CN109639479A - Based on the network flow data Enhancement Method and device for generating confrontation network - Google Patents
Based on the network flow data Enhancement Method and device for generating confrontation network Download PDFInfo
- Publication number
- CN109639479A CN109639479A CN201811496710.2A CN201811496710A CN109639479A CN 109639479 A CN109639479 A CN 109639479A CN 201811496710 A CN201811496710 A CN 201811496710A CN 109639479 A CN109639479 A CN 109639479A
- Authority
- CN
- China
- Prior art keywords
- data
- flows
- enhancing
- network model
- confrontation network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the present invention provides a kind of based on the network flow data Enhancement Method and device that generate confrontation network, which comprises obtains the data set of live network flow in target scene;According to the data set, it is trained to confrontation network model is generated;Network model is fought based on the trained generation, obtains final enhancing data on flows.The embodiment of the present invention can be adapted for various scenes, without data on flows in terms of expertise, adaptively realize data on flows enhancing, expanded the data set of network flow, improve using machine learning method optimization network performance effect.
Description
Technical field
The embodiment of the present invention belongs to technical field of communication network, fights network based on generation more particularly, to a kind of
Network flow data Enhancement Method and device.
Background technique
In recent years, the machine learning using network flow data as training set (Machine Learning, ML) is answered extensively
It uses in traffic classification, Traffic anomaly detection and optimization of network performance.In order to improve the effect of machine learning, a large amount of net is needed
Network data on flows.
However, being difficult to be collected into effective network flow data in reality, the shortage of data on flows affects engineering
The application effect of habit.Therefore, it is necessary to use rationally effective network flow data Enhancement Method resultant flow data.Lead to currently
Under communication network background, a large amount of different types of services cause the diversity of network flow data, this requires network flow data
Synthesis need dynamic with higher and flexibility.Traditional network flow data synthetic method includes parting Brownian movement
(Fractal Brown Motion, FBM), M/G/ ∞ queuing model, the model based on small echo and multi-fractal wavelet model
(Multifractal Wavelet Model,MWM).These conventional methods need to rely on a large amount of expertise establish it is different
Data on flows model, and need to rely on different expertises for different flow scenes.Since expertise has very by force
Subjectivity, cause rely on expertise synthesis data on flows and true data on flows between there are bigger differences.
In conclusion existing network flow data synthetic method is not applied for all flow scenes, universality is poor, and
Data on flows model is established dependent on expertise, there is very strong subjectivity, the data on flows of synthesis is caused to lack authenticity.
Summary of the invention
To overcome above-mentioned existing network flow data synthetic method universality poor, the data on flows of synthesis lacks authenticity
The problem of or at least be partially solved the above problem, the embodiment of the present invention provides a kind of based on the network flow for generating confrontation network
Measure data enhancement methods and device.
According to a first aspect of the embodiments of the present invention, it provides a kind of based on the network flow data enhancing for generating confrontation network
Method, comprising:
Obtain the data set of live network flow in target scene;
According to the data set, it is trained to confrontation network model is generated;
Network model is fought based on the trained generation, obtains final enhancing data on flows.
Second aspect according to embodiments of the present invention provides a kind of based on the network flow data enhancing dress for generating confrontation network
It sets, comprising:
Module is obtained, for obtaining the data set of live network flow in target scene;
Training module, for being trained to confrontation network model is generated according to the data set;
Output module obtains final enhancing data on flows for fighting network model based on the trained generation.
In terms of third according to an embodiment of the present invention, a kind of electronic equipment is also provided, comprising:
At least one processor;And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
Order is able to carry out in the various possible implementations of first aspect provided by any possible implementation based on generation
Fight the network flow data Enhancement Method of network.
4th aspect according to an embodiment of the present invention, also provides a kind of non-transient computer readable storage medium, described
Non-transient computer readable storage medium stores computer instruction, and the computer instruction makes the computer execute first aspect
Various possible implementations in provided by any possible implementation based on the network flow for generating confrontation network
Data enhancement methods.
