CN106874852A - A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods - Google Patents
A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods Download PDFInfo
- Publication number
- CN106874852A CN106874852A CN201710025371.9A CN201710025371A CN106874852A CN 106874852 A CN106874852 A CN 106874852A CN 201710025371 A CN201710025371 A CN 201710025371A CN 106874852 A CN106874852 A CN 106874852A
- Authority
- CN
- China
- Prior art keywords
- acceleration transducer
- equipment
- kurtosis
- spec
- extracted
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/13—Sensors therefor
- G06V40/1306—Sensors therefor non-optical, e.g. ultrasonic or capacitive sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
Abstract
Extracted and recognition methods the invention discloses a kind of device-fingerprint based on acceleration transducer, employ the mode of collecting device acceleration transducer data, realize the identification to current device.Compared with prior art, the method is fixed against hardware and unconventional software, improves the reliability and stability of the method；The acceleration transducer of equipment is in the fabrication process because the limitation of processing technology, there is small error between different acceleration transducers, and these errors are almost constant in the life cycle of equipment and cannot change, the mode degree of accuracy in this way relative to hardware informations such as traditional foundation mac addresses it is higher.The inventive method need not install any control and realize the identification of equipment in the case of not needing any operation bidirectional of user simultaneously, improve Consumer's Experience.By after large number quipments inspection, calculating identification accuracy 99% or so.
Description
Technical field
The invention belongs to internet arena, it is related to a kind of equipment based on acceleration transducer to refer to extraction and recognition methods.
Background technology
In the last few years, user's identification turned into application developers and operator's general need.Identifying user and they disappear
Take custom, advertiser, e-commerce platform etc. can be helped more targetedly to deliver advertisement and product, improve conversion ratio.
And for other classes need ensure account safety application, such as electronic wallet application, online shopping application, identifying user is more
It is to meet demand for security.Lawless person steals the feelings that account number cipher is logged in remaining arbitrary equipment and carries out illegal activities
Condition happens occasionally, and associated account number carrys out identifying user with intelligent terminal, can effectively lift the security of account, ensures personal
The safety of property and privacy.Normally, browser can recognize account and equipment by Cookies, and on smart mobile phone
App then by asking the ID of equipment, such as IMEI, carrys out identifying user.But these general methods caused people to
The concern of family personal secrets problem, corresponding measure has been suggested specification these behaviors.Therefore, we have to look for one kind
New mode identifying user, discriminating user account is used on which terminal device.
The content of the invention
The present invention provides a kind of device-fingerprint based on acceleration transducer and extracts and recognition methods, is participated in without user
In the case of, equipment is identified by obtaining equipment Acceleration sensor intrinsic difference, without user installation control, improve
Consumer's Experience.
Device-fingerprint based on acceleration transducer of the invention is extracted and recognition methods, is comprised the following steps
1) calling device interior micromachine shakes equipment, collecting device acceleration transducer data.
2) acceleration transducer data are pre-processed, extracts the quadratic sum root S of sampling interval I (k) and three axle components
(k), then being processed using cubic spline interpolation makes it be uniformly distributed in time domain.
3) average (Mean), standard deviation (Std.Dev), mean difference are extracted in time domain to I (k) and S (k) respectively
(Average Deviation), the degree of bias (Skewness), kurtosis (Kurtosis), RMS amplitude (RMS Amplitude),
Maximum (Highest Value), 8 features of minimum value (Lowest Value), extraction standard is poor on frequency domain
(Spec.Std.Dev), geometric center (Spec.Centroid), the degree of bias (Spec.Skewness), kurtosis
(Spec.Kurtosis), frequency spectrum wave crest (Spectral Crest), K- scramblings (Irregularity-K), J- are irregular
Property (Irregularity-J), smoothness (Smoothness), tonality coefficient (Flatness), roll-offing property (Roll Off) 10
Feature, and this is amounted to 36 features as device-fingerprint.
4) device-fingerprint is trained and is recognized using the method for machine learning.
Beneficial effect：Method used herein, employs the mode of collecting device acceleration transducer data, and it is right to realize
The identification of current device.Compared with prior art, the method is fixed against hardware and unconventional software, and improve the method can
By property and stability；In the fabrication process because the limitation of processing technology, different acceleration are passed the acceleration transducer of equipment
There is small error between sensor, and what these errors were almost constant in the life cycle of equipment and cannot change, institute
The mode degree of accuracy in this way relative to hardware informations such as traditional foundation mac addresses is higher.The inventive method is not simultaneously
The identification installed any control and equipment is realized in the case of not needing any operation bidirectional of user is needed, user's body is improve
Test.By after large number quipments inspection, calculating identification accuracy 99% or so.
