CN115759757A - Transaction risk early warning method and system and electronic equipment - Google Patents
Transaction risk early warning method and system and electronic equipment Download PDFInfo
- Publication number
- CN115759757A CN115759757A CN202211615887.6A CN202211615887A CN115759757A CN 115759757 A CN115759757 A CN 115759757A CN 202211615887 A CN202211615887 A CN 202211615887A CN 115759757 A CN115759757 A CN 115759757A
- Authority
- CN
- China
- Prior art keywords
- early warning
- transaction
- risk early
- sample data
- transaction risk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Storage Device Security (AREA)
Abstract
The invention discloses a transaction risk early warning method, a transaction risk early warning system and electronic equipment, and relates to the technical field of safety early warning. The transaction risk early warning method is mainly applied to a credible execution environment, and in the implementation process, transaction data are input into a transaction risk early warning model obtained through deep learning training based on confidential calculation, so that a transaction risk early warning result can be obtained, the credible execution environment supported by computer hardware can be utilized, the privacy safety of sample data and model data used in deep learning is protected, the safety of the deep learning model is improved, the real-time performance and the accuracy of transaction early warning are improved, and the risk in the transaction process is effectively avoided.
Description
Technical Field
The invention relates to the technical field of safety early warning, in particular to a transaction risk early warning method, a transaction risk early warning system and electronic equipment.
Background
Deep learning is a new field in machine learning research, and the motivation is to establish and simulate a neural network for analyzing and learning of human brain. In recent years, deep learning has become a research hotspot and a mainstream development direction in the field of artificial intelligence, and has been widely applied to various scenes in social production and life, such as financial systems. The characteristics of continuity, high dimensionality, time-varying property and the like of the financial data of the bank are more suitable for the deep learning model, and a great deal of innovation and application possibility can be brought to the aspects of risk prevention and control, intelligent service, customer maintenance and the like of a commercial bank by utilizing the deep learning technology.
However, as deep learning applications gradually start to become a reality, data security issues for financial systems also gradually emerge. On one hand, training for deep learning requires collecting a large amount of sample data, which is much related to privacy information of individuals, enterprises, organizations and other entities; on the other hand, the deep learning model is a core data asset with great value, and in the process of deep learning model training and application, sample data and the model face a plurality of security threats such as external stealing and malicious tampering. Although there are many technical schemes for protecting data privacy in the deep learning field at present, such as a differential privacy technology for sample data protection, a homomorphic encryption technology for model protection, federal learning, and the like, the existing technical schemes generally have disadvantages in that: the protection of the deep learning computing platform is not enough, that is, strong attackers can still steal and destroy the data in operation by breaking the computing platform and the operation environment where the deep learning is located and then directly reading and writing the memory. In the face of such threat models, the security capabilities of deep learning need to be further enhanced.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a transaction risk early warning method, a transaction risk early warning system and electronic equipment.
In order to achieve the purpose, the invention provides the following scheme:
a transaction risk early warning method, the method implemented in a trusted execution environment; the method comprises the following steps:
acquiring transaction data; the transaction data includes: trading user information, trading channels, trading modes and trading money;
acquiring a transaction risk early warning model; the transaction risk early warning model is obtained through deep learning training based on confidential calculation;
and inputting the transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
Preferably, the process of obtaining the transaction risk early warning model based on the deep learning training of the confidential calculation comprises the following steps:
extracting sample data from historical transaction stream data, and preprocessing the sample data to obtain a tag set corresponding to the sample data;
respectively encrypting the sample data and the tag set to obtain an encrypted sample data set, and forming a sample data set-key mapping relation;
in the trusted execution environment, a key is obtained based on the sample data set-key mapping relation, and the encrypted sample data set is decrypted by adopting the key to obtain plaintext sample data and a label set corresponding to the plaintext sample data;
acquiring an initial deep learning model;
training the initial deep learning model by taking the plaintext sample data as the input and taking a label set corresponding to the plaintext sample data as the output to obtain a trained deep learning model;
and taking the trained deep learning model as the transaction risk early warning model.
Preferably, the pre-treatment comprises: cleaning treatment, null value removing treatment, normalization treatment and label treatment.