The embodiment of the present invention provides a kind of based on the network flow data Enhancement Method and device that generate confrontation network, the party
Method is trained by the way that the data set input of live network flow in target scene is generated confrontation network model, is up to convergence
When the enhancing data on flows that generates of confrontation network model as final enhancing data on flows, the present embodiment can be adapted for various fields
Scape, without data on flows in terms of expertise, adaptively realize data on flows enhancing, expanded the data of network flow
Collection improves the effect using machine learning method optimization network performance.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is provided in an embodiment of the present invention based on the network flow data Enhancement Method overall flow for generating confrontation network
Schematic diagram;
Fig. 2 is that the network flow data Enhancement Method based on generation confrontation network that further embodiment of this invention provides is whole
Flow diagram;
Fig. 3 is provided in an embodiment of the present invention based on the network flow data enhancement device overall structure for generating confrontation network
Schematic diagram;
Fig. 4 is electronic equipment overall structure diagram provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
A kind of network flow data Enhancement Method based on generation confrontation network is provided in one embodiment of the invention,
Fig. 1 is the network flow data Enhancement Method overall flow schematic diagram provided in an embodiment of the present invention based on generation confrontation network,
This method comprises: S101, obtains the data set of live network flow in target scene;
Wherein, target scene is the scene for needing to carry out network flow data enhancing, can be various scenes, such as resident
Area, school zone and shopping centre etc..Using the network flow data directly acquired from target scene as live network data on flows.
Data set by the combination of multiple samples of live network data on flows as live network flow.
S102 is trained according to the data set to confrontation network model is generated;
Generating confrontation network (Generative Adversarial Nets, GANs) is the core in deep learning field
Algorithm is widely used in the fields such as image procossing and video generation, benefits from zero-sum game thought, the principle of GANs is by two
The dual training of a neural network, to realize effective data enhancing.Using data set to generate confrontation network model into
Before row training, the live network data on flows first concentrated to data is pre-processed, such as normalized.It then will pretreatment
In the GANs model that data set input afterwards is built, GANs model is trained, convergence is reached.GANs model includes
One generator and an arbiter.Wherein, generator makees enhancing processing result for carrying out enhancing processing to random noise
Enhance data on flows for centre.Arbiter is for distinguishing intermediate enhancing data on flows and real traffic data.
S103 fights network model based on the trained generation, obtains final enhancing data on flows.
The intermediate enhancing data traffic that generator exports after confrontation network model is restrained, which will be generated, enhances flow as final
Data.
The present embodiment is carried out by the way that the data set input of live network flow in target scene is generated confrontation network model
Training, the enhancing data on flows that confrontation network model generates when being up to convergence is as final enhancing data on flows, the present embodiment
Can be adapted for various scenes, without data on flows in terms of expertise, adaptively realize data on flows enhancing, expand
The data set of network flow improves the effect using machine learning method optimization network performance.
On the basis of the above embodiments, the step of the data set of live network flow in target scene is obtained in the present embodiment
Suddenly it specifically includes: obtaining the live network data on flows of target scene presetting granularity in default historical time section;It will be described default
Historical time section is divided into multiple sub- periods, and the live network data on flows of each presetting granularity in each sub- period is made
For a sample;Data set by the set of all samples as live network flow in the target scene.
Specifically, the real traffic data in default historical time section under target scene are collected, such as by residential block, school zone
Target scene is used as with one in shopping centre.According to the presetting granularity of real traffic data, such as millisecond, second, minute or hour,
It selects daily, real traffic data weekly or monthly as a sample, all real traffic data samples is merged into
Data set.And the real traffic data in sample each in target scene are normalized.In order to which preferably training generates
Confrontation network makes its fast convergence, and the data set in target scene is carried out minimax normalized.For example, default history
Period when it is 44 days a length of, each sub- period when it is one day a length of, using real traffic data total per hour as one
Granularity, will be daily, i.e., real traffic data hourly are as a sample in 24 hours, in the data set of the target scene altogether
There are 44 samples.
