CN110163099A - A kind of abnormal behaviour identification device and method based on electromagnetic leakage signal - Google Patents
A kind of abnormal behaviour identification device and method based on electromagnetic leakage signal Download PDFInfo
- Publication number
- CN110163099A CN110163099A CN201910308245.3A CN201910308245A CN110163099A CN 110163099 A CN110163099 A CN 110163099A CN 201910308245 A CN201910308245 A CN 201910308245A CN 110163099 A CN110163099 A CN 110163099A
- Authority
- CN
- China
- Prior art keywords
- signal
- leakage signal
- abnormal
- classification
- electromagnetic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/08—Measuring electromagnetic field characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of abnormal behaviour identification device and method based on electromagnetic leakage signal, identification device includes sequentially connected signal acquisition module, intelligent classification identification module and differentiates that module is presented in result, in which: the signal acquisition module is for acquiring electromagnetic background signal, electromagnetic leakage signal, normal signal, abnormal signal, critical behavior leakage signal and generating electromagnetic leakage signal image and send intelligent classification identification module to；The intelligent classification identification module is responsible for scene with/without leakage signal, to scene with/without abnormal signal and to critical behavior leakage signal progress intelligent recognition, and recognition result is sent to and differentiates that result is presented module and presents.The present invention has effectively achieved the monitoring alarm to illegal access device and unit exception behavior；And existing feature samples library can be carried out timely updating and expanding, solve repetition learning and demand of the depth network training to great amount of samples.
Description
Technical field
The invention belongs to deep learnings and image identification technical field, and in particular to a kind of based on the different of electromagnetic leakage signal
Normal Activity recognition apparatus and method.
Background technique
Artificial intelligence is the forward position direction of computer field development in recent years.Depth learning technology is then to realize artificial intelligence
One of the important means of and machine learning field hot spot, utilize deep neural network, image recognition, voice know
The multiple fields such as not realize important breakthrough.
In field of image recognition, depth learning technology is current most effective means.Deep learning image recognition technology is
Through the identification of Vehicles and Traffic Signs board, human body static behavior Classification and Identification, floristics and color identification, recognition of face and
Medically cancer identification etc. is widely used, but there is no corresponding in terms of electromagnetic information leakage for the technology
Using.
Electromagnetic leakage signal implies the running parameter, behavior state and sensitive letter of target device information processing or transmission
Breath.Electromagnetic leakage signal is acquired by channels such as power lines and forms electromagnetism " image " sample with rich connotation, utilizes depth
Degree study and image recognition technology parse above-mentioned partial parameters, behavior, information, and establish abnormal behaviour sample database and identification hand
Section, to push the electronic device information security evaluation application based on electromagnetic leakage signal.
Summary of the invention
In order to overcome the disadvantages mentioned above of the prior art, the present invention provides a kind of abnormal behaviours based on electromagnetic leakage signal
Identification device and method.
The technical solution adopted by the present invention to solve the technical problems is: a kind of abnormal behaviour based on electromagnetic leakage signal
Module is presented in identification device, including sequentially connected signal acquisition module, intelligent classification identification module and differentiation result, in which:
The signal acquisition module is for acquiring electromagnetic background signal, electromagnetic leakage signal, normal signal, abnormal signal, critical behavior
Leakage signal simultaneously generates electromagnetic leakage signal image and sends intelligent classification identification module to；The intelligent classification identification module is responsible for
To scene with/without leakage signal, to scene with/without abnormal signal and to critical behavior leakage signal progress intelligent recognition, and will
Recognition result, which is sent to, differentiates that module is presented in result；The differentiation result is presented module and is used for the knowledge of intelligent classification identification module
Other result is presented.
