CN115065458A - Electronic commerce transaction system with data encryption transmission - Google Patents

Electronic commerce transaction system with data encryption transmission Download PDF

Info

Publication number
CN115065458A
CN115065458A CN202210944450.0A CN202210944450A CN115065458A CN 115065458 A CN115065458 A CN 115065458A CN 202210944450 A CN202210944450 A CN 202210944450A CN 115065458 A CN115065458 A CN 115065458A
Authority
CN
China
Prior art keywords
particles
transaction
particle
chaotic
time
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
Application number
CN202210944450.0A
Other languages
Chinese (zh)
Other versions
CN115065458B (en
Inventor
付舒丛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Dingxin Digital Technology Co ltd
Original Assignee
Beijing University of Posts and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Priority to CN202211235695.2A priority Critical patent/CN115619555B/en
Priority to CN202211235708.6A priority patent/CN115622774B/en
Priority to CN202210944450.0A priority patent/CN115065458B/en
Publication of CN115065458A publication Critical patent/CN115065458A/en
Application granted granted Critical
Publication of CN115065458B publication Critical patent/CN115065458B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/602—Providing cryptographic facilities or services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
    • 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

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Neurology (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Artificial Intelligence (AREA)
  • Technology Law (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Bioethics (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Development Economics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

An electronic commerce transaction system with data encryption transmission comprises an intelligent parameter setting module, a block chain module, an encryption module, an electronic commerce transaction evaluation module and a transaction module, wherein the intelligent parameter setting module is used for setting quantization requirements of a user, the block chain module is used for safely storing, updating and recording data and transaction activities, the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on the data, the electronic commerce transaction evaluation module selects a support vector machine, selects a proper electronic commerce transaction type for the user, and completes transactions in the transaction module. The invention has the beneficial effects that: the method can effectively prevent the leakage of personal information and transaction information of the client, and guarantee the electronic commerce transaction of the client.