On the basis of the above embodiments, it is carried out according to the data set to confrontation network model is generated in the present embodiment
Trained step specifically includes: enhancing processing is carried out to random noise based on the generator in the generation confrontation network model,
Obtain intermediate enhancing data on flows;Based on the arbiter generated in confrontation network model to the intermediate enhancing data on flows
Differentiated with the sample in data set described after pretreatment;According to differentiate result to the parameter of the generator and arbiter into
Row adjustment, until generation confrontation network model convergence.
As shown in Fig. 2, generating confrontation network mainly includes a generator and an arbiter.Generator and arbiter are all
It is deep neural network (Deep Neural Networks, DNN) structure, including an input layer, one or more hidden layers,
An and output layer.Pretreated live network data on flows collection will be passed through as the input of arbiter, the input of generator
For random noise, enhancing result of the generator to random noise is enhanced into data on flows as centre.The training initial stage generates
The enhancing ability of device is poor, and the data on flows of enhancing does not meet the distribution of real traffic data.Random noise after generator,
The enhancing data on flows and true data on flows of output are input to arbiter, and arbiter is easy to determine enhancing data on flows
With real traffic data.The parameter of generator and arbiter is adjusted according to differentiation result, after iteration is based on adjusting parameter
Generator to random noise carry out enhancing processing and based on arbiter differentiate enhancing data on flows and real traffic data.With
The continuous progress of training, the enhancing ability of generator is more and more stronger, and the intermediate enhancing data on flows of generation increasingly meets very
The distribution of real data on flows, arbiter are difficult to differentiate between the data on flows and true data on flows of enhancing.When GANs model is restrained
Afterwards, it is 0.5 that arbiter, which judges that its input belongs to real traffic data and enhances the probability of data on flows,.Generator is in convergence
Distribution of the final enhancing data on flows distribution of output close to real traffic data.
On the basis of the above embodiments, in the present embodiment based on it is described generate confrontation network model in generator to
The step of machine noise carries out enhancing processing, obtains intermediate enhancing data on flows specifically includes: random noise is inputted the generation
The input layer of device;Enhancing processing carried out to the random noise based on one or more hidden layers of the generator, in acquisition
Between enhance data on flows;The intermediate enhancing data on flows is input to the arbiter by the output layer based on the generator
In.
Specifically, generator includes an input layer, one or more hidden layers and an output layer.It will be by pre-
The real traffic data of processing are input in arbiter, and the input of the input layer of generator is random noise, by hidden layer meter
After calculating transmission, output layer can export the data on flows of enhancing, and be entered into arbiter, and the stream of enhancing is distinguished by arbiter
Measure data and true data on flows.The hidden layer of generator can be full articulamentum, such as m full articulamentums, generator it is defeated
Enter layer to be connected with full articulamentum, be sequentially connected with by m full articulamentums, the output layer of the last one full articulamentum and generator
It is connected entirely, the activation primitive of full articulamentum is line rectification function (Rectified Linear Unit, ReLU).It generates
The data on flows of the output layer output enhancing of device, i.e., it is intermediate to enhance data on flows.The output layer of arbiter exports its input and belongs to
The probability of real traffic data and enhancing data on flows.The intermediate enhancing that generator exports after confrontation network model is restrained will be generated
Data traffic is as final enhancing data on flows.
According to the common statistical property of data on flows, such as mean value, variance and Hurst index (Hurst exponent, Hess
Refer in particular to count), data traffic and true data on flows after Contrast enhanced, the validity of the data traffic of verifying analysis enhancing with
And the adaptivity in the enhancing of different scenes down-off data.As shown in table 1, under three kinds of different scenes, the flow number of enhancing
It is very close according to mean value, variance and Hurst index between true data on flows, illustrate that the data on flows distribution of enhancing is close
The distribution of real traffic data, and can be adapted for various different scenes.
The data on flows effect enhanced under 1 different scenes of table
It provides in another embodiment of the present invention a kind of based on the network flow data enhancing dress for generating confrontation network
It sets, the device is for realizing the method in foregoing embodiments.Therefore, aforementioned based on the network flow number for generating confrontation network
According to the description and definition in each embodiment of Enhancement Method, it can be used for the understanding of each execution module in the embodiment of the present invention.
