Efficient revocable data security access control method and device based on zero trust, processor and computer readable storage medium thereof

Abstract

The application relates to a zero-trust-based efficient revocable data security access control method, wherein the method comprises the following steps: a data owner transmits an encrypted data file to a cloud server; a proxy server sends a trust value measurement request to an access control center when receiving an access request sent by a data user; when the trust measurement results of the user behavior, the cloud server and the authorization center are all normal, the authorization center generates a private key and a proxy key; the cloud server decrypts part of the ciphertext through the proxy key and sends the part of the ciphertext to the data user; and the data user decrypts the part of the ciphertext through the private key to obtain plaintext. The application also relates to a corresponding device, a processor and a storage medium thereof. The beneficial effect is that the zero-trust idea can be used to solve the problem of untrusted third parties, thereby achieving the purpose of efficient data security access.

Description

Technical Field

[0001] This invention relates to the field of information security technology, and more particularly to the field of data security technology, specifically to a zero-trust-based, highly efficient, revocable data security access control method, apparatus, processor, and computer-readable storage medium thereof. Background Technology

[0002] Access control is an effective method for ensuring secure data access, preventing unauthorized users from accessing cloud data. Only users with the necessary permissions can access the corresponding data, thus providing security and privacy protection for cloud data. Currently existing access control methods mainly include the following:

[0003] 1) Discretionary Access Control (DAC). DAC allows data owners to manage their own data and determine access permissions for other users through established rules. This access control mechanism is highly flexible, allowing data owners to manage their data according to their own wishes.

[0004] 2) Mandatory Access Control (MAC). MAC assigns a fixed security attribute to users and files, such as security level and access permissions. When a user accesses a data file, the system checks this security attribute to determine if the user has the necessary permissions. MAC offers better security protection than discretionary access control, but its strictness limits its flexibility.

[0005] 3) Access control based on attribute-based encryption (ABE). This includes the Key-Policy Attribute-Based Encryption (KP-ABE) scheme and the Ciphertext-Policy Attribute-Based Encryption (CP-ABE) scheme. ABE is a public-key encryption mechanism that uses attributes as key elements, enabling one-to-many encryption and decryption. Unlike identity encryption, which requires knowledge of the accessor's identity information for each decryption, in ABE, identity is treated as a series of attributes. A user can decrypt when their attributes meet the encryptor's preset threshold. It targets a group of users for decryption, rather than individual users.

[0006] Optional and mandatory access control are traditional access control methods, characterized by inflexibility in binding subjects and specific objects, making them unsuitable for complex user scenarios like cloud environments. Attribute encryption access control is suitable for multi-user backgrounds in cloud environments and can improve data access efficiency. However, traditional attribute encryption has two problems: First, attribute encryption relies on a trusted third party. In a real cloud environment, cloud service providers are considered semi-trusted and may act beyond their authorized scope for profit. Root Authorization Authority (CA) is trusted, but regional Authorization Authorities (CAs) may seek to obtain user information outside their managed regions for profit. Furthermore, cloud servers are semi-trusted, and they may collude to leak user privacy. Additionally, users may conspire to obtain unauthorized data files for high profits. Second, attribute encryption is computationally intensive. To reduce the computational burden on data users and improve system access efficiency, existing solutions outsource complex calculations during decryption to cloud service providers, leaving data users with only exponential operations. However, it cannot solve the problem of large computational load for data owners; third, existing attribute encryption schemes do not support efficient revocation, and when revocation occurs, the system's computational overhead is large, and a large number of key updates also bring a huge management burden to the authorization center; fourth, some new application scenarios may emerge in the cloud environment, such as the introduction of zero trust to monitor user behavior in real time to detect anomalies or the data owner wanting to change the original access policy, which will involve the problem of dynamic access policy updates. The expressiveness of access control policies in existing technical solutions is not rich enough and does not support fine-grained dynamic access policy updates. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a zero-trust-based, efficient, revocable data security access control method, apparatus, processor, and computer-readable storage medium that aims to improve data access efficiency and ensure security during the data access process.

[0008] To address the aforementioned problem one and resolve the issue of untrusted third parties, this technical solution introduces the "zero trust" security concept to construct a data security access control framework based on attribute encryption in the medical cloud. It adopts a "never trust and always verify" stance, periodically and continuously monitoring the behavior of the user and server on the network, and constructing an access control model based on behavioral trust metrics to solve the problem of untrusted third parties in previous attribute encryption schemes.

[0009] To address issue two, this technical solution continues to utilize the outsourcing approach to further optimize computational efficiency by outsourcing some of the encrypted computations to cloud servers, thereby reducing the computational burden on data owners.

[0010] Regarding question three, because outsourced encryption and decryption technology is used, when data user attribute revocation occurs, there is no need to perform complex update calculations. It is only necessary to delete the proxy key stored on the cloud server by the revoked user. During the attribute revocation stage, most of the update and re-encryption calculations are outsourced to the cloud server, and the data owner only needs to perform a small amount of calculation.

