CN115941291A - Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network - Google Patents
Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network Download PDFInfo
- Publication number
- CN115941291A CN115941291A CN202211435668.XA CN202211435668A CN115941291A CN 115941291 A CN115941291 A CN 115941291A CN 202211435668 A CN202211435668 A CN 202211435668A CN 115941291 A CN115941291 A CN 115941291A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- data
- network
- value
- wgan
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000004458 analytical method Methods 0.000 title claims abstract description 24
- 238000011156 evaluation Methods 0.000 claims abstract description 146
- 238000013136 deep learning model Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000012800 visualization Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 31
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000009826 distribution Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 13
- 238000011002 quantification Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 4
- 230000002194 synthesizing effect Effects 0.000 claims description 4
- 230000003042 antagnostic effect Effects 0.000 abstract description 5
- 230000008447 perception Effects 0.000 abstract description 4
- 230000007246 mechanism Effects 0.000 description 8
- 230000009286 beneficial effect Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241000288105 Grus Species 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an analysis system and method for DPoS block chain network security situation perception, and belongs to the technical field of block chains. The method comprises the steps of obtaining relevant characteristic data; quantizing the blockchain network value; training to generate an improved variant WGAN-GP of the antagonistic network to obtain a WGAN-GP deep learning model; performing time series prediction by using the WGAN-GP deep learning model, and filling missing values in the relevant characteristic data to obtain complete data; according to the complete data, quantifying an evaluation object by utilizing an improved TOPSIS evaluation algorithm with a weight; and displaying the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data by using a visualization system. The invention solves the problems of unstable whole network caused by inconvenient perception of the security situation of the block chain network and incapability of timely clearing hidden dangers.
Description
Technical Field
The invention belongs to the technical field of block chains, and particularly relates to an analysis system and method for DPoS block chain network security situation perception.
Background
As a new technology framework, the block chain technology integrates technologies such as peer-to-peer network, asymmetric encryption, merkel tree, consensus mechanism, and intelligent contract, and in recent years, as the application of the block chain technology gradually relates to multiple traditional and new industries such as finance, energy, internet of things, government affairs management, health care, and transportation, the block chain technology has become a focus of attention of researchers. The DPoS consensus mechanism is considered to be suitable for large-scale application scenarios due to high transaction throughput, low block-out delay and low energy consumption, attracts a large number of decentralized application developers and users to participate, and meanwhile, the user's requirements for safety and stable monitoring of a block link network are increasing. In the DPoS common identification mechanism, a special producer node is responsible for verification and packaging transaction, the number of the producer nodes in the network is fixed, all the money-holding users in the network exchange votes for voting to determine, all the nodes meeting the corresponding standards can participate in voting, the nodes with higher ranking of votes can obtain the right of a packaging block, and the rest nodes become standby nodes and only synchronize data on a chain. However, a DPoS-based block-chain network may have the following risks:
(1) The block producer node and the user holding the vote higher jointly operate the election activity, and the benefit of the common user is damaged.
(2) The elected nodes are not necessarily honest nodes, when wrong nodes or malicious nodes occur, the nodes can only be expected to be removed in the next round of election, and the malicious nodes cannot be timely removed due to the fact that voting is not positive.
(3) At present, no mechanism for evaluating nodes exists, and unqualified producer nodes are easy to select to influence the stability of the whole network.
(4) The geographical position distribution of the nodes is too centralized, and the integrity of the block chain data is threatened due to the inevitable centralization tendency.
For the building contributors and related researchers of the blockchain network, security situational awareness for the blockchain network is an effective way to check the above risks. The data visualization tool can effectively support the decision of professionals, the transmitted information amount is increased by using the multiple visual channels, meanwhile, the burden of information reduction is reduced, and convenience is brought to decision work.
With the widespread application of blockchain technology, in recent years, a variety of blockchain visualization tools have been developed by many blockchain technology enthusiasts and related researchers. One tool in the prior art is a series of visual views designed in a visual metaphor manner for a common user, so as to help the user to know an operation mechanism of a blockchain network and attract potential users to participate in ecological construction of the blockchain network.
Generally speaking, in the aspect of account addresses, transaction data and the like, data are only statistically displayed aiming at the main focus of block chain visualization research at present, deep mining and exploration are lacked, the overall security situation and evolution of a block chain network are less focused, and the block chain network visualization research on a DPoS consensus mechanism appearing later is relatively less.
Disclosure of Invention
Aiming at the defects in the prior art, the analysis system and method for DPoS block chain network security situation awareness provided by the invention solve the problems that the block chain network security situation awareness is inconvenient and the whole network is unstable due to the fact that hidden dangers cannot be timely eliminated.
In order to achieve the purpose of the invention, the invention adopts the technical scheme that: a DPoS block chain network security situation awareness oriented analysis method comprises the following steps:
s1, acquiring related characteristic data;
s2, quantifying the network value of the block chain according to the community token value of the related characteristic data;
s3, training and generating an improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
s4, performing time series prediction by using the WGAN-GP deep learning model, and filling missing values in the relevant characteristic data to obtain complete data;
s5, according to the complete data, quantifying an evaluation object by using an improved TOPSIS evaluation algorithm with a weight;
and S6, displaying the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data by using a visualization system.