The abnormal behaviour recognition methods based on electromagnetic leakage signal that the present invention also provides a kind of, including following content:
One, signal acquisition:
Reception is acquired to the leakage signal of target device or route periphery by coupling probe or antenna, generation includes
The electromagnetic leakage signal picture of the information such as time, signal strength, signal frequency；
Two, intelligent classification identifies:
Firstly, being identified to scene with/without leakage signal, if there is leakage signal, for individual equipment, identification leakage letter
Operation behavior corresponding to number；For multiple and different equipment, identifies with/without abnormal signal, if there is abnormal signal appearance, judge
Whether there is the access of illegality equipment, if the access without illegality equipment, is identified according to individual equipment critical behavior to abnormal signal
Corresponding operation behavior is identified；
Three, recognition result is presented:
After Classification and Identification, by with/without leakage signal, with/without abnormal signal, the specific ratio of critical behavior leakage signal
Similarity result is presented by display interface.
Compared with prior art, the positive effect of the present invention is:
The present invention is identified by the acquisition to electromagnetic leakage signal, is realized to illegal access device and unit exception row
For monitoring alarm.It is unevenly distributed in time, space and frequency for on-site signal and unstable feature, differentiation is live
With/without leakage signal, the intelligence and accuracy rate for threatening and differentiating are improved.It is multiple for the electromagnetic leakage signal of multiple and different equipment
Miscellaneous changeable feature, is identified with/without abnormal signal, is analyzed " responsibility " main body, is determined whether the access of illegality equipment.For
The characteristics of electromagnetic signature of the individual equipment under different configuration status or operating mode can be according to different operation behavior dynamic change,
Identify operation behavior corresponding to equipment leakage signal.The present invention can timely update existing feature samples library
With expansion, solve repetition learning and demand of the depth network training to great amount of samples.
Detailed description of the invention
Examples of the present invention will be described by way of reference to the accompanying drawings, in which:
Fig. 1 is a kind of block diagram of abnormal behaviour identification device based on electromagnetic leakage signal；
Fig. 2 is electromagnetic leakage signal intelligent classification identification device module flow diagram；
Fig. 3 is scene with/without leakage signal intelligent classification identification process flow chart；
Fig. 4 is scene with/without abnormal signal intelligent classification identification process flow chart；
Fig. 5 is critical behavior leakage signal intelligent classification identification process flow chart.
Specific embodiment
A kind of abnormal behaviour identification device based on electromagnetic leakage signal, as shown in Figure 1, including following module:
1 signal acquisition module
Signal acquisition module is acquired the leakage signal of target device or route periphery by coupling probe or antenna
It receives, generates the electromagnetic leakage signal picture comprising information such as time, signal strength, signal frequencies, frequency domain-maximum value is supported to protect
The types such as figure, time-frequency-Waterfall plot, time domain-intensity map are held, and are established based on one or more picture types with/without leakage signal
Sample database, with/without abnormal signal sample database and critical behavior leakage signal sample database.
2 intelligent classification identification modules
Intelligent classification identification module mainly includes to scene with/without leakage signal, with/without abnormal signal and to critical behavior
The intelligent recognition and classification of leakage signal realize the monitoring alarm to illegal access device and unit exception behavior, process ginseng
According to Fig. 2.Firstly, identifying to scene with/without leakage signal, if there is leakage signal, for individual equipment, leakage signal is identified
Corresponding operation behavior；For multiple and different equipment, identify with/without abnormal signal, if there is abnormal signal appearance, judge be
The no access for having illegality equipment is identified according to individual equipment critical behavior to abnormal signal institute if the access without illegality equipment
Corresponding operation behavior is identified, is finally fed back to result and is differentiated that result is presented module and carries out result presentation.
1) scene is with/without leakage signal intelligent classification identification module
Scene with/without leakage signal intelligent classification identification process flow chart referring to shown in Fig. 3, the judging process packet of the module
Include following steps:
It is pre-processed Step 1: scene is classified with/without leakage signal picture tag:
It is all unevenly distributed in time, space and frequency for on-site signal and unstable feature, by collecting
Live electromagnetic background and leakage signal one or more images carry out labeling pretreatment, by electromagnetic background signal label
" No leakage signal " classification is turned to, electromagnetic leakage signal label is turned into " having leakage signal " classification.
Step 2: whether there is or not the foundation of leakage signal sample database:
By the electromagnetic leakage signal of the electromagnetic background signal of " No leakage signal " classification and " having leakage signal " classification, respectively
It is stored under different files, establishes " No leakage signal " sample database and " having leakage signal " sample database.