Fig. 3 is the network flow data enhancement device overall structure diagram provided in an embodiment of the present invention based on generation confrontation network,
The device includes obtaining module 301, training module 302 and output module 303;Wherein:
Obtain the data set that module 301 is used to obtain live network flow in target scene;
Wherein, target scene is the scene for needing to carry out network flow data enhancing, can be various scenes, such as resident
Area, school zone and shopping centre etc..Module 301 is obtained using the network flow data directly acquired from target scene as true net
Network data on flows.Data set by the combination of multiple samples of live network data on flows as live network flow.
Training module 302 is used to be trained according to the data set to confrontation network model is generated;
GANs model is the core algorithm in deep learning field, is widely used in the neck such as image procossing and video generation
Domain, benefits from zero-sum game thought, and the principle of GANs is by the dual training of two neural networks, to realize effective number
According to enhancing.Before being trained using data set to generation confrontation network model, first to the live network flow of data concentration
Data are pre-processed, such as normalized.Then pretreated data set is inputted the GANs built by training module 302
In model, GANs model is trained, reaches convergence.GANs model includes a generator and an arbiter.Its
In, generator enhances data on flows for carrying out enhancing processing to random noise, using enhancing processing result as centre.Arbiter
For distinguishing intermediate enhancing data on flows and real traffic data.
Output module 303 is used to fight network model based on the trained generation, obtains final enhancing data on flows.
Output module 303 will generate the intermediate enhancing data traffic of generator output after confrontation network model is restrained as most
Enhancing data on flows eventually.
The present embodiment is carried out by the way that the data set input of live network flow in target scene is generated confrontation network model
Training, the enhancing data on flows that confrontation network model generates when being up to convergence is as final enhancing data on flows, the present embodiment
Can be adapted for various scenes, without data on flows in terms of expertise, adaptively realize data on flows enhancing, expand
The data set of network flow improves the effect using machine learning method optimization network performance.
On the basis of the above embodiments, module is obtained in the present embodiment to be specifically used for: obtaining in default historical time section
The live network data on flows of target scene presetting granularity;The default historical time section is divided into multiple sub- periods, it will
The live network data on flows of each presetting granularity is as a sample in each sub- period;By the set of all samples
Data set as live network flow in the target scene.
It on the basis of the above embodiments, further include preprocessing module in the present embodiment, for being carried out to the data set
Pretreatment, the pretreatment include minimax normalized.
On the basis of the above embodiments, training module is specifically used in the present embodiment: fighting network based on the generation
Generator in model carries out enhancing processing to random noise, obtains intermediate enhancing data on flows;Net is fought based on the generation
Arbiter in network model differentiates the sample in the data set after the intermediate enhancing data on flows and pretreatment;Root
According to differentiate result to the generator and arbiter parameter be adjusted, until generations confrontation network model restrain.
On the basis of the above embodiments, training module is further used in the present embodiment: described in random noise is inputted
The input layer of generator;One or more hidden layers based on the generator carry out enhancing processing to the random noise, obtain
Take intermediate enhancing data on flows;The intermediate enhancing data on flows is input to the differentiation by the output layer based on the generator
In device.
On the basis of the above embodiments, hidden layer described in the present embodiment is full articulamentum, and the full articulamentum swashs
Function living is line rectification function.
On the basis of the above embodiments, output module is specifically used in the present embodiment: the generation is fought network mould
The intermediate enhancing data traffic that generator exports after type convergence is as final enhancing data on flows.
The present embodiment provides a kind of electronic equipment, Fig. 4 is electronic equipment overall structure provided in an embodiment of the present invention signal
Figure, which includes: at least one processor 401, at least one processor 402 and bus 403;Wherein,
Processor 401 and memory 402 pass through bus 403 and complete mutual communication;
Memory 402 is stored with the program instruction that can be executed by processor 401, and the instruction of processor caller is able to carry out
Method provided by above-mentioned each method embodiment, for example, obtain the data set of live network flow in target scene;According to
The data set is trained to confrontation network model is generated;Network model is fought based on the trained generation, is obtained most
Enhancing data on flows eventually.
