelligent terminal is to training data
Concentrate each characteristic value of each behavioural characteristic sequence to be normalized, obtain multiple normalization behavioural characteristic sequences；
(2b) intelligent terminal carries out feature reconstruction respectively to multiple normalization behavioural characteristic sequences：
Interval [0,1] is divided into N number of equal interval, wherein N >=2 by (2b1) intelligent terminal；
(2b2) intelligent terminal is calculated respectively respectively normalizes behavioural characteristic sequence in each area in multiple normalization behavioural characteristic sequences
Between upper characteristic value distribution probability value, each normalization behavioural characteristic sequence obtains N number of distribution probability value, each normalization gone
The N number of distribution probability value for being characterized sequence is expressed as vector by interval division order, so that it is general to obtain multiple behavioural characteristic distributions
Rate vector, constitutes training vector set；
(3) intelligent terminal trains user behavior characteristic model using training vector set：
(3a) intelligent terminal calculates the similarity of all behavioural characteristic distribution probability vectors between any two in training vector set, obtains
ArriveIndividual Similarity value, wherein, R represents the quantity of behavioural characteristic distribution probability vector in training vector set,
Wherein R >=2；
(3b) intelligent terminal pairIndividual Similarity value is averaged, and obtains user behavior characteristic model；
(4) user preset authentication threshold value；
Authentication phase：
(5) the behavioural characteristic sequence in intelligent terminal acquisition user currently inputs certification password and input process；
(6) intelligent terminal judges whether user's current authentication password matches completely with pre- setting authentication password, if so, to current behavior
Characteristic sequence is pre-processed, and obtains current behavior feature distribution probability vector, and performs step (7), otherwise performs step
(5)；
(7) intelligent terminal obtains current average similarity：
(7a) intelligent terminal calculates current behavior feature distribution probability vector and each behavioural characteristic distribution in training vector set is general
Similarity between rate vector, obtains R current Similarity values；
(7b) intelligent terminal is averaged to R current Similarity values, obtains current average similarity；
(8) intelligent terminal obtains authentication result：
Intelligent terminal judges whether the difference percentage of current average similarity and user behavior characteristic model is less than or equal to preset
Authentication threshold value, if so, then authentication success, otherwise, authentication failure.
2. the intelligent terminal identity identifying method according to claim 1 based on similarity, it is characterised in that step (1b)
Described in behavioural characteristic sequence, refer to one group of sequence of the touch screen behavioural characteristic composition during user's input authentication password,
Including the interval time between adjacent password character, each password character duration, the incoming frequency of each password character,
Any combination of these numeric type behavioural characteristics of each touch screen pressure of password character.
3. the intelligent terminal identity identifying method according to claim 1 based on similarity, it is characterised in that step (3a)
Described in calculate the similarity of all behavioural characteristic distribution probabilities vector between any two, its calculation formula in training vector set
For：
<mrow>
<msub>
<mi>D</mi>
<mrow>
<mi>J</mi>
<mi>S</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>X</mi>
<mn>1</mn>
</msub>
<mo>,</mo>
<msub>
<mi>X</mi>
<mn>2</mn>
</msub>
<mo>,</mo>
<mo>...</mo>
<mo>,</mo>
<msub>
<mi>X</mi>
<mi>I</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>H</mi>
<mrow>
<mo>(</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>I</mi>
</munderover>
<msub>
<mi>&pi;</mi>
<mi>i</mi>
</msub>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>I</mi>
</munderover>
<msub>
<mi>&pi;</mi>
<mi>i</mi>
</msub>
<mi>H</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
Wherein,And H (Xi) it is behavioural characteristic distribution probability vector Xi={ pi1,pi2,...,piNInformation entropy function,
Its formula is：
<mrow>
<mi>H</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>-</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
<mi>log</mi>
<mi> </mi>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
</mrow>
Wherein, N represents interval division sum, pinRepresent behavioural characteristic distribution probability vector XiIn n-th of interval distribution probability.