Step 3: training process:
Using " No leakage signal " sample after labeling with " having leakage signal " sample as training picture input
To training process, training program is called to be trained training picture, by repetition learning and depth network training, it is accurate to be formed
" No leakage signal " and " having leakage signal " classification results model.
Step 4: identification process:
It regard test picture as input, operation recognizer identifies picture, using trained classification results mould
Type makes decisions, and provides similarity respectively for " No leakage signal " and " having leakage signal " two kinds of situations, and by similarity knot
Fruit output.If the test picture of input does not meet the feature of original sample database, that is, it is regarded as new sample data, it will be new
Samples pictures are added in " No leakage signal " or the sample database of " having leakage signal ", are expanded available sample library, if belonging to
In " having leakage signal " classification, warning function need to be started simultaneously.
2) scene is with/without abnormal signal intelligent classification identification module
Scene is with/without abnormal signal intelligent classification identification process flow chart referring to shown in Fig. 3.The module is mainly for multiple
The electromagnetic leakage signal of distinct device identifies with/without abnormal signal, analyzes " responsibility " main body, determine whether illegality equipment
Access.The judging process of the module includes the following steps:
It is pre-processed Step 1: classifying with/without abnormal signal picture tag:
By collection in worksite to one or more normal signal labels turn to " being no different regular signal " classification, actively swash live
It sends out and the one or more abnormal signal labels acquired turns to " having abnormal signal " classification.
Step 2: whether there is or not the foundation of abnormal signal sample database:
The normal signal for the classification that " will be no different regular signal ", the abnormal signal with " having abnormal signal " classification are respectively stored in
Under different files, " being no different regular signal " sample database and " having abnormal signal " sample database are established.
Step 3: training process:
Using " being no different regular signal " sample after labeling with " having abnormal signal " sample as training picture input
To training process, training program is called to be trained training picture, by repetition learning and depth network training, it is accurate to be formed
" being no different regular signal " and " having abnormal signal " classification results model.
Step 4: identification process:
It regard test picture as input, operation recognizer identifies picture, using trained classification results mould
Type makes decisions, and provides similarity respectively for " being no different regular signal " and " having abnormal signal " two kinds of situations, and by similarity knot
Fruit output.If it was found that the new abnormal signal for having illegal access device to generate, starts warning function and expands available sample library.
3) critical behavior leakage signal intelligent classification identification module
Electromagnetic leakage signal critical behavior intelligent classification identification process flow chart referring to Figure 5, the judgement of the module
Journey includes the following steps:
Step 1: the classification pretreatment of electromagnetic leakage signal picture tag:
It can be dynamic according to different operation behavior for electromagnetic signature of the individual equipment under different configuration status or operating mode
The characteristics of state changes carries out labeling pretreatment to leakage signal corresponding to equipment operation behavior.Pass through intelligent acquisition skill
Art, by signals such as the different working condition effectively extracted and different working modes, label turns to " critical behavior 1, crucial row respectively
Being 2 ..., critical behavior n " is different classes of.
Step 2: critical behavior leakage signal sample database is established:
By " critical behavior 1, critical behavior 2 ... critical behavior n " different classes of signal, it is respectively stored in different files
Under, establish " critical behavior leakage signal " sample database.
Step 3: training process:
Using the sample of " critical behavior 1, critical behavior 2 ... critical behavior n " after labeling as training picture
It is input to training process, training program is called to be trained training picture, by repetition learning and depth network training, is formed
The classification results model of accurate critical behavior leakage signal.
Step 4: identification process:
It regard test picture as input, operation recognizer identifies picture, using trained classification results mould
Type makes decisions, and situations a variety of for " critical behavior 1, critical behavior 2 ... critical behavior n " provide similarity respectively, and by phase
Like degree result output.If starting warning function it was found that there is abnormal operation behavior and carrying out warning reminding, and timely expand existing
There is feature samples library, updates the classification results model of critical behavior leakage signal.
3 differentiate that module is presented in result
Differentiate that module is presented in result, after Classification and Identification, the result of differentiation is clearly presented to by display interface
User respectively includes with/without leakage signal, with/without abnormal signal, the specific alignment similarity result of critical behavior leakage signal
Presentation.
