CN107273795B - Cross-device electromagnetic fingerprint database construction method and device based on machine learning - Google Patents
Cross-device electromagnetic fingerprint database construction method and device based on machine learning Download PDFInfo
- Publication number
- CN107273795B CN107273795B CN201710309422.0A CN201710309422A CN107273795B CN 107273795 B CN107273795 B CN 107273795B CN 201710309422 A CN201710309422 A CN 201710309422A CN 107273795 B CN107273795 B CN 107273795B
- Authority
- CN
- China
- Prior art keywords
- electromagnetic
- signal
- electromagnetic signal
- sample data
- equipment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 26
- 238000010276 construction Methods 0.000 title description 3
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 30
- 230000009467 reduction Effects 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000003321 amplification Effects 0.000 claims description 3
- 238000000556 factor analysis Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 230000010355 oscillation Effects 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 230000002452 interceptive effect Effects 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G06V40/1365—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Geophysics And Detection Of Objects (AREA)
- Measurement Of Length, Angles, Or The Like Using Electric Or Magnetic Means (AREA)
Abstract
The invention discloses a method for constructing a cross-device electromagnetic fingerprint database based on machine learning, which comprises the following steps: firstly, acquiring an electromagnetic signal of equipment, and performing noise reduction, specific window bandwidth interception and digital processing on the electromagnetic signal; then, machine learning is carried out on the processed electromagnetic signals, classification and marking of the electromagnetic signals are obtained through judgment, and information of the classification and marking is transmitted to a display; and finally, judging the correctness of the displayed equipment identification result by the user, if so, storing the classification and marking information into an electromagnetic fingerprint database, otherwise, learning from a new machine from the electromagnetic signal of the newly acquired equipment until the accuracy is reached. The method can accurately obtain the electromagnetic signals of the equipment, further establish the electromagnetic fingerprint database, and update and expand the electromagnetic fingerprint database while using the electromagnetic fingerprint database. The invention also discloses a device for realizing the method, which does not need the support of external marking equipment, saves the cost and enhances the interactive naturalness with the articles.
Description
Technical Field
the invention belongs to the field of electronic equipment marking, and particularly relates to a method and a device for constructing a cross-equipment electromagnetic fingerprint database based on machine learning.
Background
the rapid development of electronic technology and internet of things technology enables more and more devices to use microelectronic circuits, embedded chips and sensors, and forms unique electromagnetic signals in combination with the conductivity, electromagnetic signal patterns and the like of the devices themselves. This unique electromagnetic signal can be used to mark a unique device as an electromagnetic fingerprint.
The current cross-device information marking technology and method still have the following defects and shortcomings:
1) Additional device support is required, for example, a radio frequency tag needs to be attached to a target device;
2) the cost is high, high-standard electronic signal equipment is needed for equipment identification and marking in a severe environment, and the cost is high under the condition of large-scale deployment;
3) The environment is not friendly, radio frequency tags and the like can be subjected to the problems that the radio frequency tags and the like cannot be removed randomly, are not removed completely, are easy to damage original tags, cannot reuse the tags and the like when the radio frequency tags and the like need to be removed, and the treatment of a large amount of metal components and other compounds of waste tags is a severe test for the environment;
4) the expansibility and compatibility are not strong, and the current equipment marking technology and method often need to print a special pattern on the surface of equipment, such as a two-dimensional code, or need to stick an additional electromagnetic mark on the surface of an article, such as a radio frequency tag, or need to embed a unique electromagnetic signal identification device inside the article to distinguish other articles. Such unique tag systems or built-in identification devices often require targeted detection and interaction systems, which bring great inconvenience to cross-category device labeling and modification and expansion of the labeling;
5) The prior equipment marking technology and method cannot automatically classify and mark the obtained information and provide a user interaction interface for information confirmation after the detection of additional equipment.
Therefore, to solve the disadvantages and shortcomings of the above-mentioned device information marking system, those skilled in the art are faced with a need for improvement and solution.