Description

Electronic commerce transaction system with data encryption transmission
Technical Field
The invention relates to the field of electronic commerce, in particular to an electronic commerce transaction system for data encryption transmission.
Background
With the development of internet technology, network space has become the basis of people's survival and development in modern society, however, due to the insecurity of the internet, various information security problems exist, and network attacks include disguise, deception, eavesdropping, illegal access, tampering, repudiation, counterfeiting, denial of service, spreading viruses, and the like. The block chain is a new data structure, has dispersivity, does not need trust, is owned, managed and supervised by all nodes in the network, is not controlled by a single party, is a novel financial mode of commercial banks at present, mainly takes a core enterprise in electronic commerce as an entry point, provides related financial products and electronic commerce services through the contact of a plurality of enterprises in the electronic commerce, and can improve the business structure of the commercial banks to a certain extent, so that the commercial banks have stronger competitive advantages. Generally, the authenticity of data depends on the trust of a system center or a third-party entity, such as a main node, a central database, a system leader, a database manager and the like, once the system center is not trusted any more, the authenticity of the data is damaged and is difficult to find, so that it is necessary to encrypt the data of an e-commerce platform, the encrypted data of the digital e-commerce platform does not have the centralization of nodes, servers and databases, the operation and maintenance of the system are independent of managers, network nodes strictly package digital fingerprints of transaction information in a certain time into blocks and quickly broadcast the blocks to the whole network, and a hash technology is combined to form a tightly-linked chain among the blocks so as to form a highly-safe public account, namely a block chain, so that the block chain technology has a good effect on data encryption, but for the e-commerce platform data encryption adopting the block chain technology, the encryption process is complex, and the encrypted data is easy to be distorted and even lost, which affects the security and reliability of the data encryption of the e-commerce platform.
Disclosure of Invention
In view of the above problems, the present invention is directed to an e-commerce transaction system with encrypted data transmission.
The purpose of the invention is realized by the following technical scheme:
an electronic commerce transaction system for data encryption transmission is characterized by comprising an intelligent parameter setting module, a blockchain module, an encryption module, an electronic commerce transaction evaluation module and a transaction module, wherein the intelligent parameter setting module is used for quantifying the requirements of users and comprises transaction IDs, transaction types, timestamps, privacy levels, transaction objects, transaction addresses and transaction amounts, the intelligent parameter setting module is established in a dictionary mode and comprises keys and key values, the quantified requirements of the users are stored in the blockchain module, the blockchain module is used for safely storing, updating, recording data and transaction activities, electronic commerce transaction behaviors recorded in the blockchain are sorted, the data are grouped according to the transaction types, completely repeated data are deleted, default data are marked and supplemented in time, whether the key value result of each type of data in each group of transaction types falls into a certain interval or not is detected, the intervals comprise normal value intervals, abnormal value intervals and untrusted intervals, wherein the normal value intervals indicate that the key value stored by the user is correct, the abnormal value intervals indicate that the key value stored by the user is incorrect, the abnormal value is highlighted at the moment, the untrusted intervals indicate that the key value stored by the user is problematic, whether the result of data input is correct or not needs to be rechecked for the storage of the key value of the current time by the user, if the result of the key value input by the user is incorrect, the transaction behavior stored in the block chain is modified, and if the result of the key value input by the user is correct, the transaction behavior of the user needs to be rechecked; the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on data, the e-commerce transaction evaluation module selects a support vector machine to select a proper e-commerce transaction type for a user, and the user completes the transaction at the transaction module.
Furthermore, the intelligent parameter setting module is used for quantifying the requirements of the user, including transaction ID, transaction type, timestamp, security level, transaction object, transaction address and transaction amount, and is established in a dictionary manner, including keys and key values, and recorded as key values
Figure 914100DEST_PATH_IMAGE001
The timestamps are uniformly numbered through a hash function after being input according to standard time, the selection range of the timestamps is the combination of numbers and 26 lower case letters, and the security level comprises three levels of public security, common security and special security.
Further, by the block chain module, transaction behaviors in the block chain are sorted, data are distinguished according to transaction types, completely repeated data are deleted, default data need to be marked and supplemented in time, the key value result of each data in each group of transaction types falls into a certain interval, the interval comprises a normal value interval, an abnormal value interval and an untrusted interval, wherein the normal value interval represents that the key value stored by the user is correct, the abnormal value interval represents that the key value stored by the user is incorrect, the abnormal value can be highlighted at the moment, the untrusted interval represents that the key value stored by the user is problematic, whether the result of data input is correct or not needs to be rechecked for the storage of the key value of the current time by the user, if the result of the key value input by the user is incorrect, the transaction behaviors stored in the block chain are modified, and if the result of the key value input by the user is correct, then the transaction behavior of the user needs to be re-detected.
Furthermore, the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence to encrypt data, and the assumption is that
Figure 351904DEST_PATH_IMAGE002
Indicating an encryption operation, M indicates data that needs to be encrypted,
Figure 130504DEST_PATH_IMAGE003
representing decryption, C representing data needing decryption after encryption, and satisfying the following conditions:
Figure 670070DEST_PATH_IMAGE004
，
Figure 441585DEST_PATH_IMAGE005
and has:
Figure 750207DEST_PATH_IMAGE006
the symmetric block encryption algorithm based on the neural network chaotic sequence needs to meet the requirements that a plaintext to be encrypted is 56 bits, a check bit is 8 bits, and a total bit is 64 bits, and the method specifically comprises the following steps:
(1) firstly, carrying out initial transformation on a plaintext, dividing an information block into two parts, and then transforming a product through a function, wherein 16 times of execution are needed;
(2) after the product transformation, the two parts of information are combined to execute the inverse initial transformation operation, and the left-shifted message is changed into 48 bits;
(3) replace 32 bits of new data with the final result;
(4) and (4) executing replacement operation according to the steps (1) to (3), and finishing the encryption process after 16 execution cycles.
Further, the encryption module, assuming a given key, is adapted to encrypt the data stream
Figure 699708DEST_PATH_IMAGE007
If, if
Figure 195412DEST_PATH_IMAGE007
The generated subkey is
Figure 770618DEST_PATH_IMAGE008
，
Figure 933747DEST_PATH_IMAGE009
，
Figure 54149DEST_PATH_IMAGE010
Then, then
Figure 568307DEST_PATH_IMAGE007
Called weak key, satisfies:
Figure 436554DEST_PATH_IMAGE011
Figure 719768DEST_PATH_IMAGE012
Figure 11072DEST_PATH_IMAGE013
wherein the content of the first and second substances,
Figure 996214DEST_PATH_IMAGE014
to representBased on the symmetric block encryption algorithm of the neural network chaotic sequence,
Figure 663956DEST_PATH_IMAGE015
a decryption algorithm representing symmetric blocks based on a neural network chaotic sequence, if
Figure 536097DEST_PATH_IMAGE016
In all, there is
Figure 263882DEST_PATH_IMAGE017
Where M ', C ', and k ' are non-operational, a public key is used for encrypting data, and a private key is used for decrypting data, the same key is used for encryption and decryption each time in the conventional symmetric encryption system, and the public key encryption system uses two unrelated keys to ensure the security of the network, the security of data, and the security of the key itself, the expression is as follows:
Figure 736320DEST_PATH_IMAGE018
Figure 676594DEST_PATH_IMAGE019
Figure 934400DEST_PATH_IMAGE020
a symmetric block encryption algorithm based on a neural network chaotic sequence encrypts data of an electronic commerce platform, wherein the chaotic neural network consists of chaotic neurons, external input and internal feedback input, a single chaotic neuron is provided with feedback and external input items from the internal neuron as well as unstable items and threshold values from the neuron per se, and the chaotic neural network is composed of
Figure 567507DEST_PATH_IMAGE021
Of chaotic neural networks comprising chaotic neurons
Figure 261663DEST_PATH_IMAGE022
The equation for each neuron is as follows:
Figure 271207DEST_PATH_IMAGE023
wherein the content of the first and second substances,
Figure 383519DEST_PATH_IMAGE024
is the first
Figure 187527DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 637488DEST_PATH_IMAGE025
Is then outputted from the output of (a),
Figure 919564DEST_PATH_IMAGE026
is the first
Figure 886383DEST_PATH_IMAGE022
The continuous output function of each chaotic neuron,
Figure 110560DEST_PATH_IMAGE021
is the number of chaotic neurons that are,
Figure 795620DEST_PATH_IMAGE027
is the first
Figure 881387DEST_PATH_IMAGE028
A chaotic neuron and
Figure 702713DEST_PATH_IMAGE022
the connection weight of each chaotic neuron,
Figure 363370DEST_PATH_IMAGE029
is the first
Figure 270146DEST_PATH_IMAGE028
The axonal transformation transfer function of a chaotic neuron,
Figure 159605DEST_PATH_IMAGE030
is the number of external inputs that the user has,
Figure 101016DEST_PATH_IMAGE031
is the first
Figure 666995DEST_PATH_IMAGE028
An input and a
Figure 61068DEST_PATH_IMAGE022
The connection weight of each chaotic neuron,
Figure 754217DEST_PATH_IMAGE032
is a discrete time of (
Figure 530894DEST_PATH_IMAGE033
To get it at
Figure 284086DEST_PATH_IMAGE028
The intensity of the input,
Figure 165454DEST_PATH_IMAGE034
Is the first
Figure 396715DEST_PATH_IMAGE022
The refractoriness function of each chaotic neuron,
Figure 296407DEST_PATH_IMAGE007
is the attenuation coefficient of the degree of fire resistance,
Figure 220501DEST_PATH_IMAGE035
is a self-feedback coefficient, and
Figure 323586DEST_PATH_IMAGE036
，
Figure 358538DEST_PATH_IMAGE037
is the first
Figure 112736DEST_PATH_IMAGE022
A complete or unexcited threshold of chaotic neurons, if
Figure 207731DEST_PATH_IMAGE038
Is shown as
Figure 63692DEST_PATH_IMAGE039
A chaotic neuron in discrete time
Figure 636756DEST_PATH_IMAGE040
The iteration of the chaotic neural network is represented as follows:
Figure 511040DEST_PATH_IMAGE041
Figure 511357DEST_PATH_IMAGE042
for all neurons, function
Figure 589034DEST_PATH_IMAGE043
And
Figure 483565DEST_PATH_IMAGE044
is defined as
Figure 963088DEST_PATH_IMAGE045
In which
Figure 399886DEST_PATH_IMAGE046
As a function of the sign, i.e.:
Figure 964859DEST_PATH_IMAGE047
the external input intensity of each neuron at any time is set to the initial external input intensity value, i.e.:
Figure 128993DEST_PATH_IMAGE048
the value is 0 or 1, assuming that all firing thresholds per neuron are
Figure 630732DEST_PATH_IMAGE049
The method comprises the following steps:
Figure 238431DEST_PATH_IMAGE050
wherein, in the step (A),
Figure 290700DEST_PATH_IMAGE051
and
Figure 743679DEST_PATH_IMAGE052
are respectively the first
Figure 447061DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 225661DEST_PATH_IMAGE053
And
Figure 765227DEST_PATH_IMAGE054
internal state of (2), assume
Figure 287475DEST_PATH_IMAGE055
May take values of 1, 0 and-1, which, when in an excited state,
Figure 845365DEST_PATH_IMAGE056
when they are in the inhibition state, then
Figure 529287DEST_PATH_IMAGE057
When they are not directly connected to each other,
Figure 290569DEST_PATH_IMAGE058
based on the statistical characteristics of the neural network, the number of excitation connections and the number of inhibition connections are equal, the unpredictability of the output sequence of the neural network is increased, and because the chaotic neural network is introduced on the basis of a Hopfield neural network model with time lag, the requirement of constructing a connection matrix according to the Hopfield is met when
Figure 865776DEST_PATH_IMAGE059
When, suppose that
Figure 28904DEST_PATH_IMAGE058
To obtain the values of the connection weight matrix.
Further, for the parameters
Figure 149307DEST_PATH_IMAGE007
、
Figure 663465DEST_PATH_IMAGE035
And
Figure 514134DEST_PATH_IMAGE022
is selected from the group consisting of integers, secondly
Figure 531768DEST_PATH_IMAGE052
To make the non-periodic fluctuation centered at 0, the designed chaotic neural network is updated as follows:
Figure 88652DEST_PATH_IMAGE060
wherein
Figure 824526DEST_PATH_IMAGE061
For the correction factor, define a correction factor having
Figure 475956DEST_PATH_IMAGE062
A discrete Hopfield neural network of interconnected neurons, each neuron having a state of
Figure 613677DEST_PATH_IMAGE063
，
Figure 341461DEST_PATH_IMAGE064
Is 0 or 1, the next state
Figure 299053DEST_PATH_IMAGE065
Dependent on the current state of the neuron, i.e.
Figure 754174DEST_PATH_IMAGE066
Wherein, in the step (A),
Figure 746401DEST_PATH_IMAGE067
is a neuron
Figure 379507DEST_PATH_IMAGE022
And
Figure 339242DEST_PATH_IMAGE028
is a symmetric matrix,
Figure 348786DEST_PATH_IMAGE068
is a neuron
Figure 195520DEST_PATH_IMAGE022
The threshold value of (a) is set,
Figure 265107DEST_PATH_IMAGE064
is that
Figure 443629DEST_PATH_IMAGE069
Time of day
Figure 991285DEST_PATH_IMAGE039
The state of the individual neurons is determined,
Figure 958104DEST_PATH_IMAGE070
is that
Figure 933013DEST_PATH_IMAGE071
Time of day
Figure 867340DEST_PATH_IMAGE039
The state of the individual neurons is known,
Figure 218687DEST_PATH_IMAGE072
is that
Figure 40012DEST_PATH_IMAGE069
Time of day
Figure 185823DEST_PATH_IMAGE073
State of individual neuron, neural network in
Figure 341866DEST_PATH_IMAGE053
The energy over time is given in the following table:
Figure 496904DEST_PATH_IMAGE074
the energy function is monotonically decreased along with the evolution of the system state, and the neural network finally reaches a stable state due to limited energy, and is defined as an attractor, the attractor is a chaotic attractor, namely, the attractor is not related to the rule between the attractor and the initial state, and unpredictable relation exists between state messages in the attraction domain of each attractor, if the weight matrix is connected
Figure 907157DEST_PATH_IMAGE075
The attractors and their corresponding attraction fields will change accordingly, introducing a random transformation matrix
Figure 738716DEST_PATH_IMAGE076
Then, the original initial state is set
Figure 132788DEST_PATH_IMAGE077
And an attractor
Figure 825937DEST_PATH_IMAGE078
Respectively converted into new initial states
Figure 621855DEST_PATH_IMAGE079
And an attractor
Figure 361665DEST_PATH_IMAGE080
The method comprises the following steps:
Figure 977454DEST_PATH_IMAGE081
Figure 474295DEST_PATH_IMAGE082
the matrix of synaptic connections between neurons consists of +1, 0 and-1, with + l, 0 and-1 indicating that two neurons are in an excitatory state, no direct connection and an inhibitory state, respectively, and according to statistical probabilities, if there are more unpredictable attractors, the number of excitatory synaptic connections in the network is equal to the number of inhibitory synaptic connections, assuming that the number of samples stored in the network is 8 and the convergence field element is 20, the matrix of connecting synapses is given by:
Figure 124719DEST_PATH_IMAGE083
in order to ensure the encryption effect, it is necessary to perform security analysis on the symmetric block encryption algorithm based on the neural network chaotic sequence, and assuming that it is converted into a binary sequence C, it is derived by the following formula:
Figure 298080DEST_PATH_IMAGE084
wherein the content of the first and second substances,
Figure 666745DEST_PATH_IMAGE085
,
Figure 436118DEST_PATH_IMAGE086
the binary autocorrelation function R is as follows:
Figure 190316DEST_PATH_IMAGE087
sequence of
Figure 285311DEST_PATH_IMAGE088
And
Figure 875692DEST_PATH_IMAGE089
cross correlation function ofThe formula yields:
Figure 714335DEST_PATH_IMAGE090
on the basis, the design of a symmetric packet encryption algorithm based on a neural network chaotic sequence and an asymmetric packet encryption algorithm based on a neural network chaotic attractor is analyzed, and the method has safety in electronic commerce transaction transmission.
Furthermore, the electronic commerce transaction evaluation module optimizes kernel function parameters and punishment factors of the support vector machine by adopting a particle swarm algorithm.
Further, the particle swarm optimization is set to be carried out in a search space by adopting the following steps
Figure 588619DEST_PATH_IMAGE091
Updating the time:
step (1): performing similar updating detection on each particle in the particle swarm: order to
Figure 854515DEST_PATH_IMAGE092
Represents the second in the particle group
Figure 666614DEST_PATH_IMAGE093
The number of the particles is one,
Figure 43368DEST_PATH_IMAGE094
represents the second in the particle group
Figure 605167DEST_PATH_IMAGE095
Particles of
Figure 41964DEST_PATH_IMAGE092
And particles
Figure 606938DEST_PATH_IMAGE094
In that
Figure 787383DEST_PATH_IMAGE096
The time meets the following conditions:
Figure 370680DEST_PATH_IMAGE097
and is
Figure 978379DEST_PATH_IMAGE098
When it is, the particles are determined
Figure 765070DEST_PATH_IMAGE092
And particles
Figure 998474DEST_PATH_IMAGE094
In that
Figure 452589DEST_PATH_IMAGE096
The time instants are similar update particles, wherein,
Figure 231189DEST_PATH_IMAGE099
to represent
Figure 239596DEST_PATH_IMAGE096
Time particle
Figure 11112DEST_PATH_IMAGE092
At the location of the search space,
Figure 319734DEST_PATH_IMAGE100
represent
Figure 269235DEST_PATH_IMAGE096
Time particle
Figure 296097DEST_PATH_IMAGE094
At the location of the search space,
Figure 608654DEST_PATH_IMAGE101
to represent
Figure 37361DEST_PATH_IMAGE096
Time particle
Figure 157764DEST_PATH_IMAGE092
At the individual optimal position in the search space,
Figure 406343DEST_PATH_IMAGE102
to represent
Figure 254082DEST_PATH_IMAGE096
Time particle
Figure 537296DEST_PATH_IMAGE094
At the individual optimal position of the search space,
Figure 828600DEST_PATH_IMAGE103
is a group of particles in
Figure 813742DEST_PATH_IMAGE096
A similarity detection threshold for the time of day, and
Figure 215905DEST_PATH_IMAGE104
wherein, in the step (A),
Figure 353625DEST_PATH_IMAGE105
represent
Figure 815830DEST_PATH_IMAGE096
Time particle
Figure 288269DEST_PATH_IMAGE092
In the neighborhood of the search space, and
Figure 759701DEST_PATH_IMAGE106
wherein, in the step (A),
Figure 751928DEST_PATH_IMAGE107
to represent
Figure 385035DEST_PATH_IMAGE096
Distance position in time particle swarm
Figure 341840DEST_PATH_IMAGE099
First, the
Figure 351384DEST_PATH_IMAGE108
The position of the particles that are close to each other,
Figure 198117DEST_PATH_IMAGE109
is a given positive integer, an
Figure 251393DEST_PATH_IMAGE110
，
Figure 183577DEST_PATH_IMAGE111
Is the total number of particles in the population;
classifying the particles which are judged to be similar updating particles in the particle swarm specifically as follows: is provided with
Figure 996812DEST_PATH_IMAGE112
To represent
Figure 698052DEST_PATH_IMAGE096
The first one obtained by classifying the particles in the particle group and the like updated particles
Figure 187808DEST_PATH_IMAGE113
Class I, class II
Figure 607288DEST_PATH_IMAGE112
The particles in (a) are selected from the group of particles in the following way: randomly selecting one particle from the unclassified particles of the current particle swarm to be added into the class
Figure 693056DEST_PATH_IMAGE112
When the randomly selected particle does not have similar updating particle in the particle swarm, the selection of the particle in the swarm is stopped to add the class
Figure 514381DEST_PATH_IMAGE112
When the randomly selected particle has similar renewing particles in the particle group, adding the similar renewing particles of the randomly selected particle into the class
Figure 175038DEST_PATH_IMAGE112
And continuing to neutralize the current class with the population of particles
Figure 81815DEST_PATH_IMAGE112
Any one of the particles is a particle addition class of similarly updated particles
Figure 236852DEST_PATH_IMAGE112
In (1), until class
Figure 912684DEST_PATH_IMAGE112
Stopping selecting particles in the population to join the class when the medium particles do not have similar updating particles in the current population
Figure 481593DEST_PATH_IMAGE112
Performing the following steps;
is provided with
Figure 875666DEST_PATH_IMAGE114
To represent
Figure 568815DEST_PATH_IMAGE096
A class set obtained by classifying the particles in the particle swarm and the similar updated particles thereof at all times, an
Figure 614000DEST_PATH_IMAGE115
Wherein, in the step (A),
Figure 101614DEST_PATH_IMAGE116
representation collection
Figure 982982DEST_PATH_IMAGE114
The number of middle classes;
step (2): separately pair collections in the following manner
Figure 479822DEST_PATH_IMAGE117
In the above-mentioned classes of particles
Figure 379514DEST_PATH_IMAGE118
Updating the time: is provided with
Figure 38029DEST_PATH_IMAGE119
Presentation class
Figure 141114DEST_PATH_IMAGE112
The number of the particles in (2) is,
Figure 176066DEST_PATH_IMAGE120
given positive integer, for determining the set
Figure 195843DEST_PATH_IMAGE114
In a manner of updating the middle class particles, and
Figure 290838DEST_PATH_IMAGE121
class III of the related art
Figure 881220DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 719863DEST_PATH_IMAGE122
then, the following method is adopted to classify
Figure 325638DEST_PATH_IMAGE112
Middle particle of
Figure 591534DEST_PATH_IMAGE091
Updating the time:
Figure 669211DEST_PATH_IMAGE123
Figure 295234DEST_PATH_IMAGE124
in the above updating formula, let
Figure 774757DEST_PATH_IMAGE125
Presentation class
Figure 211554DEST_PATH_IMAGE112
To (1)
Figure 776528DEST_PATH_IMAGE126
The number of the particles is one,
Figure 206241DEST_PATH_IMAGE127
and
Figure 540270DEST_PATH_IMAGE128
respectively represent
Figure 147969DEST_PATH_IMAGE118
Time particle
Figure 934660DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 168064DEST_PATH_IMAGE129
and
Figure 356600DEST_PATH_IMAGE130
respectively represent
Figure 135200DEST_PATH_IMAGE096
Time particle
Figure 674765DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 449211DEST_PATH_IMAGE131
represents a group of particles in
Figure 757832DEST_PATH_IMAGE096
An inertial weight factor of a time of day, and
Figure 707334DEST_PATH_IMAGE132
，
Figure 734196DEST_PATH_IMAGE133
and
Figure 778244DEST_PATH_IMAGE134
respectively given a maximum inertia weight factor and a minimum inertia weight factor, and
Figure 941372DEST_PATH_IMAGE135
，
Figure 327354DEST_PATH_IMAGE136
，
Figure 825200DEST_PATH_IMAGE137
the maximum number of iterations is indicated,
Figure 689251DEST_PATH_IMAGE138
and
Figure 972465DEST_PATH_IMAGE139
are respectively in the interval
Figure 998190DEST_PATH_IMAGE140
The random number generated in the random number generator is used,
Figure 983332DEST_PATH_IMAGE141
to represent
Figure 651074DEST_PATH_IMAGE096
Time particle
Figure 788794DEST_PATH_IMAGE125
At the individual optimal position of the search space,
Figure 251000DEST_PATH_IMAGE142
to represent
Figure 720508DEST_PATH_IMAGE096
The time of day the particle swarm is at the global optimum location of the search space,
Figure 660783DEST_PATH_IMAGE143
a local learning factor representing a population of particles,
Figure 653009DEST_PATH_IMAGE144
a global learning factor representing a population of particles,
Figure 800963DEST_PATH_IMAGE143
and
Figure 511430DEST_PATH_IMAGE144
the value of (d) may take:
Figure 255395DEST_PATH_IMAGE145
，
Figure 367707DEST_PATH_IMAGE146
；
when class
Figure 420983DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 353167DEST_PATH_IMAGE147
then, the following method is adopted to classify
Figure 900823DEST_PATH_IMAGE112
Middle particle of
Figure 867642DEST_PATH_IMAGE091
Updating the time:
Figure 357398DEST_PATH_IMAGE148
Figure 776878DEST_PATH_IMAGE124
in the above-described update formula,
Figure 128225DEST_PATH_IMAGE149
to represent
Figure 949550DEST_PATH_IMAGE096
Time particle
Figure 347558DEST_PATH_IMAGE125
Inertial weight factor in the search space
Figure 254334DEST_PATH_IMAGE149
The values of (A) are set as:
Figure 143793DEST_PATH_IMAGE150
wherein the content of the first and second substances,
Figure 334472DEST_PATH_IMAGE151
presentation class
Figure 916763DEST_PATH_IMAGE152
The historical similarity coefficient of mesoparticle, and
Figure 310835DEST_PATH_IMAGE153
wherein, in the process,
Figure 3984DEST_PATH_IMAGE154
presentation class
Figure 783590DEST_PATH_IMAGE152
To (1)
Figure 536783DEST_PATH_IMAGE155
The number of the particles is one,
Figure 152572DEST_PATH_IMAGE156
to represent
Figure 649412DEST_PATH_IMAGE157
Time particle
Figure 549104DEST_PATH_IMAGE154
At the location of the search space,
Figure 207619DEST_PATH_IMAGE158
to represent
Figure 576283DEST_PATH_IMAGE159
Time of day particle
Figure 611235DEST_PATH_IMAGE160
At the location of the search space in the search space,
Figure 651520DEST_PATH_IMAGE161
is shown in the interval
Figure 215357DEST_PATH_IMAGE162
Internally generated random numbers.
The invention has the beneficial effects that: the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on data, and after electronic commerce data is encrypted and decrypted, the noise in the data is small, so that the accuracy of data encryption and decryption is ensured, the loss and distortion influence of the encrypted data are reduced, and the accuracy of the encrypted and decrypted data is ensured; the electronic commerce transaction evaluation module adopts an optimized particle swarm algorithm to optimize parameters of a support vector machine, firstly, similar update detection is carried out on particles in a particle swarm, the particles with similar update modes are classified, when the particles are updated, when the number of the particles in the similar update modes in the class of the particles is less, the class of the particles is set to be updated continuously according to the update mode of the standard particle swarm algorithm, so that the advantages of the standard particle swarm algorithm in optimization are kept, when the number of the particles in the similar update modes in the class of the particles is more, in order to ensure the diversity after particle swarm update, the similarity of the previous update step length of the class of the particles is judged by detecting the similarity of the previous positions of the class of the particles, when the previous positions of the class of the particles have larger differences, the step length when the class of the particles is updated previously is shown to have larger differences, at the moment, the inertia weight factors of the particles in the class are enhanced in the updating process, so that the particles in the class enhance the exploration of a new area, and the diversity of the positions of the particles in the class after updating is increased.