The above is only a preferred embodiment of the present invention, not restrictive meaning, it is noted that for this
For those skilled in the art, several improvement can also be made without departing from the principle of the present invention, these improvement
Also it should be regarded as protection scope of the present invention.
Claims (10)
1. a kind of abnormal behaviour identification device based on electromagnetic leakage signal, it is characterised in that: adopted including sequentially connected signal
Collect module, intelligent classification identification module and differentiate that module is presented in result, in which: the signal acquisition module is for acquiring electromagnetism back
Scape signal, electromagnetic leakage signal, normal signal, abnormal signal, critical behavior leakage signal simultaneously generate electromagnetic leakage signal image
Send intelligent classification identification module to；The intelligent classification identification module is responsible for having scene with/without leakage signal, to scene/
It is no different regular signal and intelligent recognition is carried out to critical behavior leakage signal, and recognition result is sent to and differentiates that mould is presented in result
Block；Module is presented for the recognition result of intelligent classification identification module to be presented in the differentiation result.
2. a kind of abnormal behaviour identification device based on electromagnetic leakage signal according to claim 1, it is characterised in that: institute
Stating electromagnetic leakage signal image includes electromagnetic leakage signal frequency domain-maximum value holding figure, time-frequency-Waterfall plot, time domain-intensity map.
3. a kind of abnormal behaviour identification device based on electromagnetic leakage signal according to claim 1, it is characterised in that: institute
It states intelligent classification identification module first to identify scene with/without leakage signal, if there is leakage signal, is directed to individual equipment
It identifies operation behavior corresponding to leakage signal, identifies for multiple equipment with/without abnormal signal, if there is abnormal signal appearance,
The access for whether having illegality equipment is then further judged, if the access without illegality equipment, according to individual equipment critical behavior
Identification operation behavior corresponding to abnormal signal is identified, finally by recognition result feed back to differentiate result present module
It is presented.
4. a kind of abnormal behaviour identification device based on electromagnetic leakage signal according to claim 1, it is characterised in that: institute
State differentiate result present module by display interface to scene with/without leakage signal, scene with/without abnormal signal, critical behavior
Specific alignment similarity carry out result presentation.
5. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal, it is characterised in that: including following content:
One, signal acquisition:
Reception is acquired to the leakage signal of target device or route periphery by coupling probe or antenna, when generation includes
Between, the electromagnetic leakage signal pictures of the information such as signal strength, signal frequency；
Two, intelligent classification identifies:
Firstly, identifying to scene with/without leakage signal, if there is leakage signal, for individual equipment, leakage signal institute is identified
Corresponding operation behavior；For multiple and different equipment, identify with/without abnormal signal, if there is abnormal signal appearance, judge whether
There is the access of illegality equipment, it is right to abnormal signal institute according to the identification of individual equipment critical behavior if the access without illegality equipment
The operation behavior answered is identified；
Three, recognition result is presented:
After Classification and Identification, by with/without leakage signal, with/without abnormal signal, the specific comparison phase of critical behavior leakage signal
It is presented like degree result by display interface.
6. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal according to claim 5, it is characterised in that: right
Scene includes: with/without the process that leakage signal carries out intelligent recognition
It is pre-processed Step 1: scene is classified with/without leakage signal picture tag:
Electromagnetic background signal label is turned into " No leakage signal " classification, electromagnetic leakage signal label is turned into " having leakage signal "
Classification；
Step 2: whether there is or not the foundation of leakage signal sample database:
By the electromagnetic leakage signal of the electromagnetic background signal of " No leakage signal " classification and " having leakage signal " classification, store respectively
Under different files, " No leakage signal " sample database and " having leakage signal " sample database are established；
Step 3: training process:
" No leakage signal " sample after labeling is input to instruction as training picture with " having leakage signal " sample
Practice process, training program is called to be trained training picture, by repetition learning and depth network training, is formed accurate
The classification results model of " No leakage signal " and " having leakage signal "；
Step 4: identification process:
Will test picture as input, operation recognizer picture is identified, using trained classification results model into
" No leakage signal " and " having leakage signal " two kinds of situations are provided similarity, and similarity result is defeated by row judgement respectively
Out.
7. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal according to claim 6, it is characterised in that: such as
The test picture of fruit input does not meet the feature of original sample database, then by new samples pictures be added to " No leakage signal " or
In the sample database of " having leakage signal ", available sample library is expanded, if belonging to " having leakage signal " classification, need to be started simultaneously
Warning function.
8. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal according to claim 6, it is characterised in that: right
Scene includes: with/without the process that abnormal signal carries out intelligent recognition
It is pre-processed Step 1: classifying with/without abnormal signal picture tag:
By collection in worksite to one or more normal signal labels turn to " being no different regular signal " classification, simultaneously by live actively excitation
One or more abnormal signal labels of acquisition turn to " having abnormal signal " classification；
Step 2: whether there is or not the foundation of abnormal signal sample database:
The normal signal for the classification that " will be no different regular signal ", the abnormal signal with " having abnormal signal " classification are respectively stored in difference
Under file, " being no different regular signal " sample database and " having abnormal signal " sample database are established；
Step 3: training process:
" being no different regular signal " sample after labeling is input to instruction as training picture with " having abnormal signal " sample
Practice process, training program is called to be trained training picture, by repetition learning and depth network training, is formed accurate
The classification results model of " being no different regular signal " and " having abnormal signal "；
Step 4: identification process:
Will test picture as input, operation recognizer picture is identified, using trained classification results model into
" being no different regular signal " and " having abnormal signal " two kinds of situations are provided similarity, and similarity result is defeated by row judgement respectively
Out；If it was found that the new abnormal signal for having illegal access device to generate, starts warning function and expands available sample library.
9. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal according to claim 8, it is characterised in that: right
Critical behavior leakage signal carry out intelligent recognition process include:
Step 1: the classification pretreatment of electromagnetic leakage signal picture tag:
By signal that individual equipment extracts under different working condition and different working modes respectively label turn to " critical behavior 1,
Critical behavior 2 ..., critical behavior n " it is different classes of；
Step 2: critical behavior leakage signal sample database is established:
Will " critical behavior 1, critical behavior 2 ..., critical behavior n " different classes of signal, be respectively stored in different files
Under, establish " critical behavior leakage signal " sample database；
Step 3: training process:
Using the sample of " critical behavior 1, critical behavior 2 ... critical behavior n " after labeling as training picture input
To training process, training program is called to be trained training picture, by repetition learning and depth network training, it is accurate to be formed
Critical behavior leakage signal classification results model；
Step 4: identification process:
Will test picture as input, operation recognizer picture is identified, using trained classification results model into
Row judgement, it is different classes of for " critical behavior 1, critical behavior 2 ... critical behavior n " to provide similarity respectively, and by similarity
As a result it exports；If starting warning function it was found that there is abnormal operation behavior and carrying out warning reminding, and extended to existing spy
Sample database is levied, to update the classification results model of critical behavior leakage signal.