4. the intelligent terminal identity identifying method according to claim 1 based on similarity, it is characterised in that step (6)
Described in current behavior characteristic sequence is pre-processed, realize that step is：
Current behavior characteristic sequence is normalized (6a) intelligent terminal, obtains currently normalizing behavioural characteristic sequence；
(6b) intelligent terminal carries out feature reconstruction to current normalization behavioural characteristic sequence：Intelligent terminal divides interval [0,1]
For N number of equal interval, the distribution probability value of current normalization behavioural characteristic sequence characteristic value on each interval is calculated, will be distributed
Probable value is expressed as vector by interval division order, obtains current behavior feature distribution probability vector.
5. the intelligent terminal identity identifying method according to claim 1 based on similarity, it is characterised in that step (7a)
Described in calculating current behavior feature distribution probability vector and training vector set between each behavioural characteristic distribution probability vector
Similarity, calculate the phase of all behavioural characteristic distribution probabilities vectors between any two in training vector set in itself and step (3a)
Calculation formula like degree is identical.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710459425.2A CN107194219A (en) | 2017-06-16 | 2017-06-16 | Intelligent terminal identity identifying method based on similarity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710459425.2A CN107194219A (en) | 2017-06-16 | 2017-06-16 | Intelligent terminal identity identifying method based on similarity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107194219A true CN107194219A (en) | 2017-09-22 |
Family
ID=59879153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710459425.2A Pending CN107194219A (en) | 2017-06-16 | 2017-06-16 | Intelligent terminal identity identifying method based on similarity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194219A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110245477A (en) * | 2019-05-31 | 2019-09-17 | 联想(北京)有限公司 | A kind of Information Authentication method, apparatus, equipment and storage medium |
| CN112784224A (en) * | 2019-11-08 | 2021-05-11 | 中国电信股份有限公司 | Terminal safety protection method, device and system |
| CN113032769A (en) * | 2021-04-02 | 2021-06-25 | 西安电子科技大学 | Self-adaptive continuous authentication method based on context |
| CN113726736A (en) * | 2021-07-23 | 2021-11-30 | 浙江嘉兴数字城市实验室有限公司 | Identity authentication method and system based on individual behavior cohesion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101478401A (en) * | 2009-01-21 | 2009-07-08 | 东北大学 | Authentication method and system based on key stroke characteristic recognition |
| US20100036783A1 (en) * | 2008-08-06 | 2010-02-11 | Rodriguez Ralph A | Method of and apparatus for combining artificial intelligence (AI) concepts with event-driven security architectures and ideas |
| CN101751424A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for detecting and tracking topics of online forum |
| CN103530546A (en) * | 2013-10-25 | 2014-01-22 | 东北大学 | Identity authentication method based on mouse behaviors of user |
| CN104408341A (en) * | 2014-11-13 | 2015-03-11 | 西安交通大学 | Smart phone user identity authentication method based on gyroscope behavior characteristics |
-
2017
- 2017-06-16 CN CN201710459425.2A patent/CN107194219A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100036783A1 (en) * | 2008-08-06 | 2010-02-11 | Rodriguez Ralph A | Method of and apparatus for combining artificial intelligence (AI) concepts with event-driven security architectures and ideas |
| CN101751424A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for detecting and tracking topics of online forum |
| CN101478401A (en) * | 2009-01-21 | 2009-07-08 | 东北大学 | Authentication method and system based on key stroke characteristic recognition |
| CN103530546A (en) * | 2013-10-25 | 2014-01-22 | 东北大学 | Identity authentication method based on mouse behaviors of user |