Disclosure of Invention
In view of the above, the invention provides a method and an apparatus for constructing a cross-device electromagnetic fingerprint database based on machine learning, the method can accurately obtain electromagnetic signals of devices, further establish an electromagnetic fingerprint database, and update and expand the electromagnetic fingerprint database while using the electromagnetic fingerprint database.
A method for constructing a cross-device electromagnetic fingerprint database based on machine learning comprises the following steps:
(1) acquiring an electromagnetic signal of the device by using a contact antenna;
(2) the electromagnetic sensor collects an electromagnetic signal of the equipment and transmits the electromagnetic signal to the electromagnetic processing unit;
(3) The electromagnetic processing unit carries out background noise reduction, specific window bandwidth interception and digital processing on the received electromagnetic signals to obtain digital signals serving as sample data of machine learning, and the sample data is transmitted to the signal learning unit;
(4) The signal learning unit performs machine learning on the received sample data, judges the classification and the marking of the sample data and transmits the classification and the marking information to the display;
(5) The display displays the identification result of the equipment by combining the classification of the sample data and the marking information in an icon mode;
(6) And (3) judging whether the displayed equipment identification result is accurate information of the electromagnetic signal of the equipment by the user, if so, storing the classification and marking information into an electromagnetic fingerprint database to finish the acquisition of the electromagnetic fingerprint of the equipment, and if not, skipping to execute the step (1).
in the step (2), the electromagnetic sensor continuously samples the electromagnetic signal of the device for 2 seconds at a frequency of 21.8MHz and a window size of 0.1 second to obtain electromagnetic signal data, wherein the electromagnetic signal data comprises the frequency, amplitude and waveform of the signal.
In the step (3), the specific process of background noise reduction on the received electromagnetic signal is as follows:
Firstly, before an antenna is contacted with or not contacted with equipment, obtaining a background electromagnetic signal of a current environment, performing statistical classification on the background electromagnetic signal to obtain a background electromagnetic signal sample, and analyzing the background electromagnetic signal sample to obtain a background electromagnetic signal reference value z;
then, whether the electromagnetic signal is larger than a background electromagnetic signal reference value z or not is judged, if so, the original electromagnetic signal is amplified, if not, the electromagnetic signal is set to be 0, the amplified electromagnetic signal and the electromagnetic signal set to be 0 form an electromagnetic signal after background noise reduction, and the electromagnetic signal after background noise reduction is specifically formed through a formula
Sn＝max(0,Fn-(Gn+zσ))×A
Is realized in that SnElectromagnetic signals after noise reduction for the background, FnFor the nth original electromagnetic signal in the continuous signal stream, GnThe signal is the nth signal in the background electromagnetic signal sample, z is the background electromagnetic signal reference value, sigma is the amplitude standard deviation of the nth original electromagnetic signal, and A is the electromagnetic signal amplification factor.
the background electromagnetic signal comprises the wave band, frequency and amplitude of the signal, and when the contact antenna just contacts the equipment, the environmental background signal is subtracted from the current electromagnetic signal to obtain the electromagnetic signal belonging to the equipment.
In the step (3), the electromagnetic signal subjected to background noise reduction is intercepted with a specific window bandwidth, so that the frequency range of the electromagnetic signal is selected, and the electromagnetic signal in a certain frequency range is obtained.
the specific process of the step (4) is as follows:
(4-1) carrying out sparsification on the sample data, and removing a part with higher repetition frequency in the sample data;
(4-2) normalizing the sample data after the thinning processing, and according to the amplitude of the electromagnetic signals, enabling the electromagnetic signals to be normalized to be 0-1 digits, enabling the electromagnetic signals with the maximum amplitude to be normalized to be 1 and enabling the electromagnetic signals with the minimum amplitude to be normalized to be 0 to obtain normalized sample data, wherein the digits are accurate to three digits after a decimal point;
(4-3) clustering the sample data after the normalization processing by adopting a K-means algorithm to obtain the data with the largest data size as the optimal electromagnetic signal of the equipment;
and (4-4) extracting the characteristics and the modes of the optimal electromagnetic signals to obtain the classification and the mark of the optimal electromagnetic signals.