[0011] Regarding question four, this technical solution supports dynamic fine-grained access policy updates and proxy re-encryption. When a data owner wants to change access permissions for users, they can do so by modifying the access policy. Update operations include adding, deleting, and modifying data. When abnormal user behavior is detected, and the data owner does not want the user to access their data files, they can achieve secure access to the data files by changing the user's access policy. To reduce the computational burden on the data owner, update operations can be implemented through a cloud server. When a policy update occurs, the cloud server first runs a key conversion algorithm, outputting ciphertext independent of attributes. After receiving the updated access policy from the data owner, the cloud server runs a policy update algorithm and outputs the re-encrypted ciphertext. Therefore, the main computational overhead is borne by the cloud server, significantly reducing the data owner's computational burden for re-encryption.

[0012] To achieve the above objectives, the present invention provides a zero-trust-based, highly efficient, revocable data security access control method, apparatus, processor, and computer-readable storage medium as follows:

[0013] This efficient and revocable data security access control method based on zero trust is characterized by the pre-configured data owner establishing connections with the cloud server, the authorization center, and the access control center, respectively. The access control center establishes a connection with the proxy server. The method includes the following steps:

[0014] (1) The data owner encrypts the data file to be accessed and uploads the encrypted data file to the cloud server for storage;

[0015] (2) The data user sends an access request for the required data file to the proxy server;

[0016] (3) The proxy server sends a data user's user behavior, cloud server and authorization center trust value measurement request to the access control center according to the access request.

[0017] (4) The access control center performs trust value calculations according to the trust value measurement request to obtain the corresponding trust measurement results. If all the trust measurement results are not abnormal, then step (5) is executed. If only the trust measurement result of the user behavior is abnormal, then step (6) is executed. Otherwise, step (7) is executed.

[0018] (5) The authorization center generates the private key required by the data user for decryption and the proxy key required by the cloud server and sends them to the data user and the cloud server respectively. The cloud server decrypts the data file according to the proxy key to obtain partially decrypted ciphertext and sends it to the data user. Then, the data user decrypts the partially decrypted ciphertext according to the private key to form plaintext, thereby realizing secure access to the data file.

[0019] (6) The data owner sets access restrictions on the data user;

[0020] (7) The proxy server immediately terminates the data user's access.

[0021] Preferably, step (1) specifically includes the following steps:

[0022] (1.1) The authorization center sends the first public key information to the cloud server and the first public key information and the second public key information to the data owner;

[0023] (1.2) The cloud server generates attribute encrypted ciphertext based on the first public key information and sends it to the data owner;

[0024] (1.3) The data owner encrypts the data file to be accessed based on the first public key information, the second public key information and the attribute encryption ciphertext to obtain the encrypted data file, and uploads it to the cloud server for storage.

[0025] Preferably, the authorization center includes a root authorization center CA and multiple attribute authorization centers AA connected to the root authorization center CA. Then, step (1.1) specifically includes the following steps:

[0026] (1.1.1) The root authorization center (CA) performs global data initialization by inputting a system security parameter λ and an attribute domain U, and outputs a global public parameter GP, a user identity identifier uid, and an authorization center identifier aid, and sends the authorization center identifier aid to each of the authorization centers respectively:

[0027] (Global Setup(λ,U)→{GP,uid,aid})

[0028] (1.1.2) Each of the attribute authorization centers AA outputs a public key pair PK by inputting the authorization center identifier aid. aid and private key SK aid And generate an attribute public key for each attribute managed by the attribute authorization center AA.

[0029]

[0030] (1.1.3) The authorization center sends the global public parameters and each of the attribute public keys as the first public key information to the cloud server;

[0031] In step (1.2), the cloud server inputs the global public parameter GP and the attribute public key. And output the attribute-encrypted ciphertext CT out :

[0032]

[0033] (1.1.4) The authorization center sends the authorization center identifier and the public key of each of the attribute authorization centers AA as the second public key information to the data owner;

[0034] In step (1.3), the data owner inputs the global public parameter GP and the public key PK of the relevant attribute authorization center AA. aid The attribute public key The attribute is encrypted ciphertext CT out Given plaintext M and access structure (A, ρ), output the encrypted data file CT.

[0035]

[0036] Preferably, the access control center includes an SDN controller, which is connected to the Trust Metrics Proxy (TMP) and the Access Decision Proxy (ADA). The Trust Metrics Proxy (TMP) is connected to the Trust Metrics Database (TMD), which stores real-time monitoring data of the behavior of each data user, as well as the measurement data of the cloud server and the authorization center.