The invention has the beneficial effects that: the method for quantifying the overall network security situation can effectively find abnormal nodes appearing in the network in the evolution process in time, explore the reasons of the appearance of the abnormalities through multi-view linkage of an interactive system, and present information in a visual view rather than a table form, so that common users can easily use the related information of the block chain network, researchers can easily acquire the analysis and mining results of data, defects appearing in the block chain network can be detected in time, support is provided for improvement work, and the block chain network managers are helped to verify whether an additional mechanism is effective. For time series prediction, the WGAN-GP deep learning model can generate samples more quickly because different data do not need to be generated in a sampling sequence, only backward propagation is utilized, a Markov decision chain is not needed, and the WGAN-GP deep learning model can still make prediction with higher accuracy when a sudden situation is faced.
Further, the method for acquiring the relevant feature data in step S1 includes the following steps:
s101, carrying out replay transaction in a block chain network to obtain original data of transaction information, and extracting characteristic data of a producer node from the original data;
s102, acquiring related network characteristic information in the Internet, and extracting network characteristic data from the network characteristic information;
s103, asynchronously processing the characteristic data of the producer node and the characteristic data of the network, and obtaining related characteristic data after structuring.
The beneficial effects of the above further scheme are: the method comprises the steps of obtaining original data of transaction information from a block chain network, extracting relevant network characteristic information from the Internet, processing the original data and the network characteristic information to obtain relevant characteristic data, and preparing for quantifying the whole security situation of a producer node and the network subsequently.
Further, the WGAN-GP deep learning model in the step S3 includes a generator and a discriminator;
the generator is used for synthesizing a virtual sample;
the discriminator is used for distinguishing real samples from virtual samples.
The beneficial effects of the above further scheme are: synthesizing a virtual sample similar to the real sample by using a generator, identifying the virtual sample by using an identifier, and continuously training to enable the virtual sample synthesized by the generator to be as close to the real sample as possible, so that the identifier can not identify the virtual sample, and the virtual sample synthesized by the generator can directly replace the missing related characteristic data to prepare for analyzing and processing the related characteristic data.
Further, the generator comprises a first gated loop unit GRU, a second gated loop unit GRU, a third gated loop unit GRU, a first fully-connected layer and a second fully-connected layer; the number of neurons of the first gating circulation unit GRU, the second gating circulation unit GRU and the third gating circulation unit GRU is 1024, 512 and 256 respectively;
the discriminator comprises a first convolution layer, a second convolution layer, a third full-connection layer, a fourth full-connection layer and a fifth full-connection layer; the number of neurons in the first convolutional layer, the second convolutional layer and the third convolutional layer is 32, 64 and 128 respectively; the number of neurons in the third full junction layer, the fourth full junction layer and the fifth full junction layer is 220, 220 and 1 respectively.
The beneficial effects of the above further scheme are: the generator using the three-layer gating circulation unit GRU has better performance, and the generator and the discriminator mutually confront each other to train a WGAN-GP deep learning model with good performance so as to prepare for the subsequent time series prediction.
Further, the objective function of the WGAN-GP deep learning model is:
ε～U[0,1]
wherein, L is the target function of the WGAN-GP deep learning model, lambda is the punished gradient coefficient, and P g For the false sample distribution generated by the generator, P r In order to be a true distribution of the sample,
for a randomly interpolated sample distribution, <' >>
For the dummy sample samples generated by the generator, x is a true sample, and->
Is x and>
a randomly interpolated sample on the connecting line, ε being a constant, ->
For the expectation of a false sample, ->
For the evaluation of a false sample by the discriminator, a decision is made as to whether the result is positive or negative>
D (x) is the result of the evaluation of a real sample by the discriminator for the expectation of a real sample, and->
Sample expectation for random interpolation>
And identifying the norm of the result of the random interpolation sampling for the identifier.
The beneficial effects of the above further scheme are: the improved variant WGAN-GP for generating the antagonistic network well solves the phenomenon that the gradient is 0 in the training process, and in addition, the improved variant WGAN-GP for generating the antagonistic network also adds a gradient penalty, so that the training stability of the improved variant WGAN-GP for generating the antagonistic network is greatly enhanced. For time series prediction, the WGAN-GP deep learning model can generate samples more quickly because different data do not need to be generated in a sampling sequence, only backward propagation is utilized, a Markov decision chain is not needed, and the WGAN-GP deep learning model can still make prediction with higher accuracy when a sudden situation is faced.