Preferably, the process of forming the sample data set-key mapping relationship includes:
determining a hash value of the set of encrypted sample data;
and establishing the mapping relation between the sample data set and the key based on the hash value and the key.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the transaction risk early warning method provided by the invention is mainly applied to a credible execution environment, and in the implementation process, transaction data is input into a transaction risk early warning model obtained by deep learning training based on confidential calculation, so that a transaction risk early warning result can be obtained, the credible execution environment supported by computer hardware can be utilized, the privacy safety of sample data and model data used by deep learning is protected, the safety of the deep learning model is improved, the real-time performance and the accuracy of transaction early warning are improved, and the risk in the transaction process is effectively avoided.
Corresponding to the transaction risk early warning method, the invention also provides the following implementation structure:
a transaction risk early warning system, the system implemented in a trusted execution environment; the system comprises:
the transaction data acquisition module is used for acquiring transaction data; the transaction data includes: trading user information, trading channels, trading modes and trading money;
the transaction risk early warning model acquisition module is used for acquiring a transaction risk early warning model; the transaction risk early warning model is obtained through deep learning training based on confidential calculation;
and the transaction risk early warning result generating module is used for inputting the transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
An electronic device, comprising:
a memory for storing computer logic control instructions; the computer logic control instruction is used for implementing the transaction risk early warning method;
and the processor is connected with the memory and used for calling and executing the computer logic control instruction.
Preferably, the memory is a computer-readable storage medium.
Since the technical effects achieved by the two implementation structures provided by the invention are the same as the technical effects achieved by the transaction risk early warning method provided by the invention, no further description is given here.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow chart of a transaction risk early warning method provided by the present invention;
FIG. 2 is a schematic diagram of a deep learning training process according to an embodiment of the present invention;
FIG. 3 is a diagram of an implementation architecture for deep learning training according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a transaction risk early warning system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a transaction risk early warning method, a transaction risk early warning system and electronic equipment, which can improve the safety of adopting a deep learning model, improve the real-time performance and accuracy of transaction early warning and further effectively avoid risks in a transaction process.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The invention provides a transaction risk early warning method which is mainly implemented in a trusted execution environment; confidential computing, which refers to the use of a hardware-based Trusted Execution Environment (TEE) to provide protection for data in use, is introduced into deep learning practices. The method is a hardware-based technology, data, specific functions and application programs are isolated from the common environment, the data are stored in the TEE, the data cannot be viewed or operated from the outside, and only authorized codes can access the data. The privacy computation has core characteristics of provability, data confidentiality protection, data security protection and the like.
As shown in fig. 1, the transaction risk early warning method includes:
step 100: transaction data is obtained. The transaction data includes: the transaction information, the transaction channel, the transaction mode and the transaction amount.
Step 101: and acquiring a transaction risk early warning model. The transaction risk early warning model is obtained through deep learning training based on confidential calculation.
Step 102: and inputting the transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
In the process of obtaining a transaction risk early warning model through deep learning training based on confidential calculation, the invention provides a deep learning training method of a Trusted Execution Environment (TEE) based on confidential calculation, which comprises the following specific training processes:
step 1, extracting sample data from business data, such as extracting personal information, a transaction channel, a transaction mode, money amount and the like of a customer from historical transaction flow of a bank, and preprocessing the sample data to obtain sample data and a label set. And respectively encrypting the sample data and the tag set thereof to obtain an encrypted sample data set, and submitting the sample data set-key mapping relation to a key management server. Specifically, the method comprises the following steps:
step 1-0, extracting sample data from business data, such as extracting personal information, a transaction channel, a transaction mode, amount and the like of a client from daily transaction flow of a bank, and carrying out preprocessing such as cleaning, null value removal, normalization and labeling to obtain the sample data and a label set.
Step 1-1, symmetrically encrypting the sample data and the label set by using a key K to obtain an encrypted sample data set Ds.
And 1-2, storing the encrypted sample data set Ds to a storage server.
Step 1-3, solving a hash value of the encrypted sample data set Ds, and establishing a mapping relation hash (Ds) -K.