[0037] The trust value calculation in step (4) specifically includes the following steps:

[0038] (4.1) The SDN controller forwards the trust value measurement request to the behavioral trust measurement agent TMP;

[0039] (4.2) The Trust Measurement Agent (TMP) submits a request for behavioral data related to the current access request to the Trust Measurement Database (TMD);

[0040] (4.3) The Trust Measurement Database (TMD) queries the behavioral data of the data user who made the access request, the measurement data of the cloud server and the authorization center, and returns it to the Trust Measurement Agent (TMP).

[0041] (4.4) The Trust Measurement Agent (TMP) will simultaneously send the Trust Measurement result obtained by processing the behavioral data and the measurement data to the SDN controller and the Access Decision Agent (ADA).

[0042] (4.5) The access decision agent ADA returns the decision result to the SDN controller;

[0043] (4.6) The SDN controller obtains the user behavior, the authorization center and the final trust measurement result of the cloud server based on the decision result.

[0044] Preferably, step (5) specifically includes the following steps:

[0045] (5.1) Each attribute authorization center AA inputs the global public parameter GP, the user identity identifier uid, and a set of user attributes S. uid，aid Attribute Authorization Center AA's private key SK uid and attribute public key And output the proxy key P x K uid，aid and user private key SK uid :

[0046]

[0047] (5.2) The attribute authorization center AA will output the private key SK. uid Send to the data user, and send the aforementioned proxy key P x K uid，aid Send to the cloud server;

[0048] (5.3) The cloud server inputs the global public parameter GP, the complete encrypted text CT, and the data user's proxy key P. x K uid，aid and attribute public key The decrypted ciphertext CT′ in the output section:

[0049]

[0050] (5.4) The data user input ciphertext CT, the partially decrypted ciphertext CT′, and the user private key SK uid Output plaintext M:

[0051] (Decrypt_user(CT, CT′, SK) uid →M).

[0052] Preferably, in step (6), the data owner setting access restrictions on the data user includes the data owner revoking the data user's attributes or updating the data user's access policy.

[0053] Preferably, step (6) specifically includes the following steps:

[0054] (6.1) If the trust measurement result of the user behavior indicates an attack, the cloud server needs to delete the corresponding proxy key P of the data user. x K uid，aid The operation of revoking the data user attributes is completed; if the trust measurement result of the user behavior does not indicate an attack but only a change in the trust measurement value, then steps (6.2) and (6.3) are executed to complete the operation of updating the data user access policy.

[0055] (6.2) When an access policy update event occurs, the cloud server first performs a Transform to output ciphertext without attribute information.

[0056] (6.3) The data owner inputs a new access structure (A, ρ). * , Public key PK, Key PK for each attribute x And partially decrypt the ciphertext C, and finally output the new encrypted ciphertext.

[0057] This zero-trust-based, highly efficient, revocable data security access control device is characterized by the following features:

[0058] A processor is configured to execute computer-executable instructions;

[0059] The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the aforementioned zero-trust-based, efficient, revocable data security access control method.

[0060] The processor for efficient and revocable data security access control based on zero trust is characterized in that the processor is configured to execute computer-executable instructions, which, when executed by the processor, implement the steps of the efficient and revocable data security access control method based on zero trust described above.

[0061] The computer-readable storage medium is characterized in that it stores a computer program that can be executed by a processor to implement the steps of the above-described zero-trust-based efficient revocable data security access control method.

[0062] The beneficial effects are as follows: This invention employs a more granular and secure access control mechanism that combines the zero-trust concept with attribute encryption. Attribute encryption ensures data security during transmission. Furthermore, by real-time monitoring and trust measurement of the behavior of data users, cloud servers, and authorization centers, the trust level of untrusted users is determined, and data user resource access permissions can be dynamically adjusted based on the trust measurement results. When the trust level drops to untrusted, further access to data files by the data user is blocked, thus solving the problem of data leakage caused by abnormal user behavior. Compared with existing attribute encryption technologies, this technology not only significantly reduces the computational burden on data users but also further reduces the computational burden on data owners by adopting the concept of outsourced encryption. In addition, multiple authorization centers are used to avoid single points of failure and complex key distribution issues. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the processing structure of the efficient revocable data security access control method based on zero trust according to the present invention.

[0064] Figure 2 This is a flowchart of the efficient and revocable data security access control method based on zero trust according to the present invention.

[0065] Figure 3 This is a flowchart illustrating the process by which the data owner encrypts the data file to be accessed, as described in this invention.

[0066] Figure 4 This is a flowchart illustrating the trust value calculation process of this invention.

[0067] Figure 5 This is a schematic diagram comparing the user encryption time of the present invention with that of existing technologies under simulation conditions.

[0068] Figure 6 This is a schematic diagram comparing the attribute cancellation time of the present invention with that of the prior art under simulation conditions.