Further, the method for quantifying the evaluation object in step S5 includes the steps of:
s501, obtaining a feature matrix of the evaluation object according to the complete data:
wherein X is a feature matrix of an evaluation object, X nm The value of the mth evaluation index of the nth evaluation object is shown, n is the total amount of the evaluation objects, and m is the total amount of the evaluation indexes;
s502, carrying out positive processing on negative evaluation indexes in the feature matrix to obtain positive data:
i∈[1,n]
j∈[1,m]
wherein max {. Is a maximum function, min {. Is a minimum function, x ij A value of a j-th evaluation index which is an i-th evaluation object, wherein i =1,2, …, n, j =1,2, … m;
s503, carrying out standardization processing on the forward data to obtain standardized data:
wherein z is ij Normalized data of the jth evaluation index of the ith evaluation object;
s504, obtaining a standardized feature matrix according to the standardized data;
s505, calculating the information entropy of each evaluation index according to the standardized feature matrix:
wherein E is j Information entropy of the jth evaluation index;
s506, calculating the weight of each evaluation index according to the information entropy:
wherein, w j The weight value of the jth evaluation index is obtained;
s507, defining the maximum value and the minimum value in the evaluation object according to the standardized data matrix:
Z + ＝(max{z 11 ,z 21 ,…,z n1 },max{z 12 ,z 22 ,…,z n2 },…,max{z 1m ,z 2m ,…,z nm })
Z - ＝(min{z 11 ,z 21 ,…,z n1 },min{z 12 ,z 22 ,…,z n2 },…,min{z 1m ,z 2m ,…,z nm })
wherein, Z + Maximum value for evaluation, Z - Minimum value for evaluation object, z nm A normalized data matrix of the mth evaluation index as the nth evaluation object;
s508, according to the maximum value and the minimum value in the evaluation objects and the weight of each evaluation index, calculating the distance between each evaluation object and the maximum value and the minimum value:
wherein,
is the distance between the i-th evaluation object and the maximum value>
Is the distance between the i-th evaluation object and the minimum value>
Is the maximum value of the jth evaluation index>
Is the minimum value of the jth evaluation index;
s509, calculating the score of each producer node according to the distance between each evaluation object and the maximum value and the minimum value:
wherein S is i The score of the ith evaluation object;
s5010, obtaining the evaluation ranking of the evaluation objects according to the scores of the evaluation objects;
s5011, judging whether the evaluation object is a producer node, if so, entering the step S5011, and otherwise, finishing the quantification of the evaluation object;
s5012, comparing the difference between the evaluation ranking and the ticket number ranking of each producer node to obtain the stability of the producer node, and completing the quantification of the stability of the producer node:
T i ＝Rank s -Rank v
wherein, T i For stability of the ith producer node, rank s Ranking the evaluation of the producer node, rank v Ranking the votes for the producer node.
The beneficial effects of the above further scheme are: the smaller the information entropy value of the evaluation index is, the greater the dispersion degree of the evaluation index is, the greater the influence (namely weight) of the evaluation index on the comprehensive evaluation is, so that the weight of each evaluation index is calculated by using the information entropy tool, and the influence caused by subjective factors in the evaluation process is avoided; and calculating the score of the evaluation object according to the distance between the evaluation object and the maximum value, completing the quantification of the evaluation object, and if the evaluation object is a producer node, comparing the evaluation ranking with the ticketing ranking to quantify the stable situation of the producer node, thereby avoiding the subjectivity of data, not needing an objective function, not passing the inspection, and well depicting the comprehensive influence degree of a plurality of evaluation indexes.
The invention provides an analysis system facing DPoS block chain network security situation perception, which comprises:
the first processing module is used for acquiring related characteristic data;
a second processing module for quantifying a blockchain network value according to a community token value of the relevant feature data;
the third processing module is used for training and generating an improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
the fourth processing module is used for carrying out time series prediction by utilizing the WGAN-GP deep learning model and filling missing values in the related characteristic data to obtain complete data;
the fifth processing module is used for quantifying the stability of the producer nodes and the overall network security situation by utilizing an improved TOPSIS (technique for order preference by similarity to Ideal solution) evaluation algorithm with weights according to the complete data;
and the sixth processing module is used for presenting the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data and realizing multi-view linkage.
The invention has the beneficial effects that: the modules interact to form a complete data transmission chain, the effective information is finally displayed, the reason of abnormal occurrence is explored through multi-view linkage, information is presented in a visual view rather than a table form, so that not only is it easier for a common user to know relevant information of a block chain network, but also researchers can obtain analysis mining results of data more easily, defects occurring in the block chain network are detected in time, support is provided for improvement work, and the block chain network manager is helped to verify whether an additional mechanism is effective.
Drawings
Fig. 1 is a flowchart of an analysis method for DPoS block chain network security situation awareness in the present invention.
Fig. 2 is a flowchart of a method for acquiring related feature data according to the present invention.
Fig. 3 is a data storage structure diagram of related feature data in the present invention.
Fig. 4 is a block diagram of the improved variant WGAN-GP generating an antagonistic network in the present invention.
Fig. 5 is a flowchart of a method for quantifying an evaluation object according to the present invention.
Fig. 6 is a diagram of an analysis system structure oriented to DPoS block chain network security situation awareness in the present invention.
Fig. 7 is a system architecture diagram of an analysis system for DPoS blockchain network security situation awareness in the present invention.
Fig. 8 is a system overview view of an analysis system oriented to security situation awareness of a DPoS blockchain network according to the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined by the appended claims, and all changes that can be made by the invention using the inventive concept are intended to be protected.
Example 1
As shown in fig. 1, in an embodiment of the present invention, the present invention provides a DPoS blockchain network security situation awareness oriented analysis method, including the following steps:
s1, acquiring related characteristic data;
as shown in fig. 2, the method for acquiring the relevant feature data in step S1 includes the following steps:
s101, carrying out replay transaction in a block chain network to obtain original data of transaction information, and extracting characteristic data of a producer node from the original data;
s102, obtaining relevant network characteristic information in the Internet, and extracting network characteristic data from the network characteristic information
S103, asynchronously processing the characteristic data of the producer node and the characteristic data of the network, and obtaining related characteristic data after structuring.