the specific process of the step (4-4) is as follows:
(4-4-1) describing the obtained electromagnetic signals with respect to amplitude, wavelength, waveform, and distribution of frequencies of the high-frequency and low-frequency electromagnetic signals, respectively, and extracting a feature set capable of completely describing a specific electromagnetic signal sample;
(4-4-2) taking each feature in the feature set as a factor, obtaining factor weight (factor score) of each feature by utilizing factor analysis, and using the feature corresponding to each factor and the correlation among the features as the optimal classification and marking result of the electromagnetic signal sample according to the weight; in the step (5), the equipment identification result comprises the frequency, amplitude and waveform of the electromagnetic signal.
after the electromagnetic fingerprint database is established, the electromagnetic signal of the target object is compared with the characteristics in the electromagnetic fingerprint database, so that the target object is identified, and meanwhile, the electromagnetic fingerprint database is updated by utilizing the electromagnetic signal, so that a foundation is laid for the accuracy of other target objects in the future.
According to the method, the equipment is automatically classified and marked according to the unique electromagnetic fingerprint of the equipment through machine learning, so that the use of external marking equipment (radio frequency tags) is reduced, the cost is saved, and meanwhile, the interactive naturalness with the articles is enhanced.
in another aspect, the present invention provides an apparatus for implementing the method, including:
A contact antenna for acquiring an electromagnetic signal of the device;
The electromagnetic sensor is used for acquiring an electromagnetic signal of the equipment and transmitting the electromagnetic signal to the electromagnetic processing unit;
The electromagnetic processing unit is used for carrying out background noise reduction, specific window bandwidth interception and digital processing on the received electromagnetic signals to obtain digital signals serving as sample data of machine learning and transmitting the sample data to the signal learning unit;
The signal learning unit is used for performing machine learning on the received sample data, judging to obtain the classification and the marking of the sample data and transmitting the classification and the marking information to the display;
and the display displays the identification result of the equipment in an icon combination mode according to the classification of the sample data and the marking information.
preferably, the oscillation frequency of the electromagnetic sensor is 28.8MHz, and the bandwidth sampling window range is 0-1 MHz.
The electromagnetic processing unit consists of an electromagnetic filter and an AD conversion circuit, wherein the electromagnetic filter is used for performing background noise reduction and specific window bandwidth interception processing on the received electromagnetic signals; the AD conversion circuit is used for converting the electromagnetic signals processed by the electromagnetic filter into digital signals.
preferably, the electromagnetic sensor, the electromagnetic processing unit and the signal learning unit are integrated on an embedded system circuit board, so that the distance of signal transmission is shortened, namely the transmission time is shortened, the loss of electromagnetic signals in the transmission process is greatly reduced, and the accuracy of the electromagnetic signals is ensured.
the method can be applied to information identification marks of electronic equipment, and the information is classified and marked through machine learning. The method has better expansibility and compatibility, and a targeted detection and interaction system does not need to be added externally. The method of the invention classifies and marks the marked information through machine learning, so that the cross-equipment information marking is more rapid and convenient, and the more the detection quantity is, the finer the classification is. The method has better man-machine interaction, and simply and directly provides an interactive interface for the user for information confirmation. The device of the invention does not need the support of external marking equipment, saves cost and enhances the interactive naturalness with articles. The device of the invention avoids the damage, replacement and elimination of redundant equipment, and is beneficial to environmental protection.
Drawings
FIG. 1 is a schematic structural diagram of a cross-device electromagnetic fingerprint database construction device in an embodiment;
FIG. 2 is a flowchart of a method for constructing a cross-device electromagnetic fingerprint database according to an embodiment;
Fig. 3 is a flowchart illustrating a specific process of performing machine learning on received sample data according to an embodiment.