10. a kind of abnormal behaviour recognition methods based on electromagnetic leakage signal according to claim 5, it is characterised in that:
The electromagnetic leakage signal picture supports frequency domain-maximum value to keep the types such as figure, time-frequency-Waterfall plot, time domain-intensity map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910308245.3A CN110163099B (en) | 2019-04-17 | 2019-04-17 | Device and method for identifying abnormal behaviors based on electromagnetic leakage signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910308245.3A CN110163099B (en) | 2019-04-17 | 2019-04-17 | Device and method for identifying abnormal behaviors based on electromagnetic leakage signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110163099A true CN110163099A (en) | 2019-08-23 |
| CN110163099B CN110163099B (en) | 2022-04-01 |
Family
ID=67639523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910308245.3A Active CN110163099B (en) | 2019-04-17 | 2019-04-17 | Device and method for identifying abnormal behaviors based on electromagnetic leakage signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163099B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688652A (en) * | 2019-08-30 | 2020-01-14 | 中国电子科技网络信息安全有限公司 | USB micro electromagnetic leakage signal source |
| CN111351992A (en) * | 2020-02-20 | 2020-06-30 | 南方电网科学研究院有限责任公司 | Chip security assessment method based on electromagnetic side channel analysis and related device |
| CN111949988A (en) * | 2020-07-03 | 2020-11-17 | 中国电子科技网络信息安全有限公司 | Active electromagnetic vulnerability detection system and method for physically isolated computer |
| CN112669316A (en) * | 2021-01-29 | 2021-04-16 | 南方电网调峰调频发电有限公司 | Power production abnormity monitoring method and device, computer equipment and storage medium |
| CN115272983A (en) * | 2022-09-29 | 2022-11-01 | 成都中轨轨道设备有限公司 | Contact net suspension state monitoring method and system based on image recognition |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000185634A (en) * | 1998-12-22 | 2000-07-04 | Nissan Motor Co Ltd | Brake control device |
| US20130314105A1 (en) * | 2012-05-04 | 2013-11-28 | Apple Inc. | Electronic device including device ground coupled finger coupling electrode and array shielding electrode and related methods |
| CN103926912A (en) * | 2014-05-07 | 2014-07-16 | 桂林赛普电子科技有限公司 | Smart home monitoring system based on home service robot |
| CN104854469A (en) * | 2013-03-06 | 2015-08-19 | 独立行政法人交通安全环境研究所 | Magnetism detection device |
| US20150360048A1 (en) * | 2013-02-07 | 2015-12-17 | Rocomp Global, Llc | Electromagnetic radiation targeting devices, assemblies, systems and methods |
| US9349098B1 (en) * | 2015-05-14 | 2016-05-24 | James Albert Ionson | Cognitive medical and industrial inspection system and method |
| CN205721244U (en) * | 2016-06-08 | 2016-11-23 | 徐全琦 | Intelligent cruise positioning number adopts control Operating Guideline management system |
| CN108631893A (en) * | 2017-03-17 | 2018-10-09 | 成都信息工程大学 | A kind of GMSK channel disturbances detection method |
| CN108680796A (en) * | 2018-05-17 | 2018-10-19 | 集美大学 | Electromagnetic information leakage detecting system and method for computer display |
| CN109633289A (en) * | 2018-12-28 | 2019-04-16 | 集美大学 | A kind of red information detecting method of electromagnetism based on cepstrum and convolutional neural networks |
-
2019
- 2019-04-17 CN CN201910308245.3A patent/CN110163099B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000185634A (en) * | 1998-12-22 | 2000-07-04 | Nissan Motor Co Ltd | Brake control device |
| US20130314105A1 (en) * | 2012-05-04 | 2013-11-28 | Apple Inc. | Electronic device including device ground coupled finger coupling electrode and array shielding electrode and related methods |
| US20150360048A1 (en) * | 2013-02-07 | 2015-12-17 | Rocomp Global, Llc | Electromagnetic radiation targeting devices, assemblies, systems and methods |
| CN104854469A (en) * | 2013-03-06 | 2015-08-19 | 独立行政法人交通安全环境研究所 | Magnetism detection device |
| CN103926912A (en) * | 2014-05-07 | 2014-07-16 | 桂林赛普电子科技有限公司 | Smart home monitoring system based on home service robot |
| US9349098B1 (en) * | 2015-05-14 | 2016-05-24 | James Albert Ionson | Cognitive medical and industrial inspection system and method |
| CN205721244U (en) * | 2016-06-08 | 2016-11-23 | 徐全琦 | Intelligent cruise positioning number adopts control Operating Guideline management system |
| CN108631893A (en) * | 2017-03-17 | 2018-10-09 | 成都信息工程大学 | A kind of GMSK channel disturbances detection method |
| CN108680796A (en) * | 2018-05-17 | 2018-10-19 | 集美大学 | Electromagnetic information leakage detecting system and method for computer display |