In this embodiment, a replay transaction is performed in a blockchain network to obtain original data of transaction information and obtain related network characteristic information in the internet and store the original data into a memory buffer, and the original data is asynchronously processed in the memory buffer to eliminate redundant information and is stored into a storage device after being serialized.
In this embodiment, relevant feature data required by the present invention is extracted from the acquired data, and a condition of a node is reflected from three aspects of node feature, performance, and benefit is selected for a feature item of a producer node, where the node feature includes three features of a node name, a node vote count ranking, and a node geographical location, the performance aspect includes four features of block production quantity, packed transaction quantity, CPU consumption, and network resource consumption, and the benefit aspect includes two features of a node account balance and a node daily profit condition; the network characteristic data selects the number of newly-increased users per day, contract calling times, external information reports, google trends and token unit prices; the required feature data is structured to obtain relevant feature data, the relevant feature data is stored in a MySQL database for subsequent query and use, and the data storage structure is shown in FIG. 3.
S2, quantifying the network value of the block chain according to the community token value of the related characteristic data;
in this embodiment, the community token value may represent the interest level of the participating users in the blockchain network, so the community token value is used to quantify the blockchain network value.
S3, training and generating an improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
the WGAN-GP deep learning model in the step S3 comprises a generator and a discriminator;
the generator is used for synthesizing a virtual sample;
the discriminator is used for distinguishing real samples from virtual samples.
The generator comprises a first gating circulating unit GRU, a second gating circulating unit GRU, a third gating circulating unit GRU, a first full connection layer and a second full connection layer; the number of neurons of the first gating circulation unit GRU, the second gating circulation unit GRU and the third gating circulation unit GRU is 1024, 512 and 256 respectively;
the discriminator comprises a first convolution layer, a second convolution layer, a third full-connection layer, a fourth full-connection layer and a fifth full-connection layer; the number of neurons in the first convolutional layer, the second convolutional layer and the third convolutional layer is 32, 64 and 128 respectively; the number of neurons in the third full junction layer, the fourth full junction layer and the fifth full junction layer is 220, 220 and 1 respectively.
The target function of the WGAN-GP deep learning model is as follows:
ε～U[0,1]
wherein, L is the target function of the WGAN-GP deep learning model, lambda is the punished gradient coefficient, and P g For the false sample distribution generated by the generator, P r In order to be a true distribution of the sample,
sample distributions for random interpolation>
For the dummy sample samples generated by the generator, x is a true sample, and->
Is x and->
A randomly interpolated sample on the connecting line, ε being a constant, ->
For the expectation of a false sample, ->
For the evaluation of a false sample by the discriminator, a decision is made as to whether the result is positive or negative>
D (x) is the result of the evaluation of a real sample by the discriminator for the expectation of a real sample, and->
Sample expectation for random interpolation>
The norm of the result of the discriminator on the randomly interpolated sample is identified.
In this embodiment, the model is trained by obtaining historical data of the network value of the block chain, in order to ensure the training effect of the model, the selection of the model and the adjustment of the hyper-parameters are completed in the experimental stage, the improved variant WGAN-GP generating the countermeasure network is finally selected, the framework of the improved variant WGAN-GP generating the countermeasure network selected by the experiment is shown in fig. 4, and the network value of the block chain, that is, the final predicted output value, is represented by the value of the community token. A Gated Round Unit (GRU) was selected as a generator, and 7 characteristic variables were used to predict future value of tokens, which were summed up for daily opening price, closing price, maximum price, minimum price, EOSIO google trend, network one-day new user volume and one-day contract calls, respectively, and the historical data from the previous 30 days was used herein to predict the value of the next day. In order to construct a generator with good performance, the model uses three layers of GRUs, the number of neurons is 1024, 512 and 256 respectively, two full connection layers (sense) are added, and the number of neurons in the last layer is the same as the number of variables to be output. The discriminator in this model uses a convolutional neural network to distinguish whether the input data is real or imaginary, the input data to the discriminator being from the original real data and the generated data from the generator. The discriminator contains 3 convolutional layers (Conv 1 d), with neuron numbers of 32, 64 and 128, respectively, the activation function using LeakyReLU, and finally three fully connected layers (Dense) added, with neuron numbers of 220, 220 and 1, respectively.
S4, performing time series prediction by using the WGAN-GP deep learning model, and filling missing values in the relevant characteristic data to obtain complete data;
in this embodiment, the WGAN-GP deep learning model is selected to fill the missing values in the original data, and the main purpose of this step is to obtain a complete trend of network value evolution in a selected time period, and ensure that a normal analysis process is not interrupted and a view presentation is not abnormal due to missing data items in subsequent analysis method establishment and visualization system presentation.