Brief description of the drawings
Fig. 1 is the method flow diagram of the embodiment of the present invention.
Fig. 2 is 8 computational methods of feature in time domain.
Fig. 3 is 10 computational methods of feature on frequency domain.
Specific embodiment
With reference to embodiment and Figure of description, the present invention will be further described.
The method flow of the embodiment of the present invention, refers to shown in Fig. 1.
1), it is necessary to call micromachine and acceleration in the insertion of APP login pages before operational outfit fingerprint extraction method
Spend the script of sensor.
2) when opening login page, automatic running script makes smart machine keep the vibrating state of certain hour.In vibrations
During state, the data of acceleration transducer are collected, and upload to high in the clouds.
3) acceleration transducer data are made up of four components, respectively time stamp T (k), X-axis component of acceleration Sx(k),
Y-axis component of acceleration Sy(k) and Z axis component of acceleration Sz(k).Acceleration transducer data are pre-processed beyond the clouds, is extracted
Sampling interval I (k) and quadratic sum root I (k) of three axle components, then being processed using cubic spline interpolation makes it uniformly divide in time domain
Cloth.The computational methods of wherein I (k) and S (k) are as follows：
I (k)=T (k+1)-T (k)
4) average (Mean), standard deviation (Std.Dev), mean difference are extracted in time domain to I (k) and S (k) respectively
(Average Deviation), the degree of bias (Skewness), kurtosis (Kurtosis), RMS amplitude (RMS Amplitude),
Maximum (Highest Value), 8 features of minimum value (Lowest Value), feature calculation method is as shown in Fig. 2 wherein x
It is the time domain expression-form of the initial data of feature to be extracted, N is the number of data in x；Extraction standard is poor on frequency domain
(Spec.Std.Dev), geometric center (Spec.Centroid), the degree of bias (Spec.Skewness), kurtosis
(Spec.Kurtosis), frequency spectrum wave crest (Spectral Crest), K- scramblings (Irregularity-K), J- are irregular
Property (Irregularity-J), smoothness (Smoothness), tonality coefficient (Flatness), roll-offing property (Roll Off) 10
Feature, feature calculation method are as shown in figure 3, wherein y is the frequency domain presentation form of the initial data of feature to be extracted, ymAnd yfPoint
It is amplification coefficient and frequency window, N is ymAnd yfThe number of middle data.And this is amounted to 36 features as device-fingerprint.
5) device-fingerprint is trained and is recognized using the method for machine learning.Each known device is a class sample
This, using polytypic machine learning algorithm, judges the similarity degree of equipment to be identified and each class sample, and to similarity degree
Threshold application method.Rule of thumb given threshold, when equipment to be identified is respectively less than threshold value with the similarity degree of each class, then it is assumed that
The equipment is a new equipment, is added to database, and ask more samples for training.When equipment to be identified and certain
When the similarity degree of one class or multiclass sample is more than threshold value, then it is assumed that the equipment belongs to that maximum class sample of similarity degree
Corresponding equipment.
Claims (1)
1. a kind of device-fingerprint based on acceleration transducer is extracted and recognition methods, it is characterised in that the method includes following
Step：
1) calling device interior micromachine shakes equipment, collecting device acceleration transducer data；
2) acceleration transducer data are pre-processed, extract quadratic sum root S (k) of sampling interval I (k) and three axle components,
It is set to be uniformly distributed in time domain using cubic spline interpolation again；
3) average (Mean), standard deviation (Std.Dev), mean difference (Average are extracted in time domain to I (k) and S (k) respectively
Deviation), the degree of bias (Skewness), kurtosis (Kurtosis), RMS amplitude (RMS Amplitude), maximum
(Highest Value), 8 features of minimum value (Lowest Value), on frequency domain extraction standard poor (Spec.Std.Dev),
Geometric center (Spec.Centroid), the degree of bias (Spec.Skewness), kurtosis (Spec.Kurtosis), frequency spectrum wave crest
(Spectral Crest), K- scramblings (Irregularity-K), J- scramblings (Irregularity-J), smoothness
(Smoothness), tonality coefficient (Flatness), 10 features of roll-offing property (Roll Off), and this is amounted into 36 features works
It is device-fingerprint；
4) device-fingerprint is trained and is recognized using the method for machine learning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710025371.9A CN106874852A (en) | 2017-01-13 | 2017-01-13 | A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710025371.9A CN106874852A (en) | 2017-01-13 | 2017-01-13 | A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106874852A true CN106874852A (en) | 2017-06-20 |
Family