Detailed Description
in order to more specifically describe the present invention, the following detailed description is provided for the technical solution of the present invention with reference to the accompanying drawings and the specific embodiments.
referring to fig. 1, the device for constructing the cross-device electromagnetic fingerprint database based on machine learning in the present embodiment includes: the mobile phone comprises an electromagnetic sensor, an electromagnetic processing unit, a signal learning unit and a display, wherein the electromagnetic sensor, the electromagnetic processing unit and the signal learning unit are integrated on an embedded system circuit board; the electromagnetic processing unit also comprises an electromagnetic filter and an AD conversion circuit.
fig. 2 shows a method for establishing a cross-device electromagnetic fingerprint database by using the apparatus shown in fig. 1, which specifically includes:
s01, connecting the device with the mobile phone by using the contact antenna.
and S02, the electromagnetic sensor collects the electromagnetic signal of the equipment through the contact antenna and transmits the electromagnetic signal to the electromagnetic processing unit.
in this step, the electromagnetic sensor continuously samples the electromagnetic signal of the device for 2 seconds at a frequency of 21.8MHz and a window size of 0.1 second, and obtains electromagnetic signal data, which includes the frequency, amplitude and waveform of the signal.
and S03, the electromagnetic filter carries out background noise reduction and specific window bandwidth interception processing on the received electromagnetic signal. The specific process of background noise reduction is as follows:
firstly, before an antenna is contacted with or not contacted with equipment, obtaining a background electromagnetic signal of a current environment, performing statistical classification on the background electromagnetic signal to obtain a background electromagnetic signal sample, and analyzing the background electromagnetic signal sample to obtain a background electromagnetic signal reference value z;
Then, whether the electromagnetic signal is larger than a background electromagnetic signal reference value z or not is judged, if so, the original electromagnetic signal is amplified, if not, the electromagnetic signal is set to be 0, the amplified electromagnetic signal and the electromagnetic signal set to be 0 form an electromagnetic signal after background noise reduction, and the electromagnetic signal after background noise reduction is specifically formed through a formula
Sn＝max(0,Fn-(Gn+zσ))×A
Is realized in that SnElectromagnetic signals after noise reduction for the background, FnFor the nth original electromagnetic signal in the continuous signal stream, Gnthe signal is the nth signal in the background electromagnetic signal sample, z is the background electromagnetic signal reference value, sigma is the amplitude standard deviation of the nth original electromagnetic signal, and A is the electromagnetic signal amplification factor.
s04, the AD conversion circuit converts the electromagnetic signal processed in S03 into a digital signal, and takes the digital signal as sample data for machine learning, and sends it to the signal learning unit.
And S05, the signal learning unit performs machine learning on the received sample data, judges the classification and the marking of the sample data, and transmits the classification and the marking information to the display.
Referring to fig. 3, in this step, the specific process of machine learning is:
(a) sparsifying the sample data, and removing a part with higher repetition frequency in the sample data;
(b) Normalizing the sample data after the thinning processing, and according to the amplitude of the electromagnetic signals, normalizing the electromagnetic signals into 0-1 numbers, normalizing the electromagnetic signals with the maximum amplitude into 1, and normalizing the electromagnetic signals with the minimum amplitude into 0 to obtain normalized sample data;
(c) clustering the sample data after normalization processing by adopting a K-means algorithm to obtain the data with the largest data amount as the optimal electromagnetic signal of the equipment;
(d) respectively describing the obtained electromagnetic signals according to the amplitude, wavelength, waveform and frequency distribution of the high-frequency electromagnetic signals and the low-frequency electromagnetic signals, and extracting a characteristic set capable of completely describing a specific electromagnetic signal sample;
(e) Taking each feature in the feature set as a factor, obtaining factor weight (factor score) of each feature by utilizing factor analysis, and using the feature corresponding to each factor and the correlation among the features as the optimal classification and marking result of the electromagnetic signal sample according to the weight; s06, the display displays the identification result of the equipment by the way of combining the classification of the sample data and the mark information in an icon way;
s07, the user judges whether the displayed device identification result is the accurate information of the device electromagnetic signal, if yes, the classification and marking information is stored in the electromagnetic fingerprint database to complete the acquisition of the device electromagnetic fingerprint, and if not, the user jumps to execute S01.