S5, according to the complete data, quantifying an evaluation object by using an improved TOPSIS evaluation algorithm with a weight;
as shown in fig. 5, the method for quantifying the evaluation object in step S5 includes the following steps:
s501, obtaining a feature matrix of an evaluation object according to the complete data:
wherein X is a characteristic matrix of an evaluation object, X nm The value of the mth evaluation index of the nth evaluation object is obtained, n is the total quantity of the evaluation objects, and m is the total quantity of the evaluation indexes;
s502, carrying out positive processing on negative evaluation indexes in the feature matrix to obtain positive data:
i∈[1,n]
j∈[1,m]
wherein max {. Is a maximum function, min {. Is a minimum function, x ij A value of a j-th evaluation index which is an i-th evaluation object, wherein i =1,2, …, n, j =1,2, … m;
s503, carrying out standardization processing on the forward data to obtain standardized data:
wherein z is ij Normalized data of the jth evaluation index of the ith evaluation object;
s504, obtaining a standardized feature matrix according to the standardized data;
s505, calculating the information entropy of each evaluation index according to the standardized feature matrix:
wherein, E j Information entropy of the jth evaluation index;
s506, calculating the weight of each evaluation index according to the information entropy:
wherein, w j The weight value of the jth evaluation index is obtained;
s507, defining the maximum value and the minimum value in the evaluation object according to the standardized data matrix:
Z + ＝(max{z 11 ,z 21 ,…,z n1 },max{z 12 ,z 22 ,…,z n2 },…,max{z 1m ,z 2m ,…,z nm })
Z - ＝(min{z 11 ,z 21 ,…,z n1 },min{z 12 ,z 22 ,…,z n2 },…,min{z 1m ,z 2m ,…,z nm })
wherein Z is + Maximum value for evaluation, Z - Minimum value for evaluation object, z nm A normalized data matrix of the mth evaluation index as the nth evaluation object;
s508, according to the maximum value and the minimum value in the evaluation objects and the weight of each evaluation index, calculating the distance between each evaluation object and the maximum value and the minimum value:
wherein,
the distance between the ith evaluation object and the maximum value,/>
is the distance between the i-th evaluation object and the minimum value>
For the maximum of the jth evaluation index>
Is the minimum value of the jth evaluation index;
s509, calculating the score of each producer node according to the distance between each evaluation object and the maximum value and the minimum value:
wherein S is i The score of the ith evaluation object;
s5010, obtaining the evaluation ranking of the evaluation objects according to the scores of the evaluation objects;
s5011, judging whether the evaluation object is a producer node, if so, entering the step S5011, and otherwise, finishing the quantification of the evaluation object;
s5012, comparing the difference between the evaluation ranking and the ticket number ranking of each producer node to obtain the stability of the producer node, and completing quantification of the stability of the producer node:
T i ＝Rank s -Rank v
wherein, T i For stability of the ith producer node, rank s Ranking the evaluation of producer nodes, rank v Ranking the votes for the producer node.
In the embodiment, in order to avoid the influence of subjective factors on evaluation, an entropy weight method is used for calculating the feature weight of each evaluation object, the evaluation objects comprise all producer nodes in the same period and the overall network security situation in different periods, the feature items of the producer nodes comprise node names, node ticket number ranking, node balance, node income situation, block production quantity, packed transaction quantity, CPU consumption and network resource consumption, and the feature items of the overall network security situation comprise single-node stability situation, network value after filling processing and network decentralization degree. Ordering all producer nodes by using an improved TOPSIS evaluation algorithm with weights, ordering the producer nodes according to the number of votes obtained by each node in original data, quantifying the stability of the producer nodes by analyzing the difference between the evaluation ordering and the number of votes obtained by the producer nodes, calculating the weights of all characteristics by using an entropy weight method in the same way for a network overall security situation quantification method, and then ordering the network overall security situation in a selected time sequence by using the TOPSIS evaluation algorithm; the entropy weight method is only considered from a data level and is an objective weighting method, and the principle is as follows: the smaller the variation degree of the index is, the less the amount of information is reflected, and the lower the corresponding weight value should be.
And S6, displaying the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data by using a visualization system.
In this embodiment, the producer node stability and the overall network security situation obtained through the processing flows of steps S1 to S5 are combined with effective information extracted from the complete data, where the effective information includes node geographical location distribution, voting aggressiveness, vote source information, network value evolution, and external information report data, and a visualization system is used to present the result.
Example 2
As shown in fig. 6, the present invention provides an analysis system facing DPoS blockchain network security situation awareness, including:
the first processing module is used for acquiring related characteristic data;
a second processing module for quantifying a blockchain network value according to a community token value of the relevant feature data;
the third processing module is used for training and generating improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
the fourth processing module is used for carrying out time series prediction by utilizing the WGAN-GP deep learning model and filling missing values in the relevant characteristic data to obtain complete data;
the fifth processing module is used for quantifying the stability of the nodes of the producer and the overall security situation of the network by utilizing an improved TOPSIS (technique for order preference by similarity to Ideal solution) evaluation algorithm with weight values according to the complete data;
and the sixth processing module is used for presenting the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data and realizing multi-view linkage.
In this embodiment, the system architecture diagram is shown in fig. 7, and the overview view of the system is shown in fig. 8, and the system is composed of a node geographical distribution evolution view (fig. 8 (a)), a network value evolution view (fig. 8 (B)), a network stability evolution view (fig. 8 (C)), a node stability evolution view (fig. 8 (D)), a single node stability evolution view (fig. 8 (E1)), a newly added vote source view (fig. 8 (E2)), and a control panel (fig. 8 (F)). The time period to be observed and the version number of the producer node list are selected through a control panel (figure 8 (F)), a node geographical distribution evolution view (figure 8 (A)), a network value evolution view (figure 8 (B)), and a network stability evolution view (figure 8 (C)) can present the overall situation of the network, a newly added vote source of the version can be observed by selecting a specific version number, and an interested node can jump to a single-node stability evolution view (figure 8 (E1)) to observe by selecting the node stability evolution view (figure 8 (D)). The safety situation of the whole block chain network is observed from the distribution evolution of the geographic position, the change of the external environment, the node stability and the network stability, and the reason for the abnormal phenomenon can be further explored. The design idea of each view is described as follows:
the node geographical distribution evolution view (fig. 8 (a)) is used to present the geographical location distribution evolution of the producer node, and the present embodiment adopts a view similar to a strip chart and filled with bars to present the geographical location distribution of the nodes in the network. The time is taken as the abscissa of the view, different areas are coded by the occupied range of the ordinate, the occupied proportion on the ordinate represents the proportion of the number of all nodes in the area in the whole network at the time, the nodes are arranged according to the total vote ranking obtained by all the nodes in each area from top to bottom, and the density degree of the bar distribution in the view is used for coding the frequency degree of the version alternation of the producer node list.