| CN109633289A (en) * | 2018-12-28 | 2019-04-16 | 集美大学 | A kind of red information detecting method of electromagnetism based on cepstrum and convolutional neural networks |
Non-Patent Citations (4)
| Title |
|---|
| DING L 等: ""Specific Emitter Identification Via Convolutional Neural Networks"", 《IEEE COMMUNICATIONS LEUERS》 * |
| Z. WEISHA等: ""Machine Learning-Based System Electromagnetic Environment Anomaly Detection Method"", 《 2018 INTERNATIONAL CONFERENCE ON SMART GRID AND ELECTRICAL AUTOMATION (ICSGEA)》 * |
| 寇云峰 等: ""一种针对电磁传导泄漏威胁的现场快速检测系统"", 《通信技术》 * |
| 杨豪璞 等: ""基于阶段特性的APT攻击行为分类与评估方法"", 《计算机工程与应用》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688652A (en) * | 2019-08-30 | 2020-01-14 | 中国电子科技网络信息安全有限公司 | USB micro electromagnetic leakage signal source |
| CN111351992A (en) * | 2020-02-20 | 2020-06-30 | 南方电网科学研究院有限责任公司 | Chip security assessment method based on electromagnetic side channel analysis and related device |
| CN111949988A (en) * | 2020-07-03 | 2020-11-17 | 中国电子科技网络信息安全有限公司 | Active electromagnetic vulnerability detection system and method for physically isolated computer |
| CN111949988B (en) * | 2020-07-03 | 2022-04-22 | 中国电子科技网络信息安全有限公司 | Active electromagnetic vulnerability detection system and method for physically isolated computer |
| CN112669316A (en) * | 2021-01-29 | 2021-04-16 | 南方电网调峰调频发电有限公司 | Power production abnormity monitoring method and device, computer equipment and storage medium |
| CN115272983A (en) * | 2022-09-29 | 2022-11-01 | 成都中轨轨道设备有限公司 | Contact net suspension state monitoring method and system based on image recognition |
| CN115272983B (en) * | 2022-09-29 | 2023-01-03 | 成都中轨轨道设备有限公司 | Contact net suspension state monitoring method and system based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110163099B (en) | 2022-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163099A (en) | A kind of abnormal behaviour identification device and method based on electromagnetic leakage signal | |
| CN106897738B (en) | A kind of pedestrian detection method based on semi-supervised learning | |
| CN111898523A (en) | Remote sensing image special vehicle target detection method based on transfer learning | |
| CN105787472B (en) | A kind of anomaly detection method based on the study of space-time laplacian eigenmaps | |
| CN109002744A (en) | Image-recognizing method, device and video monitoring equipment | |
| Mangalam et al. | Do deep neural networks learn shallow learnable examples first? | |
| CN108875624A (en) | Method for detecting human face based on the multiple dimensioned dense Connection Neural Network of cascade | |
| Korschens et al. | Elpephants: A fine-grained dataset for elephant re-identification | |
| CN112633071B (en) | Pedestrian re-identification data domain adaptation method based on data style decoupling content migration | |
| CN109919106A (en) | Gradual target finely identifies and description method | |
| CN109800717A (en) | Activity recognition video frame sampling method and system based on intensified learning | |
| CN112270331A (en) | Improved billboard detection method based on YOLOV5 | |
| CN110287856A (en) | A kind of security personnel's behavior analysis system, method and device | |
| CN106537387B (en) | Retrieval/storage image associated with event | |
| CN109920404A (en) | Possess the information collecting device and acquisition method of the automatic selling Advertising Management System of Intellisense effect | |
| CN109145843A (en) | A kind of full vehicle information identification system of bayonet high definition camera and method | |
| CN108537129A (en) | The mask method of training sample, device and system | |
| CN110210382A (en) | A kind of face method for detecting fatigue driving and device based on space-time characteristic identification | |
| CN104978569A (en) | Sparse representation based incremental face recognition method | |
| CN112528774A (en) | Intelligent sorting system and method for unknown radar signals in complex electromagnetic environment | |
| CN113312924A (en) | Risk rule classification method and device based on NLP high-precision analysis label | |
| CN115083229B (en) | Intelligent recognition and warning system of flight training equipment based on AI visual recognition | |
| CN111191498A (en) | Behavior recognition method and related product | |
| EP4167128A1 (en) | Signal analysis method and system based on model for acquiring and identifying noise panoramic distribution | |
| Sain et al. | Outlier detection from a mixture distribution when training data are unlabeled |
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 |