Drawings
The invention is further described with the aid of the accompanying drawings, in which, however, the embodiments do not constitute any limitation to the invention, and for a person skilled in the art, without inventive effort, further drawings may be derived from the following figures.
FIG. 1 is a schematic diagram of the present invention.
Detailed Description
The invention is further described with reference to the following examples.
Referring to fig. 1, the electronic commerce transaction system for data encryption transmission of this embodiment includes an intelligent parameter setting module, a blockchain module, an encryption module, an electronic commerce transaction evaluation module, and a transaction module, wherein the intelligent parameter setting module is configured to quantify a user's requirement, including a transaction ID, a transaction type, a timestamp, a privacy level, a transaction object, a transaction address, and a transaction amount, and is established in a dictionary manner, including a key and a key value, store the quantified user requirement in the blockchain module, the blockchain module is configured to securely store, update, record data and transaction activities, and arrange electronic commerce transaction behaviors recorded in the blockchain, group data according to the transaction type, delete completely repeated data, mark and timely supplement default data, detect whether a key value result of each data in each group of transaction types falls into a certain interval, the intervals comprise normal value intervals, abnormal value intervals and untrusted intervals, wherein the normal value intervals indicate that the key value stored by the user is correct, the abnormal value intervals indicate that the key value stored by the user is incorrect, the abnormal value is highlighted at the moment, the untrusted intervals indicate that the key value stored by the user is problematic, whether the result of data input is correct or not needs to be rechecked for the storage of the key value of the current time by the user, if the result of the key value input by the user is incorrect, the transaction behavior stored in the block chain is modified, and if the result of the key value input by the user is correct, the transaction behavior of the user needs to be rechecked; the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on data, the e-commerce transaction evaluation module selects a support vector machine to select a proper e-commerce transaction type for a user, and the user completes the transaction at the transaction module.
Specifically, the intelligent parameter setting module is used for quantifying the requirements of the user, including transaction ID, transaction type, timestamp, security level, transaction object, transaction address and transaction amount, and is established in a dictionary manner, including keys and key values, and recorded as
Figure 71317DEST_PATH_IMAGE001
The timestamps are uniformly numbered through a hash function after being input according to standard time, the selection range of the timestamps is the combination of numbers and 26 lower case letters, and the security level comprises three levels of public security, common security and special security.
Specifically, by using the blockchain module, transaction behaviors in the blockchain are sorted, data are distinguished according to transaction types, completely repeated data are deleted, default data need to be marked and supplemented in time, the key value result of each data in each group of transaction types falls into a certain interval, the interval comprises a normal value interval, an abnormal value interval and an untrusted interval, wherein the normal value interval indicates that the key value stored by the user is correct, the abnormal value interval indicates that the key value stored by the user is incorrect, the abnormal value can be highlighted at the moment, the untrusted interval indicates that the key value stored by the user is problematic, whether the result of data input is correct or not needs to be rechecked for the storage of the key value of the user at this time, if the result of the key value input by the user is incorrect, the transaction behaviors stored in the blockchain are modified, and if the result of the key value input by the user is correct, then the transaction behavior of the user needs to be re-detected.
Specifically, the encryption module encrypts data by adopting a symmetric block encryption algorithm based on a neural network chaotic sequence, and supposing that
Figure 159228DEST_PATH_IMAGE002
Indicating an encryption operation, M indicates data that needs to be encrypted,
Figure 518665DEST_PATH_IMAGE003
representing decryption, C representing data needing decryption after encryption, and satisfying the following conditions:
Figure 518982DEST_PATH_IMAGE004
，
Figure 596660DEST_PATH_IMAGE005
and has the following:
Figure 488261DEST_PATH_IMAGE006
the symmetric block encryption algorithm based on the neural network chaotic sequence needs to meet the requirements that a plaintext to be encrypted is 56 bits, a check bit is 8 bits, and a total bit is 64 bits, and the method specifically comprises the following steps:
(1) firstly, carrying out initial transformation on a plaintext, dividing an information block into two parts, and then transforming a product through a function, wherein 16 times of execution are needed;
(2) after the product transformation, the two parts of information are combined to execute the inverse initial transformation operation, and the left-shifted message is changed into 48 bits;
(3) replace 32 bits of new data with the final result;
(4) and (4) executing replacement operation according to the steps (1) to (3), and finishing the encryption process after 16 execution cycles.
Preferably, the encryption module, assuming a given key, is adapted to encrypt the encrypted data stream
Figure 967784DEST_PATH_IMAGE007
If, if
Figure 404582DEST_PATH_IMAGE007
Generated subkeyIs that
Figure 953244DEST_PATH_IMAGE008
，
Figure 133689DEST_PATH_IMAGE009
，
Figure 467719DEST_PATH_IMAGE010
Then, then
Figure 75417DEST_PATH_IMAGE007
Called weak key, satisfies:
Figure 379884DEST_PATH_IMAGE011
Figure 98442DEST_PATH_IMAGE012
Figure 552557DEST_PATH_IMAGE013
wherein the content of the first and second substances,
Figure 331157DEST_PATH_IMAGE014
represents a symmetric block encryption algorithm based on a neural network chaotic sequence,
Figure 854411DEST_PATH_IMAGE015
a decryption algorithm representing symmetric blocks based on a neural network chaotic sequence, if
Figure 376659DEST_PATH_IMAGE016
In all, there is
Figure 685281DEST_PATH_IMAGE017
Wherein, M ', C ' and k ' are non-operation, public key is used for encrypting data, private key is used for decrypting data, and the traditional symmetric encryption system uses the same key to encrypt and decrypt data each timeThe key encryption system uses two unrelated keys to ensure the security of the network, the security of data and the security of the key itself, and the expression is as follows:
Figure 634782DEST_PATH_IMAGE018
Figure 645332DEST_PATH_IMAGE019
Figure 971272DEST_PATH_IMAGE020
a symmetric block encryption algorithm based on a neural network chaotic sequence encrypts data of an electronic commerce platform, wherein the chaotic neural network consists of chaotic neurons, external input and internal feedback input, a single chaotic neuron is provided with feedback and external input items from the internal neuron as well as unstable items and threshold values from the neuron per se, and the chaotic neural network is composed of
Figure 868820DEST_PATH_IMAGE021
Of chaotic neural networks comprising chaotic neurons
Figure 238491DEST_PATH_IMAGE022
The equation for each neuron is as follows:
Figure 752649DEST_PATH_IMAGE023
wherein, the first and the second end of the pipe are connected with each other,
Figure 616699DEST_PATH_IMAGE024
is the first
Figure 634334DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 437555DEST_PATH_IMAGE025
Is then outputted from the output of (a),
Figure 173430DEST_PATH_IMAGE026
is the first
Figure 841172DEST_PATH_IMAGE022
The continuous output function of each chaotic neuron,
Figure 713313DEST_PATH_IMAGE021
is the number of chaotic neurons that are,
Figure 424786DEST_PATH_IMAGE027
is the first
Figure 647957DEST_PATH_IMAGE028
A chaotic neuron and
Figure 853810DEST_PATH_IMAGE022
the connection weight of each chaotic neuron,
Figure 846037DEST_PATH_IMAGE029
is the first
Figure 993990DEST_PATH_IMAGE028
The axonal transformation transfer function of the individual chaotic neurons,
Figure 438878DEST_PATH_IMAGE030
is the number of external inputs that the user has,
Figure 448422DEST_PATH_IMAGE031
is the first
Figure 544423DEST_PATH_IMAGE028
An input and a
Figure 348431DEST_PATH_IMAGE022
The connection weight of each chaotic neuron,
Figure 546194DEST_PATH_IMAGE032
is a discrete time(
Figure 93850DEST_PATH_IMAGE033
To get it at
Figure 312867DEST_PATH_IMAGE028
The intensity of each input,
Figure 287776DEST_PATH_IMAGE034
Is the first
Figure 972835DEST_PATH_IMAGE022
The refractoriness function of each chaotic neuron,
Figure 324182DEST_PATH_IMAGE007
is the attenuation coefficient of the degree of fire resistance,
Figure 394775DEST_PATH_IMAGE035
is a self-feedback coefficient, and
Figure 806165DEST_PATH_IMAGE036
，
Figure 447362DEST_PATH_IMAGE037
is the first
Figure 336820DEST_PATH_IMAGE022
Complete or unexcited threshold of chaotic neuron if
Figure 527499DEST_PATH_IMAGE038
Is shown as
Figure 844211DEST_PATH_IMAGE039
A chaotic neuron in discrete time
Figure 238283DEST_PATH_IMAGE040
The iteration of the chaotic neural network is represented as follows:
Figure 931433DEST_PATH_IMAGE041
Figure 976618DEST_PATH_IMAGE042
for all neurons, function
Figure 729810DEST_PATH_IMAGE043
And
Figure 611179DEST_PATH_IMAGE044
is defined as
Figure 823199DEST_PATH_IMAGE045
Wherein
Figure 473623DEST_PATH_IMAGE046
As a function of the sign, i.e.:
Figure 397716DEST_PATH_IMAGE047
the external input intensity of each neuron at any time is set to the initial external input intensity value, i.e.:
Figure 766381DEST_PATH_IMAGE048
the value is 0 or 1, assuming that all firing thresholds per neuron are
Figure 50601DEST_PATH_IMAGE049
The method comprises the following steps:
Figure 555531DEST_PATH_IMAGE050
wherein, in the step (A),
Figure 384947DEST_PATH_IMAGE051
and
Figure 240907DEST_PATH_IMAGE052
are respectively the first
Figure 63239DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 688255DEST_PATH_IMAGE053
And
Figure 954152DEST_PATH_IMAGE054
internal state of (2), assume
Figure 31829DEST_PATH_IMAGE055
May take values of 1, 0 and-1, which, when in an excited state,
Figure 923430DEST_PATH_IMAGE056
when they are in the inhibition state, then
Figure 402953DEST_PATH_IMAGE057
When they are not directly connected to each other,
Figure 839751DEST_PATH_IMAGE058
based on the statistical property of the neural network, the number of excitation connections and suppression connections is equal, the unpredictability of the output sequence of the neural network is increased, and the chaotic neural network is introduced on the basis of a Hopfield neural network model with time lag, so that the requirement of constructing a connection matrix according to Hopfield can be met when the chaotic neural network is used for solving the problem that the output sequence of the chaotic neural network is unpredictable
Figure 391342DEST_PATH_IMAGE059
When, suppose that
Figure 306209DEST_PATH_IMAGE058
To obtain the values of the connection weight matrix, the present embodiment takes the case where for M =8, the connection matrix is given by:
Figure 905817DEST_PATH_IMAGE163
in particular, for parameters
Figure 513516DEST_PATH_IMAGE007
、
Figure 815054DEST_PATH_IMAGE035
And
Figure 533611DEST_PATH_IMAGE022
is selected from the group consisting of integers, secondly
Figure 722147DEST_PATH_IMAGE052
To make the non-periodic fluctuation centered at 0, the designed chaotic neural network is updated as follows:
Figure 500747DEST_PATH_IMAGE060
wherein
Figure 289580DEST_PATH_IMAGE061
For the correction factor, define a correction factor having
Figure 811828DEST_PATH_IMAGE062
A discrete Hopfield neural network of interconnected neurons, each neuron having a state of
Figure 120450DEST_PATH_IMAGE063
，
Figure 69951DEST_PATH_IMAGE064
Is 0 or 1, the next state
Figure 80502DEST_PATH_IMAGE065
Dependent on the current state of the neuron, i.e.
Figure 406441DEST_PATH_IMAGE066
Wherein, in the step (A),
Figure 303990DEST_PATH_IMAGE067
is a neuron
Figure 670730DEST_PATH_IMAGE022
And
Figure 184888DEST_PATH_IMAGE028
is a symmetric matrix,
Figure 48939DEST_PATH_IMAGE068
is a neuron
Figure 66574DEST_PATH_IMAGE022
The threshold value of (2) is set,
Figure 872724DEST_PATH_IMAGE064
is that
Figure 343020DEST_PATH_IMAGE069
Time to
Figure 745183DEST_PATH_IMAGE039
The state of the individual neurons is known,
Figure 882903DEST_PATH_IMAGE070
is that
Figure 859955DEST_PATH_IMAGE071
Time to
Figure 817547DEST_PATH_IMAGE039
The state of the individual neurons is determined,
Figure 23400DEST_PATH_IMAGE072
is that
Figure 264894DEST_PATH_IMAGE069
Time of day
Figure 898001DEST_PATH_IMAGE073
State of individual neuron, neural network in
Figure 608468DEST_PATH_IMAGE053
The energy over time is given in the following table:
Figure 618012DEST_PATH_IMAGE074