The network value evolution view (figure 8 (B)) is used for showing the trend of token prices and contemporaneous external information, the view shares an abscissa with the node geographical distribution evolution view (figure 8 (a)), and the background presents the height of the block chain network value. And arranging the collected related information data in a view in a scatter point mode according to the release time, wherein the area size of the scatter point represents the heat of the information, and clicking the scatter point can jump to a specific information report page.
The network stability evolution view (fig. 8 (C)) is used for presenting the stability evolution situation of the whole blockchain network, and presents the relative change situation of the blockchain network stability within the range of the selected version in a line graph mode by taking the version changed by the producer node list as the abscissa.
The node stability evolution view (figure 8 (D)) is used for presenting the stability evolution situation of the producer node in each version, the abscissa of the view adopts the version number of the producer node list, the ordinate is arranged from top to bottom according to the vote number ranking obtained by the producer node, the circular node is used for representing the stability situation of the current producer node, the same node is connected by a straight line in different versions, and if the current producer node is selected in the next round of election, the straight line directly extends to the abscissa; if a candidate node is selected successfully in the previous round, a straight line directly extends from the abscissa to the node, and the two types of nodes are more interesting.
The single-node stability evolution view (fig. 8 (E1)) is used for presenting the stability evolution situation of a single producer node in a selected version range, the view is further designed on the basis of a broken line graph, the abscissa is a version number, and the ordinate represents the difference between two ranks.
The new vote source view (fig. 8 (E2)) is used to show the new vote source of the single selected producer node in the current version during the previous round of election. The view is presented by a network diagram, wherein the nodes comprise two types of nodes which are respectively a producer node and a voter, the voter is divided into two types which comprise a common User (User) and an agent (Proxy), and the thickness of a connecting line between the nodes is used for encoding the voting number.
Claims (7)
1. A DPoS block chain network security situation awareness oriented analysis method is characterized by comprising the following steps:
s1, acquiring related characteristic data;
s2, quantifying the network value of the block chain according to the community token value of the related characteristic data;
s3, training and generating an improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
s4, performing time series prediction by using the WGAN-GP deep learning model, and filling missing values in the relevant characteristic data to obtain complete data;
s5, according to the complete data, quantifying an evaluation object by using an improved TOPSIS evaluation algorithm with a weight;
and S6, displaying the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data by using a visualization system.
2. The DPoS blockchain network security situation awareness-oriented analysis method according to claim 1, wherein the method for acquiring the relevant feature data in the step S1 includes the following steps:
s101, carrying out replay transaction in a block chain network to obtain original data of transaction information, and extracting characteristic data of a producer node from the original data;
s102, acquiring related network characteristic information in the Internet, and extracting network characteristic data from the network characteristic information;
s103, asynchronously processing the producer node characteristic data and the network characteristic data, and obtaining related characteristic data after structuring.
3. The DPoS blockchain network security posture awareness oriented analysis method according to claim 1, wherein the WGAN-GP deep learning model in the step S3 comprises a generator and a discriminator;
the generator is used for synthesizing a virtual sample;
the discriminator is used for distinguishing real samples from virtual samples.
4. The DPoS block chain network security situation awareness-oriented analysis method according to claim 3, wherein the generator comprises a first gated round-robin unit GRU, a second gated round-robin unit GRU, a third gated round-robin unit GRU, a first full-connectivity layer and a second full-connectivity layer; the number of neurons of the first gating circulation unit GRU, the second gating circulation unit GRU and the third gating circulation unit GRU is 1024, 512 and 256 respectively;
the discriminator comprises a first convolution layer, a second convolution layer, a third full-connection layer, a fourth full-connection layer and a fifth full-connection layer; the number of neurons in the first convolutional layer, the second convolutional layer and the third convolutional layer is 32, 64 and 128 respectively; the number of neurons in the third full junction layer, the fourth full junction layer and the fifth full junction layer is 220, 220 and 1 respectively.
5. The DPoS blockchain network security posture awareness oriented analysis method according to claim 4, wherein an objective function of the WGAN-GP deep learning model is as follows:
ε～U[0,1]
wherein, L is an objective function of the WGAN-GP deep learning model, lambda is a penalty gradient coefficient, and P is g For the false sample distribution generated by the generator, P r In order to be a true distribution of the sample,
for a randomly interpolated sample distribution, <' >>
For the dummy sample samples generated by the generator, x is a true sample, and->
Is x and->
A randomly interpolated sample on the connecting line, ε being a constant, ->
For the expectation of a false sample, ->
For the evaluation of a false sample by a discriminator>
In anticipation of a real sample, D (x) is the result of the evaluation of a real sample by the evaluator, and>
sample expectation for random interpolation>
For a discriminatorAnd identifying the norm of the result of the random interpolation sampling sample.