And step 1-4, submitting the mapping relation hash (Ds) -K to a key management server for storage.
And 2, loading the encrypted sample data set in the TEE, and initiating authentication to a key management server to obtain a corresponding key. Specifically, the method comprises the following steps:
and 2-1, loading an encryption sample data set Ds in the TEE and respectively obtaining a hash value hash (Ds).
And 2-2, creating a digital report in a user-defined format in the TEE, and adding the hash value hash (Ds) into the digital report to obtain a digital report R.
And 2-3, deriving a signature key through the chip layer trusted root in the TEE, and signing the digital report R obtained in the step 2-2.
And 2-4, generating a self-signed certificate in the TEE, adding the digital report R in the step into an extended domain of the self-signed certificate, and then initiating a remote TLS connection with the key management server based on the certificate.
And 2-5, after receiving the TLS connection request of the remote TEE, the key management server requests authentication, and returns a key K to the TEE after the authentication is successful.
The specific authentication process comprises the following steps:
and 2-5-1, the key management server acquires a self-signed certificate from a TLS request from the TEE and takes out the digital report R from the self-signed certificate.
Step 2-5-2, the key management server takes the hash value hash (Ds) of the sample data set from the digital report R
And 2-5-3, the key management server uses a public key and a certificate chain in the TEE white list to check the signature of the digital report R, and if the signature is successfully checked, the TLS connection is kept. Otherwise, the authentication fails, and the TLS connection is closed.
And 2-5-4, searching the hash (Ds) -K from the database for the TEE successfully authenticated by the key management server, and if the hash (Ds) -K is found, sending the key K to the TEE.
And 3, decrypting the encrypted sample data set by using the secret key in the TEE to obtain sample data of a plaintext and a label set thereof, and starting deep learning training. Specifically, the method comprises the following steps:
and 3-1, after the TEE succeeds in authentication and acquires the secret key K, loading the encrypted sample data set into the TEE, and decrypting the encrypted sample data set respectively to obtain plaintext sample data and a label set.
And 3-2, taking the plaintext sample data and the tag set as input, starting deep learning training in the TEE, and destroying the sample data and the tag set after the training is finished.
And 3-3, encrypting the trained deep learning model, storing the encrypted deep learning model persistently, and loading the deep learning model to the TEE for reasoning operation when in use.
And 4, after the training is finished, obtaining a deep learning model, encrypting and storing the deep learning model, and destroying the sample data set in the TEE.
The confidential computing Trusted Execution Environment (TEE) adopted in the invention has the following main characteristics:
A. sensitive data is not visible in the TEE based on computer specific CPU instruction support.
And B, executing training of the deep learning model in the TEE, and storing the trained model in a persistent mode after the trained model is encrypted.
The key management server is characterized as follows:
A. an encrypted storage and retrieval service for sample data set-key mapping relationships is provided.
B. A trusted authentication and key provisioning service for a confidential computing environment (TEE) is provided that maintains a white list of TEEs obtained from TEE hardware vendors for matching digital report signatures submitted by remote TEEs.
And the encryption module provides certificate/key generation, signature/signature verification and Hash calculation services.
And the storage server is used for storing the encrypted sample data set (including the tag set).
Deep learning frameworks include, but are not limited to, pythrch, tensorflow, and the like.
The specific implementation process of the transaction risk early warning method provided above is described below by taking bank anti-money laundering transactions as an example, and the method is not limited to this in the practical application process.
Example one
A bank is provided, an anti-money laundering transaction early warning system is constructed, and an anti-money laundering monitoring model is trained through a deep learning method to identify whether a money laundering behavior exists in a certain transaction. The input data of deep learning is sample data and a tag set, wherein the sample data includes but is not limited to customer information, transaction channels, transaction modes, money amount and the like of banks. Obviously, the sample data contains highly sensitive private data, and the sample data needs to be kept secret highly during the deep learning training process. Similarly, the trained anti-money laundering detection model also needs to be highly confidential as a core data asset.