the energy function is monotonically decreased along with the evolution of the system state, and the neural network finally reaches a stable state due to limited energy, and is defined as an attractor, the attractor is a chaotic attractor, namely, the attractor is not related to the rule between the attractor and the initial state, and unpredictable relation exists between state messages in the attraction domain of each attractor, if the weight matrix is connected
Figure 716943DEST_PATH_IMAGE075
The attractors and their corresponding attraction fields will change accordingly, introducing a random transformation matrix
Figure 786530DEST_PATH_IMAGE076
Then, the original initial state is set
Figure 718714DEST_PATH_IMAGE077
And an attractor
Figure 531949DEST_PATH_IMAGE078
Respectively converted into new initial states
Figure 748036DEST_PATH_IMAGE079
And an attractor
Figure 722945DEST_PATH_IMAGE080
The method comprises the following steps:
Figure 142425DEST_PATH_IMAGE081
Figure 493772DEST_PATH_IMAGE082
the matrix of synaptic connections between neurons consists of +1, 0 and-1, with + l, 0 and-1 indicating that two neurons are in an excitatory state, no direct connection and an inhibitory state, respectively, and according to statistical probabilities, if there are more unpredictable attractors, the number of excitatory synaptic connections in the network is equal to the number of inhibitory synaptic connections, assuming that the number of samples stored in the network is 8 and the convergence field element is 20, the matrix of connecting synapses is given by:
Figure 564365DEST_PATH_IMAGE083
in order to ensure the encryption effect, it is necessary to perform security analysis on the symmetric block encryption algorithm based on the neural network chaotic sequence, and assuming that it is converted into a binary sequence C, it is derived by the following formula:
Figure 710176DEST_PATH_IMAGE084
wherein, the first and the second end of the pipe are connected with each other,
Figure 616952DEST_PATH_IMAGE085
,
Figure 771990DEST_PATH_IMAGE086
the binary autocorrelation function R is as follows:
Figure 697089DEST_PATH_IMAGE087
sequence of
Figure 279380DEST_PATH_IMAGE088
And
Figure 673452DEST_PATH_IMAGE089
is derived from the following equation:
Figure 366602DEST_PATH_IMAGE090
on the basis, the design of a symmetric packet encryption algorithm based on a neural network chaotic sequence and an asymmetric packet encryption algorithm based on a neural network chaotic attractor is analyzed, and the method has safety in electronic commerce transaction transmission.
Furthermore, the electronic commerce transaction evaluation module optimizes kernel function parameters and punishment factors of the support vector machine by adopting a particle swarm algorithm.
Further, the particle swarm optimization is set to be carried out in a search space by adopting the following steps
Figure 166716DEST_PATH_IMAGE091
Updating the time:
step (1): performing similar updating detection on each particle in the particle swarm: order to
Figure 919908DEST_PATH_IMAGE092
Represents the second in the particle group
Figure 535697DEST_PATH_IMAGE093
The number of the particles is one,
Figure 281805DEST_PATH_IMAGE094
represents the second in the particle group
Figure 932229DEST_PATH_IMAGE095
Particles of
Figure 856323DEST_PATH_IMAGE092
And particles
Figure 224988DEST_PATH_IMAGE094
In that
Figure 243628DEST_PATH_IMAGE096
The time meets the following conditions:
Figure 748559DEST_PATH_IMAGE097
and is
Figure 843554DEST_PATH_IMAGE098
When it is, the particles are determined
Figure 433935DEST_PATH_IMAGE092
And particles
Figure 521846DEST_PATH_IMAGE094
In that
Figure 146862DEST_PATH_IMAGE096
The time instants are similar update particles, wherein,
Figure 412758DEST_PATH_IMAGE099
represent
Figure 490436DEST_PATH_IMAGE096
Time particle
Figure 853808DEST_PATH_IMAGE092
At the location of the search space,
Figure 333331DEST_PATH_IMAGE100
to represent
Figure 770129DEST_PATH_IMAGE096
Time particle
Figure 584370DEST_PATH_IMAGE094
At the location of the search space,
Figure 764816DEST_PATH_IMAGE101
to represent
Figure 364424DEST_PATH_IMAGE096
Time particle
Figure 972123DEST_PATH_IMAGE092
At the individual optimal position of the search space,
Figure 457681DEST_PATH_IMAGE102
represent
Figure 441818DEST_PATH_IMAGE096
Time particle
Figure 895933DEST_PATH_IMAGE094
At the individual optimal position of the search space,
Figure 658221DEST_PATH_IMAGE103
is a group of particles in
Figure 932208DEST_PATH_IMAGE096
A similarity detection threshold for the time of day, and
Figure 454456DEST_PATH_IMAGE104
wherein, in the step (A),
Figure 12345DEST_PATH_IMAGE105
to represent
Figure 961847DEST_PATH_IMAGE096
Time particle
Figure 988709DEST_PATH_IMAGE092
In the neighborhood of the search space, and
Figure 49068DEST_PATH_IMAGE106
wherein, in the step (A),
Figure 461464DEST_PATH_IMAGE107
to represent
Figure 581867DEST_PATH_IMAGE096
Distance position in time particle swarm
Figure 830446DEST_PATH_IMAGE099
First, the
Figure 694496DEST_PATH_IMAGE108
The position of the particles that are close to each other,
Figure 229907DEST_PATH_IMAGE109
is a given positive integer, an
Figure 521212DEST_PATH_IMAGE110
，
Figure 257086DEST_PATH_IMAGE111
Are particlesThe total number of particles in the population;
classifying the particles judged as similar update particles in the particle swarm, specifically comprising the following steps: is provided with
Figure 924828DEST_PATH_IMAGE112
To represent
Figure 46237DEST_PATH_IMAGE096
The first to classify the particles in the particle group and their similar updated particles
Figure 508442DEST_PATH_IMAGE113
Class I, class II
Figure 731613DEST_PATH_IMAGE112
The particles in (a) are selected from the group of particles in the following way: randomly selecting one particle from the unclassified particles of the current particle swarm to be added into the class
Figure 452313DEST_PATH_IMAGE112
When the randomly selected particle does not have similar updating particle in the particle swarm, the selection of the particle in the swarm is stopped to add the class
Figure 444540DEST_PATH_IMAGE112
When the randomly selected particle has similar renewing particles in the particle group, adding the similar renewing particles of the randomly selected particle into the class
Figure 77647DEST_PATH_IMAGE112
And continuing to neutralize the current class with the population of particles
Figure 522534DEST_PATH_IMAGE112
Any one of the particles is a particle addition class of similarly updated particles
Figure 781346DEST_PATH_IMAGE112
In (1), until class
Figure 628080DEST_PATH_IMAGE112
Stopping selecting particles in the population to join the class when the medium particles do not have similar updating particles in the current population
Figure 697667DEST_PATH_IMAGE112
Performing the following steps;
is provided with
Figure 629851DEST_PATH_IMAGE114
To represent
Figure 689424DEST_PATH_IMAGE096
A class set obtained by classifying the particles in the particle swarm and the similar updated particles thereof at all times, an
Figure 390664DEST_PATH_IMAGE115
Wherein, in the step (A),
Figure 365573DEST_PATH_IMAGE116
representation collection
Figure 34320DEST_PATH_IMAGE114
The number of middle classes;
step (2): separately pair collections in the following manner
Figure 385667DEST_PATH_IMAGE117
In the above-mentioned classes of particles
Figure 206993DEST_PATH_IMAGE118
Updating the time: is provided with
Figure 618383DEST_PATH_IMAGE119
Presentation class
Figure 774426DEST_PATH_IMAGE112
The number of particles in (a) is,
Figure 929464DEST_PATH_IMAGE120
given positive integer, for determining the set
Figure 605296DEST_PATH_IMAGE114
The update method of the class-I particles, and
Figure 922008DEST_PATH_IMAGE121
class III of the related art
Figure 565348DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 258497DEST_PATH_IMAGE122
then, the following method is adopted to classify
Figure 54415DEST_PATH_IMAGE112
Middle particle of
Figure 542028DEST_PATH_IMAGE091
Updating the time:
Figure 675594DEST_PATH_IMAGE123
Figure 172434DEST_PATH_IMAGE124
in the above updating formula, let
Figure 822858DEST_PATH_IMAGE125
Representation class
Figure 746952DEST_PATH_IMAGE112
To (1)
Figure 833726DEST_PATH_IMAGE126
The number of the particles is one,
Figure 868678DEST_PATH_IMAGE127
and
Figure 639188DEST_PATH_IMAGE128
respectively represent
Figure 983450DEST_PATH_IMAGE118
Time of day particle
Figure 573831DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 412474DEST_PATH_IMAGE129
and
Figure 37491DEST_PATH_IMAGE130
respectively represent
Figure 552655DEST_PATH_IMAGE096
Time of day particle
Figure 630332DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 7087DEST_PATH_IMAGE131
represents a group of particles in
Figure 486610DEST_PATH_IMAGE096
An inertial weight factor of a time of day, and
Figure 904166DEST_PATH_IMAGE132
，
Figure 469139DEST_PATH_IMAGE133
and
Figure 649585DEST_PATH_IMAGE134
respectively given a maximum inertia weight factor and a minimum inertia weight factor, and
Figure 983614DEST_PATH_IMAGE135
，
Figure 840581DEST_PATH_IMAGE136
，
Figure 892850DEST_PATH_IMAGE137
represents the maximumThe number of iterations is,
Figure 876987DEST_PATH_IMAGE138
and
Figure 65523DEST_PATH_IMAGE139
are respectively in the interval
Figure 93391DEST_PATH_IMAGE140
The random number generated in the random number generator is used,
Figure 632956DEST_PATH_IMAGE141
to represent
Figure 889625DEST_PATH_IMAGE096
Time particle
Figure 447514DEST_PATH_IMAGE125
At the individual optimal position of the search space,
Figure 131437DEST_PATH_IMAGE142
to represent
Figure 158299DEST_PATH_IMAGE096
The time of day the particle swarm is at the global optimum location of the search space,
Figure 205276DEST_PATH_IMAGE143
a local learning factor representing a population of particles,
Figure 368405DEST_PATH_IMAGE144
a global learning factor that represents a population of particles,
Figure 754387DEST_PATH_IMAGE143
and
Figure 2965DEST_PATH_IMAGE144
the value of (d) may take:
Figure 116284DEST_PATH_IMAGE145
，
Figure 399497DEST_PATH_IMAGE146
；
class III
Figure 690802DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 426676DEST_PATH_IMAGE147
then, the following method is adopted to classify
Figure 78106DEST_PATH_IMAGE112
Middle particle of
Figure 481406DEST_PATH_IMAGE091
Updating the time:
Figure 678032DEST_PATH_IMAGE148
Figure 901203DEST_PATH_IMAGE124
in the above-described update formula,
Figure 356324DEST_PATH_IMAGE149
to represent
Figure 348551DEST_PATH_IMAGE096
Time particle
Figure 247237DEST_PATH_IMAGE125
Inertial weight factor in the search space
Figure 957704DEST_PATH_IMAGE149
The values of (A) are set as:
Figure 971444DEST_PATH_IMAGE150
wherein the content of the first and second substances,
Figure 83757DEST_PATH_IMAGE151
presentation class
Figure 622185DEST_PATH_IMAGE152
The historical similarity coefficient of mesoparticle, and
Figure DEST_PATH_IMAGE164
wherein, in the step (A),
Figure 538058DEST_PATH_IMAGE154
presentation class
Figure 85714DEST_PATH_IMAGE152
To
Figure 52533DEST_PATH_IMAGE155
The number of the particles is one,
Figure 542289DEST_PATH_IMAGE156
to represent
Figure 961769DEST_PATH_IMAGE157
Time particle
Figure 313116DEST_PATH_IMAGE154
At the location of the search space,
Figure 134441DEST_PATH_IMAGE158
to represent
Figure 795098DEST_PATH_IMAGE159
Time particle
Figure 701875DEST_PATH_IMAGE160
At the location of the search space,
Figure 325754DEST_PATH_IMAGE161
is shown in the interval
Figure 267165DEST_PATH_IMAGE162
Internally generated random numbers.
The invention has the beneficial effects that: the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on data, and after electronic commerce data is encrypted and decrypted, the noise in the data is small, so that the accuracy of data encryption and decryption is ensured, the loss and distortion influence of the encrypted data are reduced, and the accuracy of the encrypted and decrypted data is ensured; the electronic commerce transaction evaluation module adopts an optimized particle swarm algorithm to optimize parameters of a support vector machine, firstly, similar update detection is carried out on particles in a particle swarm, the particles with similar update modes are classified, when the particles are updated, when the number of the particles in the similar update modes in the class of the particles is less, the class of the particles is set to be updated continuously according to the update mode of the standard particle swarm algorithm, so that the advantages of the standard particle swarm algorithm in optimization are kept, when the number of the particles in the similar update modes in the class of the particles is more, in order to ensure the diversity after particle swarm update, the similarity of the previous update step length of the class of the particles is judged by detecting the similarity of the previous positions of the class of the particles, when the previous positions of the class of the particles have larger differences, the step length when the class of the particles is updated previously is shown to have larger differences, at the moment, the inertia weight factors of the particles in the class are enhanced in the updating process, so that the particles in the class strengthen the exploration of a new area, and the diversity of the updated positions of the particles in the class is increased, on the contrary, when the previous positions of the particles in the class have greater similarity, the step length of the particles in the class during previous updating is indicated to have greater similarity, at the moment, the diversity of the particles in the class after updating is increased by setting the randomness of the current inertia weight factor values of the particles in the class, so that the optimizing precision and the convergence speed of the particle swarm algorithm are enhanced, and the support vector machine optimized based on the improved particle swarm algorithm has higher accuracy in the electronic commerce transaction evaluation process.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the protection scope of the present invention, although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims (8)