The present embodiment provides a kind of non-transient computer readable storage medium, non-transient computer readable storage medium storages
Computer instruction, computer instruction make computer execute method provided by above-mentioned each method embodiment, for example, obtain mesh
Mark the data set of live network flow in scene;According to the data set, it is trained to confrontation network model is generated;Based on instruction
The generation confrontation network model perfected, obtains final enhancing data on flows.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed;And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of based on the network flow data Enhancement Method for generating confrontation network, which is characterized in that
Include:
Obtain the data set of live network flow in target scene;
According to the data set, it is trained to confrontation network model is generated;
Network model is fought based on the trained generation, obtains final enhancing data on flows.
2. the method according to claim 1, wherein obtaining the data set of live network flow in target scene
Step specifically includes:
Obtain the live network data on flows of target scene presetting granularity in default historical time section;
The default historical time section is divided into multiple sub- periods, by each sub- period each presetting granularity it is true
Network flow data is as a sample;
Data set by the set of all samples as live network flow in the target scene.
3. the method according to claim 1, wherein according to the data set, to generate confrontation network model into
Before the step of row training further include:
The data set is pre-processed, the pretreatment includes minimax normalized.
4. according to the method described in claim 2, it is characterized in that, according to the data set, to generate confrontation network model into
The step of row training specifically includes:
Enhancing processing is carried out to random noise based on the generator generated in confrontation network model, acquisition is intermediate to enhance flow
Data;
Based on the arbiter generated in confrontation network model to the number after the intermediate enhancing data on flows and pretreatment
Differentiated according to the sample of concentration;
The parameter of the generator and arbiter is adjusted according to differentiation result, until generation confrontation network model is received
It holds back.
5. according to the method described in claim 4, it is characterized in that, based on the generator pair generated in confrontation network model
The step of random noise carries out enhancing processing, obtains intermediate enhancing data on flows specifically includes:
Random noise is inputted to the input layer of the generator;
Enhancing processing is carried out to the random noise based on one or more hidden layers of the generator, obtains intermediate enhancing stream
Measure data;
The intermediate enhancing data on flows is input in the arbiter by the output layer based on the generator.
6. according to the method described in claim 5, it is characterized in that, the hidden layer be full articulamentum, the full articulamentum
Activation primitive is line rectification function.
7. according to the method described in claim 4, it is characterized in that, being obtained based on the trained generation confrontation network model
The step of taking final enhancing data on flows specifically includes:
The intermediate enhancing data traffic for generating generator output after confrontation network model is restrained is enhanced into flow as final
Data.
8. a kind of based on the network flow data enhancement device for generating confrontation network characterized by comprising
Module is obtained, for obtaining the data set of live network flow in target scene;
Training module, for being trained to confrontation network model is generated according to the data set;
Output module obtains final enhancing data on flows for fighting network model based on the trained generation.
9. a kind of electronic equipment characterized by comprising
At least one processor, at least one processor and bus;Wherein,
The processor and memory complete mutual communication by the bus;
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to instruct energy
Enough methods executed as described in claim 1 to 7 is any.
10. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Computer instruction is stored up, the computer instruction makes the computer execute the method as described in claim 1 to 7 is any.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811496710.2A CN109639479B (en) | 2018-12-07 | 2018-12-07 | Network traffic data enhancement method and device based on generation countermeasure network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811496710.2A CN109639479B (en) | 2018-12-07 | 2018-12-07 | Network traffic data enhancement method and device based on generation countermeasure network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109639479A true CN109639479A (en) | 2019-04-16 |
CN109639479B CN109639479B (en) | 2020-10-20 |
Family
ID=66072037
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811496710.2A Active CN109639479B (en) | 2018-12-07 | 2018-12-07 | Network traffic data enhancement method and device based on generation countermeasure network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109639479B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110602078A (en) * | 2019-09-04 | 2019-12-20 | 南京邮电大学 | Application encryption traffic generation method and system based on generation countermeasure network |
CN112398674A (en) * | 2019-08-16 | 2021-02-23 | 中国移动通信集团浙江有限公司 | Method and device for generating VNFD configuration template for describing virtual network functions |
CN112688928A (en) * | 2020-12-18 | 2021-04-20 | 中国科学院信息工程研究所 | Network attack flow data enhancement method and system combining self-encoder and WGAN |
CN112884075A (en) * | 2021-03-23 | 2021-06-01 | 北京天融信网络安全技术有限公司 | Traffic data enhancement method, traffic data classification method and related device |
CN112884121A (en) * | 2021-02-05 | 2021-06-01 | 武汉纺织大学 | Traffic identification method based on generation of confrontation deep convolutional network |
CN113158390A (en) * | 2021-04-29 | 2021-07-23 | 北京邮电大学 | Network attack traffic generation method for generating countermeasure network based on auxiliary classification |
WO2021169366A1 (en) * | 2020-02-25 | 2021-09-02 | 华为技术有限公司 | Data enhancement method and apparatus |
CN113726545A (en) * | 2021-06-23 | 2021-11-30 | 清华大学 | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement |
CN116737793A (en) * | 2023-05-29 | 2023-09-12 | 南方电网能源发展研究院有限责任公司 | Carbon emission stream generation method, model training method, device and computer equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103023927A (en) * | 2013-01-10 | 2013-04-03 | 西南大学 | Method and system for intrusion detection based on non-negative matrix factorization under sparse representation |
CN104796416A (en) * | 2015-04-08 | 2015-07-22 | 中国科学院信息工程研究所 | Botnet simulation method and botnet simulation system |
CN107451619A (en) * | 2017-08-11 | 2017-12-08 | 深圳市唯特视科技有限公司 | A kind of small target detecting method that confrontation network is generated based on perception |
CN108322349A (en) * | 2018-02-11 | 2018-07-24 | 浙江工业大学 | The deep learning antagonism attack defense method of network is generated based on confrontation type |
CN108446667A (en) * | 2018-04-04 | 2018-08-24 | 北京航空航天大学 | Based on the facial expression recognizing method and device for generating confrontation network data enhancing |
CN108805418A (en) * | 2018-05-22 | 2018-11-13 | 福州大学 | A kind of traffic data fill method fighting network based on production |
-
2018
- 2018-12-07 CN CN201811496710.2A patent/CN109639479B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103023927A (en) * | 2013-01-10 | 2013-04-03 | 西南大学 | Method and system for intrusion detection based on non-negative matrix factorization under sparse representation |
CN104796416A (en) * | 2015-04-08 | 2015-07-22 | 中国科学院信息工程研究所 | Botnet simulation method and botnet simulation system |
CN107451619A (en) * | 2017-08-11 | 2017-12-08 | 深圳市唯特视科技有限公司 | A kind of small target detecting method that confrontation network is generated based on perception |
CN108322349A (en) * | 2018-02-11 | 2018-07-24 | 浙江工业大学 | The deep learning antagonism attack defense method of network is generated based on confrontation type |
CN108446667A (en) * | 2018-04-04 | 2018-08-24 | 北京航空航天大学 | Based on the facial expression recognizing method and device for generating confrontation network data enhancing |
CN108805418A (en) * | 2018-05-22 | 2018-11-13 | 福州大学 | A kind of traffic data fill method fighting network based on production |
Non-Patent Citations (3)
Title |
---|
夏正敏: "基于分形的网络流量分析及异常检测技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
柯健等: "基于生成对抗网络的语音增强算法研究", 《信息技术与网络安全》 * |
陈文兵等: "基于条件生成式对抗网络的数据增强方法", 《计算机应用》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112398674A (en) * | 2019-08-16 | 2021-02-23 | 中国移动通信集团浙江有限公司 | Method and device for generating VNFD configuration template for describing virtual network functions |