after the electromagnetic fingerprint database is established, the electromagnetic signal of the target object is compared with the characteristics in the electromagnetic fingerprint database, so that the target object is identified, and meanwhile, the electromagnetic fingerprint database is updated by utilizing the electromagnetic signal, so that a foundation is laid for the accuracy of other target objects in the future.
The above-mentioned embodiments are intended to illustrate the technical solutions and advantages of the present invention, and it should be understood that the above-mentioned embodiments are only the most preferred embodiments of the present invention, and are not intended to limit the present invention, and any modifications, additions, equivalents, etc. made within the scope of the principles of the present invention should be included in the scope of the present invention.
Claims (7)
1. a method for constructing a cross-device electromagnetic fingerprint database based on machine learning comprises the following steps:
(1) Acquiring an electromagnetic signal of the device by using a contact antenna;
(2) the electromagnetic sensor collects an electromagnetic signal of the equipment and transmits the electromagnetic signal to the electromagnetic processing unit;
(3) the electromagnetic processing unit carries out background noise reduction, specific window bandwidth interception and digital processing on the received electromagnetic signals to obtain digital signals serving as sample data of machine learning, and the sample data is transmitted to the signal learning unit, wherein the specific process of carrying out background noise reduction on the received electromagnetic signals is as follows:
firstly, before an antenna is contacted with or not contacted with equipment, obtaining a background electromagnetic signal of a current environment, performing statistical classification on the background electromagnetic signal to obtain a background electromagnetic signal sample, and analyzing the background electromagnetic signal sample to obtain a background electromagnetic signal reference value z;
Then, whether the electromagnetic signal is larger than a background electromagnetic signal reference value z or not is judged, if yes, the original electromagnetic signal is amplified, if not, the electromagnetic signal is set to be 0, the amplified electromagnetic signal and the electromagnetic signal set to be 0 form the electromagnetic signal after background noise reduction, and the formula is specifically adopted
Sn＝max(0,Fn-(Gn+zσ))×A
Is realized in that Snelectromagnetic signals after noise reduction for the background, FnFor the nth original electromagnetic signal in the continuous signal stream, Gnthe method comprises the steps of taking an nth signal in a background electromagnetic signal sample, taking z as a background electromagnetic signal reference value, taking sigma as an amplitude standard deviation of an nth original electromagnetic signal, and taking A as an electromagnetic signal amplification factor;
(4) The signal learning unit performs machine learning on the received sample data, judges the classification and the marking of the sample data and transmits the classification and the marking information to the display;
(5) The display displays the identification result of the equipment by combining the classification of the sample data and the marking information in an icon mode;
(6) and (3) judging whether the displayed equipment identification result is accurate information of the electromagnetic signal of the equipment by the user, if so, storing the classification and marking information into an electromagnetic fingerprint database to finish the acquisition of the electromagnetic fingerprint of the equipment, and if not, skipping to execute the step (1).
2. the method for constructing the cross-device electromagnetic fingerprint database based on machine learning according to claim 1, wherein the specific process of the step (4) is as follows:
(4-1) carrying out sparsification on the sample data, and removing a part with high repetition frequency in the sample data;
(4-2) normalizing the sample data after the thinning processing, and according to the amplitude of the electromagnetic signals, normalizing the electromagnetic signals into 0-1 numbers, wherein the electromagnetic signals with the maximum amplitude are normalized into 1, and the electromagnetic signals with the minimum amplitude are normalized into 0 to obtain normalized sample data;
(4-3) clustering the sample data after the normalization processing by adopting a K-means algorithm to obtain the data with the largest data size as the optimal electromagnetic signal of the equipment;
And (4-4) extracting the characteristics and the modes of the optimal electromagnetic signals to obtain the classification and the mark of the optimal electromagnetic signals.