6. The DPoS blockchain network security situation awareness-oriented analysis method according to claim 1, wherein the quantification method of the evaluation object in the step S5 comprises the following steps:
s501, obtaining a feature matrix of the evaluation object according to the complete data:
wherein X is a feature matrix of an evaluation object, X nm The value of the mth evaluation index of the nth evaluation object is obtained, n is the total quantity of the evaluation objects, and m is the total quantity of the evaluation indexes;
s502, carrying out positive processing on the negative evaluation indexes in the feature matrix to obtain positive data:
wherein max {. Is a maximum function, min {. Is a minimum function, x ij A value of a j-th evaluation index which is an i-th evaluation object, wherein i =1,2, …, n, j =1,2, … m;
s503, carrying out standardization processing on the forward data to obtain standardized data:
wherein z is ij Normalized data of a jth evaluation index for an ith evaluation object;
s504, obtaining a standardized feature matrix according to the standardized data;
s505, calculating the information entropy of each evaluation index according to the standardized feature matrix:
wherein E is j Information entropy of the jth evaluation index;
s506, calculating the weight of each evaluation index according to the information entropy:
wherein w j The weight value of the jth evaluation index is obtained;
s507, defining the maximum value and the minimum value in the evaluation object according to the standardized data matrix:
Z + ＝(max{z 11 ,z 21 ,…,z n1 },max{z 12 ,z 22 ,…,z n2 },…,max{z 1m ,z 2m ,…,z nm })
Z - ＝(min{z 11 ,z 21 ,…,z n1 },min{z 12 ,z 22 ,…,z n2 },…,min{z 1m ,z 2m ,…,z nm })
wherein, Z + Maximum value for evaluation, Z - Minimum value for evaluation object, z nm A normalized data matrix of the mth evaluation index as the nth evaluation object;
s508, according to the maximum value and the minimum value in the evaluation objects and the weight of each evaluation index, calculating the distance between each evaluation object and the maximum value and the minimum value:
wherein,
is the distance between the i-th evaluation object and the maximum value>
Is the distance between the i-th evaluation object and the minimum value>
For the maximum of the jth evaluation index>
Is the minimum value of the jth evaluation index;
s509, calculating the score of each producer node according to the distance between each evaluation object and the maximum value and the minimum value:
wherein S is i Score for the ith evaluation object;
s5010, obtaining the evaluation ranking of the evaluation objects according to the scores of the evaluation objects;
s5011, judging whether the evaluation object is a producer node, if so, entering the step S5011, and otherwise, finishing the quantification of the evaluation object;
s5012, comparing the difference between the evaluation ranking and the ticket number ranking of each producer node to obtain the stability of the producer node, and completing quantification of the stability of the producer node:
T i ＝Rank s -Rank v
wherein, T i For stability of the ith producer node, rank s Ranking the evaluation of producer nodes, rank v And ranking the votes obtained by the producer nodes.
7. The DPoS blockchain network security posture awareness oriented analysis system of any one of claims 1 to 6, comprising:
the first processing module is used for acquiring related characteristic data;
a second processing module for quantifying a blockchain network value according to a community token value of the relevant feature data;
the third processing module is used for training and generating an improved variant WGAN-GP of the countermeasure network by taking the historical data of the block chain network value as a real sample to obtain a WGAN-GP deep learning model;
the fourth processing module is used for carrying out time series prediction by utilizing the WGAN-GP deep learning model and filling missing values in the relevant characteristic data to obtain complete data;
the fifth processing module is used for quantifying the stability of the producer nodes and the overall network security situation by utilizing an improved TOPSIS (technique for order preference by similarity to Ideal solution) evaluation algorithm with weights according to the complete data;
and the sixth processing module is used for presenting the stability of the producer nodes, the overall network security situation and the effective information in the related characteristic data and realizing multi-view linkage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211435668.XA CN115941291A (en) | 2022-11-16 | 2022-11-16 | Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211435668.XA CN115941291A (en) | 2022-11-16 | 2022-11-16 | Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115941291A true CN115941291A (en) | 2023-04-07 |
Family
ID=86556687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211435668.XA Pending CN115941291A (en) | 2022-11-16 | 2022-11-16 | Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115941291A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110309657A (en) * | 2019-06-20 | 2019-10-08 | 四川大学 | The safety risk estimating method of block chain |
| CN112070373A (en) * | 2020-08-25 | 2020-12-11 | 深圳信息职业技术学院 | Method and system for grading quality of block chain item |
| CN113114492A (en) * | 2021-04-01 | 2021-07-13 | 哈尔滨理工大学 | Security situation perception algorithm based on Markov differential game block chain model |