As shown in fig. 2, the hardware configuration for implementing this embodiment includes:
a confidential computing Trusted Execution Environment (TEE), specific types include, but are not limited to, intel-SGX, ARM-TrustZone, AMD-SEV, or the like. TEE is characterized in that when deep learning training is performed, sample data and models will be isolated in a hardware protected "enclave" with data only available and not visible.
And the key management server is used for carrying out credible authentication on the TEE so as to prove that the TEE of the opposite terminal is real and effective. After the authentication is successful, the key of the encrypted sample data is sent to the TEE.
The sample data set (including the tag set) of the deep learning is extracted from the actual banking data in this embodiment, and includes the personal information of the bank customer, the transaction channel, the transaction mode, the transaction amount, and the like. And after the generation of sample data, encrypting the generated sample data, storing the encrypted sample data in a storage server, submitting a corresponding key to a key management server, loading the encrypted sample data set from the storage server and decrypting the encrypted sample data set when deep learning training starts to be performed, inputting the encrypted sample data set into deep learning for training, and destroying the sample data set after training is completed.
Deep learning training frameworks include, but are not limited to, pythrch, tensorflow, openCV, and the like.
And (5) deeply learning the model. The deep learning model obtained through training is encrypted and persisted by the TEE. In a particular model usage scenario (e.g., money laundering transaction authentication using the model in this example), the model will be loaded into the TEE for inferencing operations.
Specifically, the implementation steps of this embodiment are shown in fig. 3, and include:
step 1, generating sample data and a tag set from transaction business of a bank, wherein the characteristic dimension of the sample data is determined according to the actual data source and the deep learning requirement and can comprise basic information of both transaction parties, transaction channel information, a transaction mode, transaction amount and other necessary characteristic dimensions. The tag set contains a classification designation (0 or 1) of whether there is money laundering behavior for each piece of sample data.
And 2, symmetrically encrypting the sample data and the label set (hereinafter collectively referred to as the sample data set), wherein the encryption algorithm comprises but is not limited to AES, RC and the like.
And 3, storing the encrypted sample data set to a storage server.
And 4, solving hash values (hash algorithms include but are not limited to MD5, SHA-1, SHA256 and the like) of the encrypted sample data set, and storing the hash values and the encryption keys in a key management server after one-to-one correspondence.
And 5, the TEE starts to execute deep learning training, and the encrypted sample data set is loaded from the storage server at first, wherein the loading mode comprises but is not limited to online network transmission, such as HTTP, TCP and the like. Or off-line medium transmission such as USB, mobile hard disk and the like.
Step 6, carrying out remote trusted authentication on the TEE, and specifically comprising the following steps:
A. the TEE-based hardware (typically referred to as CPU chips and on-chip ROM) generates a digital report including the TEE vendor ID, version number, TEE base attributes, etc.
B. The hash value is solved for the encrypted set of sample data loaded into the TEE and added to the digital report. The digital report is signed using a key derived from the hardware root of trust. The signed digital report is unique to each TEE.
C. And maintaining a TEE authentication white list at one side of the key management server, wherein an authentication public key, a certificate chain and other TEE authentication information which are disclosed by a TEE manufacturer are stored, and the white list synchronizes the authentication information with the TEE manufacturer regularly in an online or offline mode.
D. Before the deep learning training begins, a remote authentication request based on TLS is initiated by the TEE side to the key management server, and the request contains a signed digital report in the form of a TLS certificate.
E. And after receiving the authentication request of the TEE, the key management server acquires the digital report, verifies the digital report by relying on a public key and a certificate chain in a white list, and meanwhile verifies whether the hash value of the encrypted sample set in the digital report exists in a key database.
F. And if the key manager successfully checks the TEE and the key database has the corresponding hash value, the authentication is successful, and the key corresponding to the hash value is sent to the TEE. Otherwise, the authentication fails, and the authentication link is closed.
And 7, the TEE obtains a secret key after the remote authentication is successful, decrypts the sample set and starts deep learning training.
And 8, obtaining a deep network model after training is completed, and encrypting the deep network model by the TEE and then persistently storing the deep network model.
And 9, loading the model by the TEE and executing inference operation when the model is used for reasoning.
Example two
This embodiment is still exemplified by the bank anti-money-laundering application, and further describes the application practice and the resulting beneficial effects of the present invention in deep learning.