3. The method for constructing the cross-device electromagnetic fingerprint database based on machine learning as claimed in claim 2, wherein the specific process of step (4-4) is as follows:
(4-4-1) describing the obtained electromagnetic signals with respect to amplitude, wavelength, waveform, and distribution of frequencies of the high-frequency and low-frequency electromagnetic signals, respectively, and extracting a feature set capable of completely describing a specific electromagnetic signal sample;
(4-4-2) taking each feature in the feature set as a factor, obtaining the factor weight of each feature by using factor analysis, and using the feature corresponding to each factor and the correlation among the features as the optimal classification and marking result of the electromagnetic signal sample according to the weight.
4. an apparatus for implementing the method of any one of claims 1 to 3, comprising:
a contact antenna for acquiring an electromagnetic signal of the device;
the electromagnetic sensor is used for acquiring an electromagnetic signal of the equipment and transmitting the electromagnetic signal to the electromagnetic processing unit;
the electromagnetic processing unit is used for carrying out background noise reduction, specific window bandwidth interception and digital processing on the received electromagnetic signals to obtain digital signals serving as sample data of machine learning and transmitting the sample data to the signal learning unit;
The signal learning unit is used for performing machine learning on the received sample data, judging to obtain the classification and the marking of the sample data and transmitting the classification and the marking information to the display;
and the display displays the identification result of the equipment in an icon combination mode according to the classification of the sample data and the marking information.
5. the apparatus of claim 4, wherein the oscillation frequency of the electromagnetic sensor is 28.8MHz, and the bandwidth sampling window is in the range of 0-1 MHz.
6. The apparatus of claim 4, wherein the electromagnetic processing unit is composed of an electromagnetic filter and an AD conversion circuit, the electromagnetic filter is used for performing background noise reduction and specific window bandwidth interception processing on the received electromagnetic signal; the AD conversion circuit is used for converting the electromagnetic signals processed by the electromagnetic filter into digital signals.
7. the apparatus of claim 4, wherein the electromagnetic sensor, the electromagnetic processing unit and the signal learning unit are integrated on an embedded system circuit board.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710309422.0A CN107273795B (en) | 2017-05-04 | 2017-05-04 | Cross-device electromagnetic fingerprint database construction method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710309422.0A CN107273795B (en) | 2017-05-04 | 2017-05-04 | Cross-device electromagnetic fingerprint database construction method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107273795A CN107273795A (en) | 2017-10-20 |
| CN107273795B true CN107273795B (en) | 2019-12-17 |
Family
ID=60074361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710309422.0A Active CN107273795B (en) | 2017-05-04 | 2017-05-04 | Cross-device electromagnetic fingerprint database construction method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107273795B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11131787B2 (en) * | 2018-05-28 | 2021-09-28 | Samsung Electronics Co., Ltd. | Electromagnetic sensor and mobile device including the same |
| CN109946538B (en) * | 2019-03-14 | 2020-06-19 | 上海交通大学 | Mobile equipment monitoring method and system based on magnetic induction signals |
| CN110941020B (en) * | 2019-12-16 | 2022-06-10 | 杭州安恒信息技术股份有限公司 | Detection method and device for stolen shooting equipment based on electromagnetic leakage |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110161A (en) * | 2007-08-31 | 2008-01-23 | 北京科技大学 | System for automatic cab model recognition and automatic vehicle flowrate detection and method thereof |