| US20210320847A1 (en) * | 2018-10-15 | 2021-10-14 | Netz Forecasts Ltd. | Systems and methods for network stabilization prediction |
| US20220021711A1 (en) * | 2020-07-20 | 2022-01-20 | Cgi Federal | Security Platform and Method for Efficient Access and Discovery |
| CN114168948A (en) * | 2021-12-17 | 2022-03-11 | 北京华清信安科技有限公司 | Network security situation comprehensive analysis method |
| WO2022080583A1 (en) * | 2020-10-14 | 2022-04-21 | 고려대학교 세종산학협력단 | Deep learning-based bitcoin block data prediction system taking into account time series distribution characteristics |
| CN114638021A (en) * | 2022-03-18 | 2022-06-17 | 北京邮电大学 | Internet of things lightweight block chain system security evaluation method |
| CN115174165A (en) * | 2022-06-21 | 2022-10-11 | 重庆移通学院 | Self-adaptive security situation cognitive system based on block chain under cloud edge-side architecture |
-
2022
- 2022-11-16 CN CN202211435668.XA patent/CN115941291A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210320847A1 (en) * | 2018-10-15 | 2021-10-14 | Netz Forecasts Ltd. | Systems and methods for network stabilization prediction |
| CN110309657A (en) * | 2019-06-20 | 2019-10-08 | 四川大学 | The safety risk estimating method of block chain |
| US20220021711A1 (en) * | 2020-07-20 | 2022-01-20 | Cgi Federal | Security Platform and Method for Efficient Access and Discovery |
| CN112070373A (en) * | 2020-08-25 | 2020-12-11 | 深圳信息职业技术学院 | Method and system for grading quality of block chain item |
| WO2022080583A1 (en) * | 2020-10-14 | 2022-04-21 | 고려대학교 세종산학협력단 | Deep learning-based bitcoin block data prediction system taking into account time series distribution characteristics |
| CN113114492A (en) * | 2021-04-01 | 2021-07-13 | 哈尔滨理工大学 | Security situation perception algorithm based on Markov differential game block chain model |
| CN114168948A (en) * | 2021-12-17 | 2022-03-11 | 北京华清信安科技有限公司 | Network security situation comprehensive analysis method |
| CN114638021A (en) * | 2022-03-18 | 2022-06-17 | 北京邮电大学 | Internet of things lightweight block chain system security evaluation method |
| CN115174165A (en) * | 2022-06-21 | 2022-10-11 | 重庆移通学院 | Self-adaptive security situation cognitive system based on block chain under cloud edge-side architecture |
Non-Patent Citations (4)
| Title |
|---|
| JEYAKUMAR SAMANTHA THARANI: "Graph Based Visualisation Techniques for Analysis of Blockchain Transactions", 2021 IEEE 46TH CONFERENCE ON LOCAL COMPUTER NETWORKS (LCN), 7 September 2021 (2021-09-07) * |
| 朱旭光: "交易数据全生命周期区块链隐私保护评估方法", 应用科学学报, 31 July 2022 (2022-07-31) * |
| 温啸林: "基于DPoS共识机制的区块链社区演化的可视分析方法", 计算机科学, 15 January 2022 (2022-01-15) * |
| 胡腾: "区块链分布式计算环境安全威胁研究", 中国博士学位论文全文数据库（信息科技辑）, 15 January 2022 (2022-01-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Saaty | Applications of analytical hierarchies | |
| Kou et al. | An integrated expert system for fast disaster assessment | |
| Piao et al. | Housing price prediction based on CNN | |
| Zahradníčková et al. | Scenarios as a strong support for strategic planning | |
| CN113537807B (en) | Intelligent wind control method and equipment for enterprises | |
| Nichols et al. | Evaluation of environmental assessment methods | |
| Sun | A study on the educational use of statistical package for the social sciences | |
| Zou et al. | Fast comprehensive flood risk assessment based on game theory and cloud model under parallel computation (P-GT-CM) | |
| CN113947322A (en) | Figure matching method and system based on FP-Growth algorithm | |
| Fu | Optimization study of multidimensional big data matrix model in enterprise performance evaluation system | |
| Bączkiewicz et al. | Multi-criteria temporal assessment of afordable and clean energy systems in European countries using the DARIA-TOPSIS method | |
| CN117455158A (en) | Device and method for generating grouping decision scheme of agent in typical scene | |
| Pareek et al. | A review report on knowledge discovery in databases and various techniques of data mining | |
| Liu et al. | The design of a moral education evaluation system for college students based on a deep learning model | |
| Chen et al. | A PROMETHEE II approach based on probabilistic hesitant fuzzy linguistic information with applications to multi-criteria group decision-making (ICSSE 2020) | |
| CN115941291A (en) | Analysis system and method for security situation awareness of DPoS (distributed denial of service) block chain network | |
| Omran et al. | Intelligent decision support system for the Egyptian food security | |
| Devulapalli et al. | Business analysis during the pandemic crisis using deep learning models | |
| CN112580992A (en) | Illegal collective risk monitoring system of similar financial enterprises | |
| Babaei Keshteli et al. | A combined machine learning algorithms and Interval DEA method for measuring predicting the efficiency | |
| Nandar et al. | Jobs Prediction Model based on Myanmar Labor Force Survey by using Artificial Neural Network | |
| Mondal et al. | Evaluation of the Main Barriers of RMG in Industry 4.0 Implementation by Using Analytical Hierarchy Process (AHP) | |
| CN117971511B (en) | Collaborative visual simulation platform | |
| Peng et al. | Research and Big Data Analysis on the Risk Factors of Banks Based on the AHP-EWM Model | |
| Shi et al. | The prediction of the basketball league championship based on AHP |
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 |