| CN104408341A (en) * | 2014-11-13 | 2015-03-11 | 西安交通大学 | Smart phone user identity authentication method based on gyroscope behavior characteristics |
Non-Patent Citations (2)
| Title |
|---|
| JIANHUA LIN: ""Divergence measures based on the Shannon entropy"", 《IEEE TRANSACTIONS ON INFORMATION THEORY》 * |
| SHUANGYUAN QIAO 等: ""A Secure Authentication Method of Intelligent Terminals Based on Jensen-Shannon Divergence"", 《2017 INTERNATIONAL CONFERENCE ON NETWORKING AND NETWORK APPLICATIONS (NANA)》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110245477A (en) * | 2019-05-31 | 2019-09-17 | 联想(北京)有限公司 | A kind of Information Authentication method, apparatus, equipment and storage medium |
| CN110245477B (en) * | 2019-05-31 | 2021-06-15 | 联想(北京)有限公司 | Information verification method, device, equipment and storage medium |
| CN112784224A (en) * | 2019-11-08 | 2021-05-11 | 中国电信股份有限公司 | Terminal safety protection method, device and system |
| CN112784224B (en) * | 2019-11-08 | 2024-01-30 | 中国电信股份有限公司 | Terminal safety protection method, device and system |
| CN113032769A (en) * | 2021-04-02 | 2021-06-25 | 西安电子科技大学 | Self-adaptive continuous authentication method based on context |
| CN113032769B (en) * | 2021-04-02 | 2022-10-04 | 西安电子科技大学 | Self-adaptive continuous authentication method based on context |
| CN113726736A (en) * | 2021-07-23 | 2021-11-30 | 浙江嘉兴数字城市实验室有限公司 | Identity authentication method and system based on individual behavior cohesion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10558792B2 (en) | Touch-screen user key-press behavior pattern construction and analysis system and identity recognition method thereof | |
| Yang et al. | BehaveSense: Continuous authentication for security-sensitive mobile apps using behavioral biometrics | |
| US11133929B1 (en) | System and method of biobehavioral derived credentials identification | |
| Shahzad et al. | Secure unlocking of mobile touch screen devices by simple gestures: You can see it but you can not do it | |
| Shahzad et al. | Behavior based human authentication on touch screen devices using gestures and signatures | |
| US9563926B2 (en) | System and method of encoding content and an image | |
| Buriro et al. | Please hold on: Unobtrusive user authentication using smartphone's built-in sensors | |
| Stanciu et al. | On the effectiveness of sensor-enhanced keystroke dynamics against statistical attacks | |
| Lu et al. | Safeguard: User reauthentication on smartphones via behavioral biometrics | |
| CN107194219A (en) | Intelligent terminal identity identifying method based on similarity | |
| Buriro et al. | Dialerauth: A motion-assisted touch-based smartphone user authentication scheme | |
| Ren et al. | Signature verification using critical segments for securing mobile transactions | |
| US9202035B1 (en) | User authentication based on biometric handwriting aspects of a handwritten code | |
| Wang et al. | Improving reliability: User authentication on smartphones using keystroke biometrics | |
| Gu et al. | Wione: one-shot learning for environment-robust device-free user authentication via commodity wi-fi in man–machine system | |
| CN106572097A (en) | Mobile device-based mixed identity authentication method | |
| Kobayashi et al. | A behavior authentication method using wi-fi bssids around smartphone carried by a user | |
| Van Nguyen et al. | Finger-drawn pin authentication on touch devices | |
| CN110990811A (en) | Identity authentication method and device | |
| Temper et al. | Touch to authenticate—Continuous biometric authentication on mobile devices | |
| Pryor et al. | Machine learning algorithms in user authentication schemes | |
| CN103297237B (en) | Identity registration and authentication method, system, personal authentication apparatus and certificate server | |
| Bhardwaj et al. | A novel behavioural biometric technique for robust user authentication | |
| Yang et al. | Bubblemap: Privilege mapping for behavior-based implicit authentication systems | |
| Alpar | Biometric touchstroke authentication by fuzzy proximity of touch locations |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170922 |