1. An electronic commerce transaction system for data encryption transmission is characterized by comprising an intelligent parameter setting module, a blockchain module, an encryption module, an electronic commerce transaction evaluation module and a transaction module, wherein the intelligent parameter setting module is used for quantifying the requirements of users and comprises transaction IDs, transaction types, timestamps, privacy levels, transaction objects, transaction addresses and transaction amounts, the intelligent parameter setting module is established in a dictionary mode and comprises keys and key values, the quantified requirements of the users are stored in the blockchain module, the blockchain module is used for safely storing, updating, recording data and transaction activities, electronic commerce transaction behaviors recorded in the blockchain are sorted, the data are grouped according to the transaction types, completely repeated data are deleted, default data are marked and supplemented in time, whether the key value result of each type of data in each group of transaction types falls into a certain interval or not is detected, the intervals comprise normal value intervals, abnormal value intervals and untrusted intervals, wherein the normal value intervals indicate that the key value stored by the user is correct, the abnormal value intervals indicate that the key value stored by the user is incorrect, the abnormal value is highlighted at the moment, the untrusted intervals indicate that the key value stored by the user is problematic, whether the result of data input is correct or not needs to be rechecked for the storage of the key value of the current time by the user, if the result of the key value input by the user is incorrect, the transaction behavior stored in the block chain is modified, and if the result of the key value input by the user is correct, the transaction behavior of the user needs to be rechecked; the encryption module adopts a symmetric block encryption algorithm based on a neural network chaotic sequence and an asymmetric encryption algorithm based on a neural network chaotic attractor to carry out secondary encryption on data, the e-commerce transaction evaluation module selects a support vector machine to select a proper e-commerce transaction type for a user, and the user completes the transaction at the transaction module.
2. The system of claim 1, wherein the intelligent parameter setting module is configured to quantify user requirements, including transaction ID, transaction type, timestamp, security level, transaction object, transaction address, and transaction amount, and is established in a dictionary manner, including keys and key values, and recorded as the key values
Figure 571102DEST_PATH_IMAGE001
The timestamps are uniformly numbered through a hash function after being input according to standard time, the selection range of the timestamps is the combination of numbers and 26 lower case letters, and the security level comprises three levels of public security, common security and special security.
3. The system of claim 1, wherein the blockchain module is configured to organize transaction behaviors in the blockchain, distinguish data according to transaction types, delete completely duplicated data, and default data needs to be labeled and supplemented in time, and the key value result of each data in each group of transaction types falls into a certain interval, where the interval includes a normal value interval, an abnormal value interval, and an untrusted interval, where the normal value interval indicates that the key value stored by the user is correct, the abnormal value interval indicates that the key value stored by the user is incorrect, the abnormal value is highlighted at the time, the untrusted interval indicates that the key value stored by the user is problematic, the user needs to recheck the storage of the key value of this time to see whether the result of data input is correct, and if the result of the key value input by the user is incorrect, the transaction behavior stored in the blockchain is modified, and if the result of the key value input by the user is correct, the transaction behavior of the user needs to be detected again.
4. The system of claim 1, wherein the encryption module encrypts the data using a symmetric block encryption algorithm based on a neural network chaotic sequence, assuming that the data is encrypted by the symmetric block encryption algorithm
Figure 412544DEST_PATH_IMAGE002
Indicating an encryption operation, M indicates data that needs to be encrypted,
Figure 661123DEST_PATH_IMAGE003
representing decryption, C representing data needing decryption after encryption, and satisfying the following conditions:
Figure 525173DEST_PATH_IMAGE004
，
Figure 57655DEST_PATH_IMAGE005
and has:
Figure 348959DEST_PATH_IMAGE006
the symmetric block encryption algorithm based on the neural network chaotic sequence needs to meet the requirements that a plaintext to be encrypted is 56 bits, a check bit is 8 bits, and a total bit is 64 bits, and the method specifically comprises the following steps:
(1) firstly, carrying out initial transformation on a plaintext, dividing an information block into two parts, and then transforming a product through a function, wherein 16 times of execution are needed;
(2) after the product transformation, the two parts of information are combined to execute the inverse initial transformation operation, and the left-shifted message is changed into 48 bits;
(3) replace 32 bits of new data with the final result;
(4) and (4) executing replacement operation according to the steps (1) to (3), and finishing the encryption process after 16 execution cycles.
5. The system of claim 4, wherein the encryption module assumes a given key
Figure 350413DEST_PATH_IMAGE007
If it is determined that
Figure 18155DEST_PATH_IMAGE007
The generated subkey is
Figure 405142DEST_PATH_IMAGE008
，
Figure 867348DEST_PATH_IMAGE009
，
Figure 90519DEST_PATH_IMAGE010
Then, then
Figure 296372DEST_PATH_IMAGE007
Called weak key, satisfies:
Figure 537867DEST_PATH_IMAGE011
Figure 436552DEST_PATH_IMAGE012
Figure 350282DEST_PATH_IMAGE013
wherein the content of the first and second substances,
Figure 890985DEST_PATH_IMAGE014
represents a symmetric block encryption algorithm based on a neural network chaotic sequence,
Figure 452897DEST_PATH_IMAGE015
a decryption algorithm representing symmetric blocks based on a neural network chaotic sequence, if
Figure 256905DEST_PATH_IMAGE016
In all, there is
Figure 454668DEST_PATH_IMAGE017
Where M ', C ', and k ' are non-operational, a public key is used for encrypting data, and a private key is used for decrypting data, the same key is used for encryption and decryption each time in the conventional symmetric encryption system, and the public key encryption system uses two unrelated keys to ensure the security of the network, the security of data, and the security of the key itself, the expression is as follows:
Figure 251592DEST_PATH_IMAGE018
Figure 218411DEST_PATH_IMAGE019
Figure 458899DEST_PATH_IMAGE020
a symmetric block encryption algorithm based on a neural network chaotic sequence encrypts data of an electronic commerce platform, wherein the chaotic neural network consists of chaotic neurons, external input and internal feedback input, a single chaotic neuron is provided with feedback and external input items from the internal neuron as well as unstable items and threshold values from the neuron per se, and the chaotic neural network is composed of
Figure 878379DEST_PATH_IMAGE021
Of chaotic neural networks comprising chaotic neurons
Figure 478994DEST_PATH_IMAGE022
The equation for each neuron is as follows:
Figure 300319DEST_PATH_IMAGE023
wherein the content of the first and second substances,
Figure 711709DEST_PATH_IMAGE024
is the first
Figure 618485DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 757211DEST_PATH_IMAGE025
The output of (a) is obtained,
Figure 698623DEST_PATH_IMAGE026
is the first
Figure 546493DEST_PATH_IMAGE022
The continuous output function of each chaotic neuron,
Figure 910872DEST_PATH_IMAGE021
is the number of chaotic neurons that are,
Figure 604021DEST_PATH_IMAGE027
is the first
Figure 337622DEST_PATH_IMAGE028
A chaotic neuron and
Figure 340082DEST_PATH_IMAGE022
the connection weight of each chaotic neuron,
Figure 221450DEST_PATH_IMAGE029
is the first
Figure 718291DEST_PATH_IMAGE028
The axonal transformation transfer function of a chaotic neuron,
Figure 368715DEST_PATH_IMAGE030
is the number of external inputs that the user has,
Figure 273567DEST_PATH_IMAGE031
is the first
Figure 642232DEST_PATH_IMAGE028
An input and a
Figure 677184DEST_PATH_IMAGE022
The connection weight of each chaotic neuron,
Figure 447694DEST_PATH_IMAGE032
is a discrete time of (
Figure 526377DEST_PATH_IMAGE033
To get it at
Figure 382337DEST_PATH_IMAGE028
The intensity of each input,
Figure 220980DEST_PATH_IMAGE034
Is the first
Figure 580418DEST_PATH_IMAGE022
The refractoriness function of each chaotic neuron,
Figure 95581DEST_PATH_IMAGE007
is the attenuation coefficient of the degree of fire resistance,
Figure 173259DEST_PATH_IMAGE035
is a self-feedback coefficient, and
Figure 815593DEST_PATH_IMAGE036
，
Figure 295116DEST_PATH_IMAGE037
is the first
Figure 981181DEST_PATH_IMAGE022
Complete or unexcited threshold of chaotic neuron if
Figure 546154DEST_PATH_IMAGE038
Is shown as
Figure 726600DEST_PATH_IMAGE039
A chaotic neuron in discrete time
Figure 60629DEST_PATH_IMAGE040
The iteration of the chaotic neural network is represented as follows:
Figure 920525DEST_PATH_IMAGE041
Figure 972795DEST_PATH_IMAGE042
for all neurons, function
Figure 691352DEST_PATH_IMAGE043
And
Figure 145467DEST_PATH_IMAGE044
is defined as
Figure 173335DEST_PATH_IMAGE045
Wherein
Figure 978480DEST_PATH_IMAGE046
As a function of the sign, i.e.:
Figure 500728DEST_PATH_IMAGE047
the external input intensity of each neuron at any time is set to the initial external input intensity value, i.e.:
Figure 809350DEST_PATH_IMAGE048
values of 0 or 1, each of which is assumed to beAll firing thresholds of neurons are
Figure 8119DEST_PATH_IMAGE049
The method comprises the following steps:
Figure 769402DEST_PATH_IMAGE050
wherein, in the step (A),
Figure 95341DEST_PATH_IMAGE051
and
Figure 258469DEST_PATH_IMAGE052
are respectively the first
Figure 628139DEST_PATH_IMAGE022
A chaotic neuron in discrete time
Figure 142297DEST_PATH_IMAGE053
And
Figure 6348DEST_PATH_IMAGE054
internal state of (2), assume
Figure 23982DEST_PATH_IMAGE055
May take values of 1, 0 and-1, which, when in an excited state,
Figure 850641DEST_PATH_IMAGE056
when they are in the inhibition state, then
Figure 586516DEST_PATH_IMAGE057
When they are not directly connected to each other,
Figure 254258DEST_PATH_IMAGE058
equalizing the number of excitatory connections and inhibitory connections based on statistical properties of the neural network and increasing unpredictability of the neural network output sequence since the chaotic neural network is at Ho with a time lagIntroduced on the basis of the model of the pfield neural network, and therefore, according to the requirement of constructing a connection matrix by the Hopfield
Figure 126399DEST_PATH_IMAGE059
When, suppose
Figure 103451DEST_PATH_IMAGE058
To obtain the values of the connection weight matrix.
6. The system of claim 5, wherein the parameter is a parameter of the electronic commerce transaction system
Figure 326622DEST_PATH_IMAGE007
、
Figure 532475DEST_PATH_IMAGE035
And
Figure 524702DEST_PATH_IMAGE022
is selected from the group consisting of integers, secondly
Figure 672655DEST_PATH_IMAGE052
To make the non-periodic fluctuation centered at 0, the designed chaotic neural network is updated as follows:
Figure 117543DEST_PATH_IMAGE060
wherein
Figure 127088DEST_PATH_IMAGE061
For the correction factor, define a correction factor having
Figure 239400DEST_PATH_IMAGE062
A discrete Hopfield neural network of interconnected neurons, each neuron having a state of
Figure 292676DEST_PATH_IMAGE063
，
Figure 490439DEST_PATH_IMAGE064
Is 0 or 1, the next state
Figure 38095DEST_PATH_IMAGE065
Dependent on the current state of the neuron, i.e.
Figure 4914DEST_PATH_IMAGE066
Wherein, in the step (A),
Figure 232020DEST_PATH_IMAGE067
is a neuron
Figure 917080DEST_PATH_IMAGE022
And
Figure 268426DEST_PATH_IMAGE028
is a symmetric matrix,
Figure 542282DEST_PATH_IMAGE068
is a neuron
Figure 953672DEST_PATH_IMAGE022
The threshold value of (a) is set,
Figure 860448DEST_PATH_IMAGE064
is that
Figure 749906DEST_PATH_IMAGE069
Time of day
Figure 940585DEST_PATH_IMAGE039