The basic flow of the bank for realizing the anti-money laundering application based on the invention is as follows:
step 1, extracting sample data required by deep learning from historical transaction business data, and generating a sample set after feature engineering.
And 2, training a money laundering transaction monitoring model according to the deep learning training method provided by the invention, wherein the model is stored in a ciphertext format in a lasting manner.
And 3, reloading the monitoring model into the TEE during actual application, and starting the online reasoning service of the model.
And 4, monitoring daily transaction behaviors by the bank, extracting transaction business data according to batches, and inputting the transaction business data into the TEE for reasoning and calculation after characteristic engineering.
And 5, calculating the probability of money laundering behavior of each transaction by model reasoning, and outputting the probability as a result.
And 6, taking the money laundering probability as an important basis by the bank to take corresponding early warning or intervention measures.
The application example is characterized in that:
firstly, sample data of deep learning training exists in a ciphertext state after being generated, even when training operation is executed, the sample data is isolated and protected by TEE hardware in a memory, and any third party without trust cannot read and tamper the sample data even if ROOT authority is obtained.
Secondly, the trained model data is stored in a ciphertext format, and even when inference operation is executed, the model is also isolated and protected by TEE hardware and cannot be read and tampered.
According to the invention, a secret computing technology is introduced into the deep learning field, so that the security threat from a computing platform in the deep learning technology practice is effectively responded, the data privacy protection of the deep learning full process is realized, and more favorable conditions are created for the popularization and the application of the deep learning technology.
Based on the above description, the present invention also has the following advantages over the prior art:
1. compared with the existing deep learning training method, the method provided by the invention improves the data security in the deep learning process, and realizes the privacy protection of the whole process of deep learning data processing, model training and model application.
2. In the invention, the hash value of the deep-learning sensitive data is used as the identity label to run through the whole process, so that the sensitive data can be efficiently stored, retrieved, authenticated and transmitted.
In addition, corresponding to the transaction risk early warning method provided by the invention, the invention also provides the following implementation structure:
a transaction risk early warning system is implemented in a trusted execution environment. As shown in fig. 4, the system includes:
a transaction data obtaining module 400, configured to obtain transaction data. The transaction data includes: the transaction information, the transaction channel, the transaction mode and the transaction amount.
A transaction risk early warning model obtaining module 401, configured to obtain a transaction risk early warning model. The transaction risk early warning model is obtained through deep learning training based on confidential calculation.
A transaction risk early warning result generating module 402, configured to input transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
An electronic device, comprising:
a memory for storing computer logic control instructions. The computer logic control instruction is used for implementing the transaction risk early warning method.
And the processor is connected with the memory and used for calling and executing the computer logic control instruction.
Wherein, the memory is a computer scale storage medium.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the description of the method part.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (7)
1. A transaction risk early warning method is characterized in that the method is implemented in a trusted execution environment; the method comprises the following steps:
acquiring transaction data; the transaction data includes: trading user information, trading channels, trading modes and trading amounts;
acquiring a transaction risk early warning model; the transaction risk early warning model is obtained through deep learning training based on confidential calculation;
and inputting the transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
2. The transaction risk early warning method according to claim 1, wherein the process of obtaining the transaction risk early warning model based on the deep learning training of the confidential calculation comprises:
extracting sample data from historical transaction running water data, and preprocessing the sample data to obtain a tag set corresponding to the sample data;
respectively encrypting the sample data and the tag set to obtain an encrypted sample data set, and forming a sample data set-key mapping relation;
in the trusted execution environment, a key is obtained based on the sample data set-key mapping relation, and the encrypted sample data set is decrypted by adopting the key to obtain plaintext sample data and a tag set corresponding to the plaintext sample data;
obtaining an initial deep learning model;
training the initial deep learning model by taking the plaintext sample data as the input and taking a label set corresponding to the plaintext sample data as the output to obtain a trained deep learning model;
and taking the trained deep learning model as the transaction risk early warning model.
3. The transaction risk early warning method according to claim 2, wherein the preprocessing comprises: cleaning treatment, null value removing treatment, normalization treatment and label treatment.