[0069] Figure 7 This is a schematic diagram comparing the user re-encryption time of the present invention with that of the prior art under simulation conditions. Detailed Implementation

[0070] To more clearly describe the technical content of the present invention, the following description is provided in conjunction with specific embodiments.

[0071] Before describing the embodiments of the present invention in detail, it should be noted that, in the following, the terms “comprising,” “including,” or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0072] Please see Figure 1 and Figure 2 As shown, this efficient and revocable data security access control method based on zero trust establishes connections between pre-configured data owners and the cloud server, authorization center, and access control center, respectively. The access control center establishes a connection with the proxy server. The authorization center includes a root authorization center (CA) and multiple attribute authorization centers (AA) connected to the root authorization center CA. The access control center includes an SDN controller, which is connected to the trust measurement agent (TMP) and the access decision agent (ADA). The trust measurement agent (TMP) is connected to the trust measurement database (TMD), which stores real-time monitored behavioral data of each data user, as well as measurement data from the cloud server and the authorization center.

[0073] The method includes the following steps:

[0074] (1) The data owner encrypts the data file to be accessed and uploads the encrypted data file to the cloud server for storage;

[0075] (2) The data user sends an access request for the required data file to the proxy server;

[0076] (3) The proxy server sends a request to the SDN controller to measure the user's behavior, cloud server and authorization center trust value based on the access request.

[0077] (4) The SDN controller performs trust value calculations based on the trust value measurement request to obtain the trust measurement results of the user behavior of the data user, the root authorization center CA and the cloud server. If the trust measurement results of the three are not abnormal, then step (5) is executed. If only the trust measurement result of the user behavior of the data user is abnormal, then step (6) is executed. Otherwise, step (7) is executed.

[0078] (5) The Attribute Authorization Center (AA) sends the private key required for data user decryption and the proxy key required for cloud server to data user and cloud server respectively. The cloud server obtains part of the decrypted ciphertext through outsourced decryption algorithm and sends it to data user. Data user decrypts the corresponding ciphertext according to private key to form plaintext, thereby realizing secure access to data files.

[0079] (6) Data owners set access restrictions on data users;

[0080] (7) The SDN controller notifies the agent server to immediately terminate the data user's access.

[0081] As a preferred embodiment of the present invention, such as Figure 3 As shown,

[0082] (1.1) The authorization center sends the first public key information to the cloud server and the first public key information and the second public key information to the data owner;

[0083] (1.2) The cloud server generates attribute encrypted ciphertext based on the first public key information and sends it to the data owner;

[0084] (1.3) The data owner encrypts the data file to be accessed based on the first public key information, the second public key information and the attribute encryption ciphertext, and uploads the encrypted data file to the cloud server for storage.

[0085] As a preferred embodiment of the present invention, such as Figure 4 As shown, step (1.1) specifically includes the following steps:

[0086] (1.1.1) The root authorization center (CA) performs global data initialization. By inputting a system security parameter λ and an attribute domain U, it outputs the global common parameter GP, the user identity identifier uid, and the authorization center identifier aid, and sends the authorization center identifier aid to each authorization center respectively:

[0087] (Global Setup(λ,U)→{GP,uid,aid})

[0088] (1.1.2) Each attribute authorization center (AA) initializes the key required to generate the encrypted data file and sends it to the cloud server. Each attribute authorization center (AA) outputs a pair of public keys (PK) by inputting the authorization center identifier (aid). aid and private key SK aid And generate a public key for each attribute managed by the AA authorization center.

[0089]

[0090] (1.1.3) The authorization center sends the global public parameters and the public keys of each attribute as the first public key information to the cloud server;

[0091] In step (1.2), the cloud server inputs the global public parameter GP and the attribute public key. It also outputs partially encrypted ciphertext CT. out That is, attribute-encrypted ciphertext:

[0092]

[0093] (1.1.4) The authorization center sends the authorization center identifier and the public key of each attribute authorization center AA as the second public key information to the data owner;

[0094] In step (1.3), the data owner encrypts the data file using a partial encrypted ciphertext. The data owner inputs the global public parameter GP and the public key PK of the relevant attribute authorization center AA. aid Attribute public key Partially encrypted ciphertext CT out Given plaintext M (the data file to be encrypted) and access structure (A, ρ), output the complete encrypted ciphertext CT (the encrypted data file):

[0095]

[0096] Please see Figure 1 and Figure 5 As shown, in a preferred embodiment of the present invention, step (4) of performing the trust value calculation specifically includes the following steps:

[0097] (4.1) The SDN controller forwards the trust value metric request to the behavioral trust metric agent TMP;

[0098] (4.2) The Trust Measurement Agent (TMP) submits a request for behavioral data related to the current access request to the Trust Measurement Database (TMD);

[0099] (4.3) The Trust Measurement Database (TMD) queries the data behavior of the requested object (i.e., the behavioral data of the data user who made the access request, the measurement data of the cloud server and the authorization center) and returns it to the Trust Measurement Agent (TMP);

[0100] (4.4) The Trust Measurement Agent (TMP) sends the trust measurement results to both the SDN controller and the Behavior Trust-Based Access Decision Agent (ADA).