ID=59157661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710025371.9A Pending CN106874852A (en) | 2017-01-13 | 2017-01-13 | A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106874852A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108664785A (en) * | 2018-04-04 | 2018-10-16 | 浙江大学 | A kind of device-fingerprint extraction and authentication method based on CPU module electromagnetic radiation |
| CN111141284A (en) * | 2019-12-28 | 2020-05-12 | 西安交通大学 | Intelligent building personnel thermal comfort level and thermal environment management system and method |
| CN113111726A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction and identification method based on homologous signals |
| CN113111725A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction identification system based on homologous signal |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105241445A (en) * | 2015-10-20 | 2016-01-13 | 深圳大学 | Method and system for acquiring indoor navigation data based on intelligent mobile terminal |
| CN105962559A (en) * | 2016-07-06 | 2016-09-28 | 张远海 | Smartband with fingerprint recognition function |
-
2017
- 2017-01-13 CN CN201710025371.9A patent/CN106874852A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105241445A (en) * | 2015-10-20 | 2016-01-13 | 深圳大学 | Method and system for acquiring indoor navigation data based on intelligent mobile terminal |
| CN105962559A (en) * | 2016-07-06 | 2016-09-28 | 张远海 | Smartband with fingerprint recognition function |
Non-Patent Citations (2)
| Title |
|---|
| SANORITA DEY: "AccelPrint:Accelerometers are Different by Birth", 《HTTP://SCHOLARCOMMONS.SC.EDU/ETD/》 * |
| SANORITA DEY等: "AccelPrint: Imperfections of Accelerometers Make Smartphones Trackable", 《IN PROCEEDINGS OF THE NETWORK AND DISTRIBUTED SYSTEM SECURITY SYMPOSIUM (NDSS)》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108664785A (en) * | 2018-04-04 | 2018-10-16 | 浙江大学 | A kind of device-fingerprint extraction and authentication method based on CPU module electromagnetic radiation |
| CN108664785B (en) * | 2018-04-04 | 2020-12-08 | 浙江大学 | Equipment fingerprint extraction and authentication method based on CPU module electromagnetic radiation |
| CN111141284A (en) * | 2019-12-28 | 2020-05-12 | 西安交通大学 | Intelligent building personnel thermal comfort level and thermal environment management system and method |
| CN113111726A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction and identification method based on homologous signals |
| CN113111725A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction identification system based on homologous signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106874852A (en) | A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods | |
| CN107831907A (en) | Identity identifying method and device based on Gait Recognition | |
| CN105279405A (en) | Keypress behavior pattern construction and analysis system of touch screen user and identity recognition method thereof | |
| WO2017107422A1 (en) | Method and device for user gender identification | |
| CN107437074B (en) | Identity authentication method and device | |
| CN106295675B (en) | A kind of Activity recognition method based on smart phone of high accuracy | |
| US20180038909A1 (en) | Analog circuit fault diagnosis method using single testable node | |
| CN110058699B (en) | User behavior identification method based on intelligent mobile device sensor | |
| WO2019192235A1 (en) | User identity authentication method and system based on mobile device | |
| Qin et al. | A fuzzy authentication system based on neural network learning and extreme value statistics | |
| KR20170091318A (en) | Authentication apparatus and method based on electrocardiographic signals | |
| CN106875203A (en) | A kind of method and device of the style information for determining commodity picture | |
| Al-Naffakh et al. | Unobtrusive gait recognition using smartwatches | |
| CN105046882B (en) | Fall down detection method and device | |
| CN106529377A (en) | Age estimating method, age estimating device and age estimating system based on image | |
| CN109766683B (en) | Protection method for sensor fingerprint of mobile intelligent device | |
| CN113946590A (en) | Method, device and equipment for updating integral data and storage medium | |
| CN106228133A (en) | User authentication method and device | |
| US20200242481A1 (en) | Method for providing data associated with original data and electronic device and storage medium for the same | |
| CN108664785B (en) | Equipment fingerprint extraction and authentication method based on CPU module electromagnetic radiation | |
| CN108965585B (en) | User identity recognition method based on smart phone sensor | |
| Jakkala et al. | Deep CSI learning for gait biometric sensing and recognition | |
| CN110765939A (en) | Identity recognition method and device, mobile terminal and storage medium | |
| CN107077617A (en) | fingerprint extraction method and device | |
| CN104657349B (en) | A kind of forum postings characteristic recognition method and device |
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: 20170620 |