4. The transaction risk early warning method according to claim 2, wherein the forming process of the sample data set-key mapping relationship comprises:
determining a hash value of the set of encrypted sample data;
and establishing the sample data set-key mapping relation based on the hash value and the key.
5. A transaction risk early warning system, wherein the system is implemented in a trusted execution environment; the system comprises:
the transaction data acquisition module is used for acquiring transaction data; the transaction data includes: trading user information, trading channels, trading modes and trading money;
the transaction risk early warning model acquisition module is used for acquiring a transaction risk early warning model; the transaction risk early warning model is obtained through deep learning training based on confidential calculation;
and the transaction risk early warning result generating module is used for inputting the transaction data into the transaction risk early warning model to obtain a transaction risk early warning result.
6. An electronic device, comprising:
a memory for storing computer logic control instructions; the computer logic control instructions are used for implementing the transaction risk early warning method according to any one of claims 1 to 4;
and the processor is connected with the memory and used for calling and executing the computer logic control instruction.
7. The electronic device of claim 6, wherein the memory is a computer-readable storage medium.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211615887.6A CN115759757A (en) | 2022-12-15 | 2022-12-15 | Transaction risk early warning method and system and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211615887.6A CN115759757A (en) | 2022-12-15 | 2022-12-15 | Transaction risk early warning method and system and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115759757A true CN115759757A (en) | 2023-03-07 |
Family
ID=85346102
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211615887.6A Pending CN115759757A (en) | 2022-12-15 | 2022-12-15 | Transaction risk early warning method and system and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115759757A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117852019A (en) * | 2024-03-06 | 2024-04-09 | 天逸财金科技服务(武汉)有限公司 | Digital asset circulation method and system based on cryptography |
-
2022
- 2022-12-15 CN CN202211615887.6A patent/CN115759757A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117852019A (en) * | 2024-03-06 | 2024-04-09 | 天逸财金科技服务(武汉)有限公司 | Digital asset circulation method and system based on cryptography |
CN117852019B (en) * | 2024-03-06 | 2024-05-24 | 天逸财金科技服务(武汉)有限公司 | Digital asset circulation method and system based on cryptography |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Dasgupta et al. | A survey of blockchain from security perspective | |
CN108681966B (en) | Information supervision method and device based on block chain | |
CN111770200B (en) | Information sharing method and system | |
CN111770198B (en) | Information sharing method, device and equipment | |
Lin et al. | Internet of things intrusion detection model and algorithm based on cloud computing and multi-feature extraction extreme learning machine | |
CN111770199B (en) | Information sharing method, device and equipment | |
CN111460525B (en) | Block chain-based data processing method, device and storage medium | |
US7194759B1 (en) | Used trusted co-servers to enhance security of web interaction | |
CN111818186B (en) | Information sharing method and system | |
CN113302610B (en) | Trusted platform based on blockchain | |
US11533175B1 (en) | Systems and methods for post-quantum cryptography on a smartcard | |
EP4092984A1 (en) | Data processing method and apparatus, device and medium | |
CN115380303A (en) | Trusted platform based on block chain | |
CN111814193B (en) | Information sharing method, device and equipment | |
CN111585995B (en) | Secure wind control information transmission and processing method and device, computer equipment and storage medium | |
CN110417557B (en) | Intelligent terminal peripheral data security control method and device | |
CN116226289A (en) | Electronic certificate management method, device, equipment and storage medium based on blockchain | |
CN115759757A (en) | Transaction risk early warning method and system and electronic equipment | |
Singh et al. | Cloud computing security using blockchain technology | |
CN115085934A (en) | Contract management method based on block chain and combined key and related equipment | |
CN114978664A (en) | Data sharing method and device and electronic equipment | |
Jayapandian et al. | A novel approach to enhance multi level security system using encryption with fingerprint in cloud | |
WO2022141057A1 (en) | Blockchain anonymous user auditing method, electronic device, and storage medium | |
Agarwal et al. | Detecting malicious accounts showing adversarial behavior in permissionless blockchains | |
CN116401718A (en) | Block chain-based data protection method and device, electronic equipment and storage medium |
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 |