[0101] (4.5) The access decision agent ADA returns the decision results to the SDN controller;

[0102] (4.6) The SDN controller obtains the data user behavior, root authorization center (CA), and cloud server's final trust measurement results based on the decision results.

[0103] As a preferred embodiment of the present invention, such as Figure 6 As shown, step (5) specifically includes the following steps:

[0104] (5.1) Each attribute authorization center AA inputs the global public parameter GP, the user identity identifier uid, and a set of user attributes S. uid，aid Attribute Authorization Center AA's private key SK uid and attribute public key And output the proxy key P x K uid，aid and user private key SK uid :

[0105]

[0106] (5.2) The Attribute Authorization Center (AA) will output the private key SK. uid Send to data users, including the proxy key P x K uid，aid Send to the cloud server;

[0107] (5.3) The cloud server inputs the global public parameter GP, the complete encrypted text CT, and the data user's proxy key P. x K uid，aid and attribute public key The decrypted ciphertext CT′ in the output section:

[0108]

[0109] (5.4) Data: User input ciphertext CT, partially decrypted ciphertext CT′, and user private key SK uid Output plaintext M:

[0110] (Decrypt-user(CT, CT′, SK) uid →M).

[0111] As a preferred embodiment of the present invention, such as Figure 7 As shown, step (6) specifically includes the following steps:

[0112] (6.1) If the trust measurement result of a data user's user behavior indicates that they have been attacked, the cloud server needs to delete the proxy key P corresponding to that data user. x K uid，aid Complete the operation of revoking data user attributes; if the trust measurement result of user behavior does not indicate an attack but only a change in trust measurement value, then execute steps (6.2) and (6.3) to complete the operation of updating data user access policy;

[0113] (6.2) When an access policy update event occurs, the cloud server first performs a Transform to output ciphertext without attribute information.

[0114] (6.3) The data owner inputs a new access structure (A, ρ). * , Public key PK, Key PK for each attribute x And partially decrypt the ciphertext C, and finally output the new encrypted ciphertext.

[0115] To address the shortcomings of existing technologies, the technical solutions employed in this paper employ the following specific technical means:

[0116] First, in traditional attribute-based encryption schemes, cloud servers are considered semi-trusted entities. On the one hand, they execute user commands according to regulations; on the other hand, they may engage in unauthorized activities for high profits, such as leaking user privacy. The root authority (CA) is a trusted center, but each attribute authority (AA) may try to obtain user information outside its managed area for profit. Furthermore, since cloud servers are semi-trusted, they may collude to leak user privacy. Additionally, users may collude to obtain data files beyond their authorized access, inevitably leading to data privacy leaks. Zero Trust acknowledges the shortcomings of traditional perimeter security architectures, arguing that all hosts, regardless of their location on the network, should be considered internet hosts, and the network they reside on, whether the internet or an internal network, must be considered a dangerous network full of threats. A key characteristic of the Zero Trust concept is default distrust, always employing verification and least privilege. These features significantly reduce the risk of attacks on cloud service providers and middleware servers. This invention draws on the zero-trust security concept, adopting a "never trust and always verify" approach to periodically and continuously monitor the behavior of the user and server on the network. It constructs an access control model based on behavioral trust metrics to solve the problem of untrusted third parties in previous attribute encryption schemes.

[0117] Secondly, this technical solution employs fully outsourced encryption and decryption technology to simultaneously address the issue of high computational load for both the encryptor and decryptor. First, the cloud server performs calculations on the generated items corresponding to the user attributes during the encryption process, and then transmits the resulting partial ciphertext to the data owner. Specifically, the generated items corresponding to the aforementioned user attributes are typically presented as an attribute set containing multiple user attribute data, making the encryption process computationally complex and time-consuming. To alleviate the computational burden on the data owner and improve efficiency, this technical solution outsources the computationally intensive and time-consuming encryption process to the cloud server. Furthermore, because the encrypted plaintext remains with the data owner throughout the encryption process and is not intercepted by the cloud server, there is no risk of data privacy leakage. During decryption, the cloud server first generates a conversion key and sends it to the user's proxy server. The proxy server performs partial complex attribute decryption and then transmits the partial decrypted ciphertext to the data user. The data user can then perform simple decryption calculations on their own server to retrieve the plaintext. This idea of ​​outsourcing attribute calculations in both the encryption and decryption processes greatly reduces the complexity of attribute encryption and decryption for data owners and users, and also prevents the leakage of plaintext, thus protecting data privacy.