| CN102571230A (en) * | 2011-12-22 | 2012-07-11 | 中国人民解放军总参谋部第六十三研究所 | Distributed collaborative signal identification method based on blind estimation of higher order statistics and signal to noise ratio |
| CN103294199A (en) * | 2013-06-09 | 2013-09-11 | 华东理工大学 | Silent information identifying system based on facial muscle sound signals |
| CN106529393A (en) * | 2016-09-18 | 2017-03-22 | 北京航空航天大学 | Electromagnetic signal spectrum data classification method with ESMD sample entropy combined with FCM |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6963659B2 (en) * | 2000-09-15 | 2005-11-08 | Facekey Corp. | Fingerprint verification system utilizing a facial image-based heuristic search method |
| US7310432B2 (en) * | 2004-06-26 | 2007-12-18 | Artinnet Corp. | Ported system for personal identity verification |
-
2017
- 2017-05-04 CN CN201710309422.0A patent/CN107273795B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110161A (en) * | 2007-08-31 | 2008-01-23 | 北京科技大学 | System for automatic cab model recognition and automatic vehicle flowrate detection and method thereof |
| CN102571230A (en) * | 2011-12-22 | 2012-07-11 | 中国人民解放军总参谋部第六十三研究所 | Distributed collaborative signal identification method based on blind estimation of higher order statistics and signal to noise ratio |
| CN103294199A (en) * | 2013-06-09 | 2013-09-11 | 华东理工大学 | Silent information identifying system based on facial muscle sound signals |
| CN106529393A (en) * | 2016-09-18 | 2017-03-22 | 北京航空航天大学 | Electromagnetic signal spectrum data classification method with ESMD sample entropy combined with FCM |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107273795A (en) | 2017-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107273795B (en) | Cross-device electromagnetic fingerprint database construction method and device based on machine learning | |
| CN111368934B (en) | Image recognition model training method, image recognition method and related device | |
| Wang et al. | WiCAR: WiFi-based in-car activity recognition with multi-adversarial domain adaptation | |
| CN105912916B (en) | A kind of unlocked by fingerprint method and terminal | |
| US20160371520A1 (en) | Wearable radio frequency identification enabled devices | |
| US8131074B2 (en) | Specific emitter identification using histogram of oriented gradient features | |
| CN109726726B (en) | Event detection method and device in video | |
| CN105912920A (en) | Fingerprint unlocking method and terminal | |
| JP2010086201A (en) | Mobile terminal device, program for the mobile terminal device, image recognition system and stolen vehicle retrieval system | |
| CN106452716A (en) | Unmanned aerial vehicle signal recognition and detection algorithm based on Hash fingerprint | |
| CN107292359B (en) | Cross-equipment article marking method and device based on electromagnetic signal feature recognition | |
| Li et al. | Log-likelihood clustering-enabled passive RF sensing for residential activity recognition | |
| CN111830321B (en) | Unmanned aerial vehicle detection and identification method based on radio frequency fingerprint | |
| CN107480608A (en) | Anti-fake processing method and related product | |
| Xue et al. | Radio frequency identification for drones using spectrogram and CNN | |
| CN210664405U (en) | PCB copper thickness measuring system | |
| CN112001198A (en) | RFID card reader, card reading method and method for assisting in positioning RFID tag | |
| CN112840308B (en) | Font optimizing method and related equipment | |
| WO2021240996A1 (en) | Learning device, reader, and learning program | |
| CN106651406A (en) | Payment terminal detecting device, system and method | |
| CN103902946B (en) | Signal analysis system meeting ISO/IEC15693 standard and working method thereof | |
| CN113282925A (en) | Malicious file detection method and device, terminal equipment and storage medium | |
| JP2003270326A (en) | Pulse signal analyzer and pulse signal analyzing method | |
| Liu et al. | RISAR: RIS-assisted Human Activity Recognition with Commercial Wi-Fi Devices | |
| CN101605331B (en) | Consumption method, consumption device and consumption system of mobile terminal |
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 |