CN112398674B (en) * | 2019-08-16 | 2022-08-23 | 中国移动通信集团浙江有限公司 | Method and device for generating VNFD configuration template for describing virtual network functions |
CN110602078A (en) * | 2019-09-04 | 2019-12-20 | 南京邮电大学 | Application encryption traffic generation method and system based on generation countermeasure network |
CN113379045B (en) * | 2020-02-25 | 2022-08-09 | 华为技术有限公司 | Data enhancement method and device |
WO2021169366A1 (en) * | 2020-02-25 | 2021-09-02 | 华为技术有限公司 | Data enhancement method and apparatus |
CN113379045A (en) * | 2020-02-25 | 2021-09-10 | 华为技术有限公司 | Data enhancement method and device |
CN112688928A (en) * | 2020-12-18 | 2021-04-20 | 中国科学院信息工程研究所 | Network attack flow data enhancement method and system combining self-encoder and WGAN |
CN112884121A (en) * | 2021-02-05 | 2021-06-01 | 武汉纺织大学 | Traffic identification method based on generation of confrontation deep convolutional network |
CN112884075A (en) * | 2021-03-23 | 2021-06-01 | 北京天融信网络安全技术有限公司 | Traffic data enhancement method, traffic data classification method and related device |
CN113158390A (en) * | 2021-04-29 | 2021-07-23 | 北京邮电大学 | Network attack traffic generation method for generating countermeasure network based on auxiliary classification |
CN113726545A (en) * | 2021-06-23 | 2021-11-30 | 清华大学 | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement |
CN113726545B (en) * | 2021-06-23 | 2022-12-23 | 清华大学 | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement |
CN116737793A (en) * | 2023-05-29 | 2023-09-12 | 南方电网能源发展研究院有限责任公司 | Carbon emission stream generation method, model training method, device and computer equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109639479B (en) | 2020-10-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109639479A (en) | Based on the network flow data Enhancement Method and device for generating confrontation network | |
Hu et al. | 3D separable convolutional neural network for dynamic hand gesture recognition | |
CN106982359B (en) | A kind of binocular video monitoring method, system and computer readable storage medium | |
CN109558966B (en) | Processing system for intelligently judging evidence and predicting occurrence of event | |
CN110033279A (en) | The suspicious account trading confirmation method and device of knowledge based graphical spectrum technology | |
CN107423442A (en) | Method and system, storage medium and computer equipment are recommended in application based on user's portrait behavioural analysis | |
CN110298663A (en) | Based on the wide fraudulent trading detection method learnt deeply of sequence | |
CN110084603A (en) | Method, detection method and the corresponding intrument of training fraudulent trading detection model | |
CN108416663A (en) | The method and device of the financial default risk of assessment | |
CN109376634A (en) | A kind of Bus driver unlawful practice detection system neural network based | |
CN110113634A (en) | A kind of information interaction method, device, equipment and storage medium | |
CN112700324A (en) | User loan default prediction method based on combination of Catboost and restricted Boltzmann machine | |
CN110084468A (en) | A kind of Risk Identification Method and device | |
CN110020099A (en) | A kind of the user's recommended method and device of video friend-making | |
CN109670927A (en) | The method of adjustment and its device of credit line, equipment, storage medium | |
Wang et al. | A hybrid bandit model with visual priors for creative ranking in display advertising | |
CN110647905B (en) | Method for identifying terrorist-related scene based on pseudo brain network model | |
CN115660147A (en) | Information propagation prediction method and system based on influence modeling between propagation paths and in propagation paths | |
CN110147851A (en) | Method for screening images, device, computer equipment and storage medium | |
CN103281555B (en) | Half reference assessment-based quality of experience (QoE) objective assessment method for video streaming service | |
CN109829593A (en) | The credit rating of target object determines method, apparatus, storage medium and electronic device | |
CN107909288A (en) | Paying behaviors analysis method based on SOM neural network clustering algorithms | |
CN112989955B (en) | Human body action recognition method based on space-time double-flow heterogeneous grafting convolutional neural network | |
CN110493026A (en) | Cordless communication network incoming call complaint handling method and device | |
CN106128101B (en) | The visualization device of magnitude of traffic flow short-term prediction result |
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 |