[0118] Third, this technical solution supports efficient attribute revocation. Because it adopts outsourced encryption and decryption technology, the proxy keys related to user attributes are stored in the cloud service. When data user attribute revocation occurs, there is no need to perform complex update calculations. It is only necessary to delete the proxy key of the revoked data user stored on the cloud server. During the attribute revocation stage, most of the update and re-encryption calculations are outsourced to the cloud server. The data owner only needs to complete a small amount of calculation, thus achieving efficient attribute revocation.

[0119] Fourth, compared to the previously proposed technical solutions, which failed to analyze user behavior or identify users with suspicious identities, this solution addresses the vulnerability of hackers who, with the increasing interconnectivity of global networks, constantly evolve their methods to launch malicious attacks by mimicking normal user behavior. If a user's client is compromised and their private key is obtained, they can successfully decrypt data files, leading to privacy breaches. This technical solution introduces the concept of zero trust. Based on the access control center's measurement of the trust level of untrusted users, when the trust level drops to untrustworthy, the access policy can be dynamically updated to resolve data leaks caused by abnormal user behavior. The specific solution is as follows: The dynamic policy update idea in this solution is: when user behavior is abnormal, the data owner needs to update the access policy for user behavior, including deletion and updates. To simplify the computational burden, the data owner does not need to re-encrypt and upload files. Instead, based on the trust measurement results from the access control center, the cloud server is notified to directly update the access policy for the data files. This part of the work is handled by the cloud server, thereby reducing the burden on the data owner. Specifically, the data owner only needs to upload the new access policy to the cloud server, which greatly reduces the data owner's computational burden. We employ an outsourced decryption approach: first, the Transform algorithm is called to convert the ciphertext into simple, attribute-independent ciphertext; then, the ciphertext is re-encrypted using a new access policy. During this process, the data owner only needs to define the new access policy to complete the entire policy update. The specific algorithm is as follows: When an access policy update event occurs, the cloud server first executes Transform to output ciphertext without attribute information. When the cloud server receives a new access policy uploaded by the data owner, it executes the PolicyUpdate algorithm to obtain the updated ciphertext. This algorithm is similar to a re-encryption algorithm, except that the third part of the ciphertext and the attribute-related permutations are replaced. In this way, the cloud server bears the main computational load, greatly reducing the burden on users.

[0120] like Figure 1 As shown, the entities involved in this technical solution and their specific functions are described below:

[0121] 1) The root authorization center (CA) is responsible for managing the various attribute authorization centers (AA), such as hospitals, insurance centers, and some research centers. Each attribute authorization center (AA1 to AAk) generates and manages the keys of users in that area, which simplifies the complexity of key distribution and management.

[0122] 2) The data owner is responsible for encrypting the data file to be encrypted and uploading it to the cloud server.

[0123] 3) The data user sends an access request for the data file that needs to be encrypted to the proxy server.

[0124] 4) The cloud server is responsible for securely storing the data files to be encrypted and performing partial encryption and decryption operations on some of the data files according to the instructions.

[0125] 5) The proxy server is responsible for part of the decryption work and performs related operations according to the instructions.

[0126] 6) Description of Access Control Center:

[0127] 6a) Behavioral Trust Measurement Agent (hereinafter referred to as TMP): A device that measures user behavioral trust. It deploys behavioral measurement middleware, mainly to identify network attack behaviors with long latency and high concealment in the internal network, such as APT and DNS attacks. It periodically measures the behavioral trust level through measurement data monitored in real time in the trust measurement database.

[0128] 6b) Behavior Trust-Based Access Decision Agent (ADA): The device that distributes access policies based on user trust levels has deployed access decision middleware. It can design corresponding decision functions for different trust levels according to different security strength requirements. After making a decision, it sends the policy data packet to the SDN controller for distribution.

[0129] 6c) Trust Measurement Database (hereinafter referred to as TMD): Stores historical behavioral measurement data of the measured object. When the object logs into the network, the behavioral measurement agent will evaluate the object's initial trust level based on the historical behavioral measurement data in the database.

[0130] 6d) SDN controller: It dynamically adjusts the access permissions of the client and the server by issuing flow tables, which follows the principle of continuous monitoring and verification of zero-trust network.

[0131] In practical applications, please refer to Figure 2 As shown, the implementation steps of the zero-trust-based efficient revocable data security access control method of the present invention are as follows:

[0132] Step 1: The data owner encrypts the data file to be encrypted and then uploads the encrypted data file to the cloud server.

[0133] 1.1) The root authorization center (CA) performs global data initialization, taking a system security parameter λ and an attribute domain U as input, and outputting the global public parameter GP, the user identity identifier uid, and the authorization center identifier aid.

[0134] (Global Setup(λ,U)→{GP,uid,aid})

[0135] 1.2) The Attribute Authorization Center (AA) initializes and generates the key required for encrypting the file and sends it to the cloud server. Each Attribute Authorization Center (AA) inputs its Authorization Center Identifier (AID) and outputs a pair of public keys (PK) for the Attribute Authorization Center (AA). aid and private key SK aid And generate a public key for each attribute managed by the AA (Attribute Authorization Center).