The state of the individual neurons is known,
Figure 257297DEST_PATH_IMAGE070
is that
Figure 651369DEST_PATH_IMAGE071
Time to
Figure 344519DEST_PATH_IMAGE039
The state of the individual neurons is known,
Figure 389704DEST_PATH_IMAGE072
is that
Figure 142896DEST_PATH_IMAGE069
Time of day
Figure 24265DEST_PATH_IMAGE073
State of individual neuron, neural network in
Figure 255526DEST_PATH_IMAGE053
The energy over time is given in the following table:
Figure 152288DEST_PATH_IMAGE074
the energy function is monotonically decreased along with the evolution of the system state, and the neural network finally reaches a stable state due to limited energy, and is defined as an attractor, the attractor is a chaotic attractor, namely, the attractor is not related to the rule between the attractor and the initial state, and unpredictable relation exists between state messages in the attraction domain of each attractor, if the weight matrix is connected
Figure 76382DEST_PATH_IMAGE075
The attractors and their corresponding attraction fields will change accordingly, introducing a random transformation matrix
Figure 445046DEST_PATH_IMAGE076
Then, the original initial state is set
Figure 479998DEST_PATH_IMAGE077
And an attractor
Figure 234196DEST_PATH_IMAGE078
Respectively converted into new initial states
Figure 329191DEST_PATH_IMAGE079
And an attractor
Figure 185152DEST_PATH_IMAGE080
The method comprises the following steps:
Figure 758216DEST_PATH_IMAGE081
Figure 632500DEST_PATH_IMAGE082
the matrix of synaptic connections between neurons consists of +1, 0 and-1, with + l, 0 and-1 indicating that two neurons are in an excitatory state, no direct connection and an inhibitory state, respectively, and according to statistical probabilities, if there are more unpredictable attractors, the number of excitatory synaptic connections in the network is equal to the number of inhibitory synaptic connections, assuming that the number of samples stored in the network is 8 and the convergence field element is 20, the matrix of connecting synapses is given by:
Figure 898396DEST_PATH_IMAGE083
in order to ensure the encryption effect, it is necessary to perform security analysis on the symmetric block encryption algorithm based on the neural network chaotic sequence, and assuming that it is converted into a binary sequence C, it is derived by the following formula:
Figure 976073DEST_PATH_IMAGE084
wherein the content of the first and second substances,
Figure 618407DEST_PATH_IMAGE085
,
Figure 347198DEST_PATH_IMAGE086
the binary autocorrelation function R is as follows:
Figure 783995DEST_PATH_IMAGE087
sequence of
Figure 348969DEST_PATH_IMAGE088
And
Figure 263835DEST_PATH_IMAGE089
the cross correlation function of (a) is given by:
Figure 115641DEST_PATH_IMAGE090
on the basis, the design of a symmetric packet encryption algorithm based on a neural network chaotic sequence and an asymmetric packet encryption algorithm based on a neural network chaotic attractor is analyzed, and the method has safety in electronic commerce transaction transmission.
7. The system of claim 1, wherein the e-commerce transaction evaluation module optimizes the kernel function parameters and penalty factors of the SVM using a PSO algorithm.
8. The system of claim 7, wherein the particle swarm algorithm is configured to perform the following steps in the search space
Figure 723340DEST_PATH_IMAGE091
Updating the time:
step (1): performing similar updating detection on each particle in the particle swarm: order to
Figure 775610DEST_PATH_IMAGE092
Represents the second in the particle group
Figure 494167DEST_PATH_IMAGE093
The number of the particles is one,
Figure 197550DEST_PATH_IMAGE094
represents the second in the particle group
Figure 976150DEST_PATH_IMAGE095
Particles of
Figure 515715DEST_PATH_IMAGE092
And particles
Figure 37964DEST_PATH_IMAGE094
In that
Figure 595853DEST_PATH_IMAGE096
The time meets the following conditions:
Figure 545354DEST_PATH_IMAGE097
and is
Figure 306637DEST_PATH_IMAGE098
When it is, the particles are determined
Figure 632576DEST_PATH_IMAGE092
And particles
Figure 44972DEST_PATH_IMAGE094
In that
Figure 165374DEST_PATH_IMAGE096
The time instants are similar update particles, wherein,
Figure 679532DEST_PATH_IMAGE099
to represent
Figure 543583DEST_PATH_IMAGE096
Time particle
Figure 807556DEST_PATH_IMAGE092
At the location of the search space,
Figure 364439DEST_PATH_IMAGE100
to represent
Figure 100314DEST_PATH_IMAGE096
Time particle
Figure 502476DEST_PATH_IMAGE094
At the location of the search space,
Figure 889464DEST_PATH_IMAGE101
to represent
Figure 617249DEST_PATH_IMAGE096
Time particle
Figure 574840DEST_PATH_IMAGE092
At the individual optimal position of the search space,
Figure 46273DEST_PATH_IMAGE102
to represent
Figure 287767DEST_PATH_IMAGE096
Time particle
Figure 920874DEST_PATH_IMAGE094
At the individual optimal position of the search space,
Figure 631341DEST_PATH_IMAGE103
is a group of particles in
Figure 640885DEST_PATH_IMAGE096
A similarity detection threshold for the time of day, and
Figure 736886DEST_PATH_IMAGE104
wherein, in the step (A),
Figure 806473DEST_PATH_IMAGE105
to represent
Figure 738657DEST_PATH_IMAGE096
Time particle
Figure 551893DEST_PATH_IMAGE092
In the neighborhood of the search space, and
Figure 770909DEST_PATH_IMAGE106
wherein, in the step (A),
Figure 745818DEST_PATH_IMAGE107
to represent
Figure 430877DEST_PATH_IMAGE096
Distance position in time particle swarm
Figure 782224DEST_PATH_IMAGE099
First, the
Figure 852817DEST_PATH_IMAGE108
The position of the particles that are close to each other,
Figure 998628DEST_PATH_IMAGE109
is a given positive integer, an
Figure 905404DEST_PATH_IMAGE110
，
Figure 60442DEST_PATH_IMAGE111
Is the total number of particles in the population;
classifying the particles which are judged to be similar updating particles in the particle swarm specifically as follows: is provided with
Figure 985541DEST_PATH_IMAGE112
To represent
Figure 567832DEST_PATH_IMAGE096
The first one obtained by classifying the particles in the particle group and the like updated particles
Figure 961905DEST_PATH_IMAGE113
Class I, class II
Figure 920633DEST_PATH_IMAGE112
The particles in (a) are selected from the group of particles in the following way: randomly selecting one particle from the unclassified particles of the current particle swarm to be added into the class
Figure 965819DEST_PATH_IMAGE112
When the randomly selected particle does not have similar updating particle in the particle swarm, the selection of the particle in the swarm is stopped to add the class
Figure 453432DEST_PATH_IMAGE112
When the randomly selected particle has similar renewing particles in the particle group, adding the similar renewing particles of the randomly selected particle into the class
Figure 69221DEST_PATH_IMAGE112
And continuing to neutralize the current class with the population of particles
Figure 566061DEST_PATH_IMAGE112
Any one of the particles is a particle addition class of similar update particles
Figure 728403DEST_PATH_IMAGE112
In (1), until class
Figure 386917DEST_PATH_IMAGE112
Stopping selecting particles in the population to join the class when the medium particles do not have similar updating particles in the current population
Figure 755581DEST_PATH_IMAGE112
Performing the following steps;
is provided with
Figure 790534DEST_PATH_IMAGE114
To represent
Figure 544732DEST_PATH_IMAGE096
A class set obtained by classifying the particles in the particle swarm and the similar updated particles thereof at all times, an
Figure 639727DEST_PATH_IMAGE115
Wherein, in the step (A),
Figure 495687DEST_PATH_IMAGE116
representation collection
Figure 334330DEST_PATH_IMAGE114
The number of middle classes;
step (2): separately pair collections in the following manner
Figure 943035DEST_PATH_IMAGE117
In the above-mentioned classes of particles
Figure 208931DEST_PATH_IMAGE118
Updating the time: is provided with
Figure 286609DEST_PATH_IMAGE119
Presentation class
Figure 928943DEST_PATH_IMAGE112
The number of the particles in (2) is,
Figure 657733DEST_PATH_IMAGE120
for a given positive integer, for determining the set
Figure 94531DEST_PATH_IMAGE114
The update method of the class-I particles, and
Figure 659504DEST_PATH_IMAGE121
class III of the related art
Figure 839950DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 691756DEST_PATH_IMAGE122
then, the following method is adopted to classify
Figure 299455DEST_PATH_IMAGE112
Middle particle of
Figure 86145DEST_PATH_IMAGE091
Updating the time:
Figure 70282DEST_PATH_IMAGE123
Figure 773664DEST_PATH_IMAGE124
in the above updating formula, let
Figure 552264DEST_PATH_IMAGE125
Presentation class
Figure 826251DEST_PATH_IMAGE112
To (1)
Figure 348499DEST_PATH_IMAGE126
The number of the particles is one,
Figure 113847DEST_PATH_IMAGE127
and
Figure 63348DEST_PATH_IMAGE128
respectively represent
Figure 339478DEST_PATH_IMAGE118
Time particle
Figure 399837DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 828545DEST_PATH_IMAGE129
and
Figure 401477DEST_PATH_IMAGE130
respectively represent
Figure 650056DEST_PATH_IMAGE096
Time particle
Figure 310845DEST_PATH_IMAGE125
At the location and step size of the search space,
Figure 797321DEST_PATH_IMAGE131
represents a group of particles in
Figure 337892DEST_PATH_IMAGE096
An inertial weight factor of a time of day, and
Figure 73767DEST_PATH_IMAGE132
，
Figure 741509DEST_PATH_IMAGE133
and
Figure 879229DEST_PATH_IMAGE134
respectively given a maximum inertia weight factor and a minimum inertia weight factor, and
Figure 593632DEST_PATH_IMAGE135
，
Figure 816803DEST_PATH_IMAGE136
，
Figure 288235DEST_PATH_IMAGE137
the maximum number of iterations is indicated,
Figure 280462DEST_PATH_IMAGE138
and
Figure 162836DEST_PATH_IMAGE139
are respectively in the interval
Figure 873303DEST_PATH_IMAGE140
The random number generated in the random number generator is used,
Figure 882848DEST_PATH_IMAGE141
to represent
Figure 729581DEST_PATH_IMAGE096
Time of day particle
Figure 48436DEST_PATH_IMAGE125
At the individual optimal position in the search space,
Figure 980620DEST_PATH_IMAGE142
to represent
Figure 793855DEST_PATH_IMAGE096
The time of day the particle swarm is at the global optimum location of the search space,
Figure 760674DEST_PATH_IMAGE143
a local learning factor representing a population of particles,
Figure 250430DEST_PATH_IMAGE144
a global learning factor representing a population of particles,
Figure 404331DEST_PATH_IMAGE143
and
Figure 21257DEST_PATH_IMAGE144
the value of (d) may take:
Figure 577003DEST_PATH_IMAGE145
，
Figure 234731DEST_PATH_IMAGE146
；
class III
Figure 141507DEST_PATH_IMAGE112
The medium particle satisfies:
Figure 296545DEST_PATH_IMAGE147
then, the following method is adopted to classify
Figure 972377DEST_PATH_IMAGE112
Middle particle of
Figure 803935DEST_PATH_IMAGE091
Updating the time:
Figure 198008DEST_PATH_IMAGE148
Figure 891157DEST_PATH_IMAGE124
in the above-described update formula,
Figure 687075DEST_PATH_IMAGE149
to represent
Figure 423956DEST_PATH_IMAGE096
Time of day particle
Figure 305324DEST_PATH_IMAGE125
Inertial weight factor in the search space
Figure 802164DEST_PATH_IMAGE149
The values of (A) are set as:
Figure 452588DEST_PATH_IMAGE150
wherein, the first and the second end of the pipe are connected with each other,
Figure 625950DEST_PATH_IMAGE151
representation class
Figure 729035DEST_PATH_IMAGE152
The historical similarity coefficient of mesoparticle, and
Figure 763987DEST_PATH_IMAGE153
wherein, in the step (A),
Figure 534497DEST_PATH_IMAGE154
representation class
Figure 147268DEST_PATH_IMAGE152
To (1)
Figure 472071DEST_PATH_IMAGE155
The number of the particles is one,
Figure 576293DEST_PATH_IMAGE156
to represent
Figure 935730DEST_PATH_IMAGE157
Time particle
Figure 716473DEST_PATH_IMAGE154
At the location of the search space,
Figure 794150DEST_PATH_IMAGE158
to represent
Figure 170905DEST_PATH_IMAGE159
Time particle
Figure 650428DEST_PATH_IMAGE125
At the location of the search space,
Figure 336493DEST_PATH_IMAGE160
is shown in the interval
Figure 901467DEST_PATH_IMAGE161
Internally generated random numbers.
CN202210944450.0A 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission Active CN115065458B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN202211235695.2A CN115619555B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission function
CN202211235708.6A CN115622774B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system based on improved particle swarm optimization and supporting data encryption transmission of vector machine
CN202210944450.0A CN115065458B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210944450.0A CN115065458B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission

Related Child Applications (2)

Application Number Title Priority Date Filing Date
CN202211235695.2A Division CN115619555B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission function
CN202211235708.6A Division CN115622774B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system based on improved particle swarm optimization and supporting data encryption transmission of vector machine

Publications (2)

Publication Number Publication Date
CN115065458A true CN115065458A (en) 2022-09-16
CN115065458B CN115065458B (en) 2022-11-15

Family

ID=83208514

Family Applications (3)

Application Number Title Priority Date Filing Date
CN202210944450.0A Active CN115065458B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission
CN202211235708.6A Active CN115622774B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system based on improved particle swarm optimization and supporting data encryption transmission of vector machine
CN202211235695.2A Active CN115619555B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission function

Family Applications After (2)

Application Number Title Priority Date Filing Date
CN202211235708.6A Active CN115622774B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system based on improved particle swarm optimization and supporting data encryption transmission of vector machine
CN202211235695.2A Active CN115619555B (en) 2022-08-08 2022-08-08 Electronic commerce transaction system with data encryption transmission function

Country Status (1)

Country Link
CN (3) CN115065458B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116308655A (en) * 2023-03-17 2023-06-23 北京未来链技术有限公司 Intelligent community commodity big data intelligent bill spelling and settlement system based on blockchain

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116405283B (en) * 2023-04-06 2023-11-24 广州大有网络科技有限公司 Data encryption authentication system based on information data protection
CN116308813B (en) * 2023-05-17 2023-08-08 青岛农村商业银行股份有限公司 Different-industry combined financial equity safety management system
CN117035836A (en) * 2023-10-08 2023-11-10 深圳市焕想科技有限公司 Electronic commerce transaction data processing method and system based on artificial intelligence

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109165224A (en) * 2018-08-24 2019-01-08 东北大学 A kind of indexing means being directed to keyword key on block chain database
CN110611701A (en) * 2018-08-21 2019-12-24 汇链丰(北京)科技有限公司 Parameter configuration and transaction processing method based on block chain
US20200371833A1 (en) * 2019-05-24 2020-11-26 International Business Machines Corporation Anomalous transaction commitment prevention for database
CN112927072A (en) * 2021-01-20 2021-06-08 北京航空航天大学 Block chain-based anti-money laundering arbitration method, system and related device
CN113723954A (en) * 2021-06-15 2021-11-30 复旦大学 Method for detecting and supervising abnormal transaction nodes in block chain
US20210397891A1 (en) * 2020-06-17 2021-12-23 Capital One Services, Llc Anomaly analysis using a blockchain, and applications thereof

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001144746A (en) * 1999-11-11 2001-05-25 Japan Science & Technology Corp Enchiper system and decoding system using chaos neural network
JP2003023421A (en) * 2001-07-09 2003-01-24 C4 Technology Inc Encryption method, program thereof, recording medium recorded with the program, encryption device, decoding method, and decoder
CN101977112B (en) * 2010-11-04 2013-10-09 厦门大学 Public key cipher encrypting and decrypting method based on neural network chaotic attractor
CN106485348A (en) * 2016-09-22 2017-03-08 中国银联股份有限公司 A kind of Forecasting Methodology of transaction data and device
CN107231214B (en) * 2017-06-12 2020-07-28 哈尔滨工程大学 Optimal multi-user detection method based on evolutionary chaotic quantum neural network
US10542046B2 (en) * 2018-06-07 2020-01-21 Unifyvault LLC Systems and methods for blockchain security data intelligence
CN110852604A (en) * 2019-11-08 2020-02-28 湖南商学院 Dynamic trust calculation method based on mobile Agent
CN112632842A (en) * 2020-12-23 2021-04-09 国网北京市电力公司 Trading harmony based power grid and building energy consumption trading matching method and system
CN112765271B (en) * 2020-12-31 2023-02-07 杭州趣链科技有限公司 Block chain transaction index storage method and device, computer equipment and medium
CN113554511B (en) * 2021-06-23 2023-06-27 河海大学 Active power distribution network power transaction method based on blockchain and particle swarm optimization
CN114331436A (en) * 2021-12-30 2022-04-12 江苏九一网络科技有限公司 E-commerce safe transaction system based on block chain

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110611701A (en) * 2018-08-21 2019-12-24 汇链丰(北京)科技有限公司 Parameter configuration and transaction processing method based on block chain
CN109165224A (en) * 2018-08-24 2019-01-08 东北大学 A kind of indexing means being directed to keyword key on block chain database
US20200371833A1 (en) * 2019-05-24 2020-11-26 International Business Machines Corporation Anomalous transaction commitment prevention for database
US20210397891A1 (en) * 2020-06-17 2021-12-23 Capital One Services, Llc Anomaly analysis using a blockchain, and applications thereof
CN112927072A (en) * 2021-01-20 2021-06-08 北京航空航天大学 Block chain-based anti-money laundering arbitration method, system and related device
CN113723954A (en) * 2021-06-15 2021-11-30 复旦大学 Method for detecting and supervising abnormal transaction nodes in block chain

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116308655A (en) * 2023-03-17 2023-06-23 北京未来链技术有限公司 Intelligent community commodity big data intelligent bill spelling and settlement system based on blockchain
CN116308655B (en) * 2023-03-17 2023-09-15 北京未来链技术有限公司 Intelligent community commodity big data intelligent bill spelling and settlement system based on blockchain

Also Published As

Publication number Publication date
CN115622774B (en) 2024-01-12
CN115619555B (en) 2023-06-09
CN115622774A (en) 2023-01-17
CN115065458B (en) 2022-11-15
CN115619555A (en) 2023-01-17

Similar Documents

Publication Publication Date Title
Akcora et al. Bitcoinheist: Topological data analysis for ransomware detection on the bitcoin blockchain
CN115065458B (en) Electronic commerce transaction system with data encryption transmission
Feng et al. Mining network data for intrusion detection through combining SVMs with ant colony networks
Wazirali An improved intrusion detection system based on KNN hyperparameter tuning and cross-validation
Pradeep Mohan Kumar et al. Intrusion detection system based on GA‐fuzzy classifier for detecting malicious attacks
McDermott et al. Investigation of computational intelligence techniques for intrusion detection in wireless sensor networks.
Prabhakaran et al. Hybrid semantic deep learning architecture and optimal advanced encryption standard key management scheme for secure cloud storage and intrusion detection
Subaira et al. Efficient classification mechanism for network intrusion detection system based on data mining techniques: a survey
Panigrahi et al. Two-stage database intrusion detection by combining multiple evidence and belief update
Torabi et al. Practical autoencoder based anomaly detection by using vector reconstruction error
Benmalek et al. Security of federated learning: Attacks, defensive mechanisms, and challenges
Kadry et al. Intrusion detection model using optimized quantum neural network and elliptical curve cryptography for data security
Moorthy et al. A study of Intrusion Detection using data mining
Ge et al. A review of secure federated learning: Privacy leakage threats, protection technologies, challenges and future directions
Wang et al. Practical blockchain-based steganographic communication via adversarial ai: A case study in bitcoin
Hasan et al. Optimization algorithms for intrusion detection system: a review
Babbar et al. FRHIDS: Federated Learning Recommender Hydrid Intrusion Detection System Model in Software Defined Networking for Consumer Devices
Shorfuzzaman Detection of cyber attacks in IoT using tree-based ensemble and feedforward neural network
Tauscher et al. Learning to detect: A data-driven approach for network intrusion detection
Luqman et al. Privacy and Security Implications of Cloud-Based AI Services: A Survey
Pal et al. InfoSec: A Comprehensive Study
Vähäkainu et al. Adversarial Poisoning Attack's Impact on Prediction Functionality of ML-Based Feedback Loop System in Cyber-Physical Context
Kannari et al. Network Intrusion Detection, Prevention and Sustainable Mechanisms: A Survey.
Chen et al. VerFHS: Verifiable image retrieval on forward privacy in blockchain-enabled IoT
Selvamani et al. An efficacious intellectual framework for host based intrusion detection system

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20221024

Address after: Room 1401, No. 169, Yingchun Street, Laishan District, Yantai City, Shandong Province, 264000

Applicant after: Shandong Dingxin Digital Technology Co.,Ltd.

Address before: 100000 Beijing city Haidian District Xitucheng Road No. 10

Applicant before: Beijing University of Posts and Telecommunications

GR01 Patent grant
GR01 Patent grant