[0136]

[0137] 1.3) Input the global public parameter GP and attribute public key into the cloud server. The output section contains encrypted ciphertext CT. out .

[0138]

[0139] 1.4) The data owner encrypts the data file to be encrypted using a partially encrypted ciphertext. The data owner inputs the global public parameter GP and the public key PK of the relevant attribute authorization center AA. aid Attribute public key Partially encrypted ciphertext CT out Given plaintext M and access structure (A, ρ), output the encrypted complete ciphertext CT;

[0140]

[0141] 1.5) The data owner will upload the finally generated encrypted file to the cloud server for storage.

[0142] Step 2: The data user submits a request to the proxy server to access the required encrypted data file.

[0143] Step 3: The proxy server sends a trust value measurement request to the SDN controller for data users, cloud servers, and each authorization center. The SDN controller calculates the final trust measurement result through the behavior measurement module and returns it to the proxy server and the root authorization center (CA). If all requirements are met, proceed to the next step. If only user behavior is abnormal, go to step 5. In any other case, terminate access.

[0144] 3.1) The proxy server sends trust value measurement requests to the SDN controller for data users, cloud servers, and various authorization centers;

[0145] Please refer to Figure 1 and Figure 3 The specific implementation of the trust value measurement operation is as follows:

[0146] 3.1a) The SDN controller forwards the metric request to the TMP;

[0147] 3.1b) TMP submits a request for behavioral data of the application to TMD;

[0148] 3.1c) The TMD queries the requested object's data behavior and returns it to the TMP;

[0149] 3.1d) TMP sends the measurement results to both the SDN controller and ADA simultaneously;

[0150] 3.1e) The ADA returns the decision results to the SDN controller;

[0151] 3.2) The SDN controller returns the final trust measurement result to the proxy server and CA;

[0152] 3.3) If the returned result is normal, proceed to step 4;

[0153] 3.4) If there are only abnormal user behaviors and they are not attack exceptions, then proceed to step 5;

[0154] 3.5) Access will be terminated in any other situation where an attack is detected.

[0155] Step 4: The Attribute Authorization Center (AA) generates the private key required for data user decryption and the proxy key required by the cloud server. The cloud server obtains the partially decrypted ciphertext through an outsourced decryption algorithm and sends it to the data user. The data user then decrypts the encrypted file to form plaintext.

[0156] 4.1) The attribute authorization center AA inputs the global public parameter GP, the user identity identifier uid, and a set of user attributes S. uid，aid SK, the private key of the authorization center uid and attribute public key Output proxy key P x K uid，aid and user private key SK uid .

[0157]

[0158] 4.2) The Attribute Authorization Center AA will transfer the private key SK uid Send to data users, including the proxy key P x K uid，aid Send it to the cloud server.

[0159] 4.3) Input the global public parameter GP, encrypted CT, and user's proxy key P into the cloud server. x K uid，aid and attribute public key The output section contains the decrypted ciphertext CT′.

[0160]

[0161] 4.4) Data: User input ciphertext CT, partially decrypted ciphertext CT′, and user private key SK uid Output plaintext M.

[0162] (Decrypt_user(CT, CT′, SK) uid →M)

[0163] Step 5: Based on the results of abnormal user behavior, the data owner decides to revoke the data user's attributes or update the data user's access policy. The file is then re-encrypted using the new access policy and uploaded again.

[0164] 5.1) When an access policy update event occurs, the cloud server first executes Transform to output encrypted data without attribute information.

[0165] 5.2) The data owner inputs a new access structure (A, ρ). * Public key PK, key PK for each attribute x There is also a portion of the decrypted ciphertext C, and then a new encrypted ciphertext is output.

[0166] Compared with existing technologies, this technical solution has the following technical advantages:

[0167] (1) Complexity Analysis

[0168] To analyze the complexity of this technical solution, we compared it with existing technical solutions. The relevant symbols and their corresponding meanings are shown in Table 1. The calculation results for each solution are shown in Table 2.

[0169] Table 1. Symbols and their corresponding meanings

[0170]

[0171]

[0172] Table 2 Calculation Costs

[0173] plan User encryption time Attribute Revocation Time Re-encryption time Existing technical solutions <![CDATA[lT p +(3n+l+1)T m ]]> <![CDATA[2lT P +(n+1)T m ]]> <![CDATA[(3n+l+1)T m ]]> This technical solution <![CDATA[T P +T m ]]> <![CDATA[T m ]]> <![CDATA[T m ]]>

[0174] We will analyze the efficiency of this technical solution from the following aspects: First, regarding the encryption time performed by the data owner, in our technical solution, the most complex operations in the encryption process are outsourced to the CSP. The data owner only needs to complete a simple public key encryption, exponentiation operation, and parameter correction operation. These calculations are unrelated to attributes, so the encryption time is a constant value, greatly reducing the computational overhead for both the data owner and the user. Second, this technical solution supports effective attribute revocation without updating the user's private key. In existing technical solutions, each revocation event requires the data owner to recalculate and send a new ciphertext component corresponding to the revoked attribute to all remaining users. In our technical solution, when attribute revocation occurs, only the key stored in the cloud server needs to be updated; users do not need to update their keys, and the revocation time is also a constant value. Third, in this technical solution, when a data user's access policy changes, the data owner does not need to re-encrypt the file upload. Instead, based on the trust measurement result of the access control center, the cloud server is notified to directly update the data file's access policy. This part of the work is handled by the cloud server, thereby reducing the burden on the data owner. The data owner only needs to upload the new access policy to the cloud server. This greatly reduces the computational burden on data owners.

[0175] Please see Figures 5 to 7 As shown, to more effectively illustrate the data access efficiency of the technical solution mentioned in this invention, we compared this technical solution with existing technical solutions in three dimensions: user encryption time, attribute revocation time, and re-encryption time, and provided simulation diagrams:

[0176] The simulation environment is a Lenovo laptop with an Intel(R) Core(TM) i5-3210M CPU @ 2.50GHz and 4.0GB RAM. The simulation is performed using C language on a Linux environment with Windows 7 as the operating system. Ubuntu 12.04 is installed on VMware Workstation 10.0 within Windows 7. It is assumed that 100 attributes are managed by 10 licensees, with each licensee managing an average of 10 attributes.

[0177] (2) Security Analysis

[0178] As shown in Table 3 below, this technical solution has undergone a security analysis from the following aspects:

[0179] First, due to the uncertainty of user identities and the large user base, this technical solution adopts an outsourced, multi-authorization-center attribute-based encryption scheme. Currently, most existing attribute encryption schemes use a single authorization center, and all key generation and distribution are handled by this center. Therefore, if an authorization center experiences information leakage, user privacy becomes a serious vulnerability. To protect user privacy and improve system security, we introduce the concept of multiple authorization centers into the solution. This multi-authorization-center approach categorizes and processes massive amounts of attributes. Each user's private key is a collection distributed by various independent authorization centers, such as hospitals, insurance companies, or research institutions. These institutions can generate corresponding private keys based on the user's attributes and send them to the user. The set of attributes possessed by the user is the set of attributes jointly distributed by all institutions. This avoids privacy leaks and single points of failure caused by a single authorization center managing all users' private keys.

[0180] Secondly, in existing attribute encryption schemes, cloud servers are considered semi-trusted entities. On the one hand, they execute user commands according to regulations; on the other hand, they may perform actions beyond the regulations for high profits, such as leaking user privacy. The root authority (CA) is a trusted entity, but each attribute authority (AA) may try to obtain user information outside its managed area for profit. Furthermore, since cloud servers are semi-trusted, they may collude with each other to leak user privacy. In addition, users may also collude to obtain data files beyond their permissions, inevitably leading to data privacy leaks. This technical solution introduces a zero-trust security concept, using an access control center to monitor cloud servers in real time. If abnormal behavior measurement results are returned, communication is immediately blocked to ensure the security of the data owner's PHR data.

[0181] Third, compared with the previously proposed technical solutions, which did not analyze user behavior or identify users with suspicious identities, this solution addresses the vulnerability of users to malicious attacks by mimicking normal user behavior and constantly changing their methods, given the interconnectedness of global networks. If a user's private key is compromised, they can successfully decrypt data files, leading to privacy breaches. This technical solution introduces the concept of zero trust, using an access control center to measure the trust level of untrusted users and dynamically adjust their access permissions based on the measurement results. When a user's trust level drops to untrustworthy or their behavior is considered malicious, access to data files is blocked, thus resolving data leaks caused by abnormal user behavior.

[0182] Table 3 Security Analysis

[0183]

[0184]

[0185] (1) Compared with existing technologies, this technical solution can not only greatly reduce the computational overhead of data users, but also further reduce the computational overhead of data owners by adopting the idea of ​​outsourced encryption. In addition, multiple authorization centers are used to avoid single points of failure and complex key distribution problems.

[0186] (2) This technical solution designs corresponding algorithms for user and attribute revocation in the public domain, supporting efficient user attribute revocation. When user attribute revocation occurs, there is no need to perform complex update operations; it is only necessary to delete the proxy key of the revoked user stored in the cloud server. For attribute revocation, most of the update and re-encryption calculations are outsourced to the cloud server, and the data owner only needs to perform a small amount of computation, thus achieving efficient attribute revocation.

[0187] (3) This technical solution adopts a more granular and secure access control mechanism that combines the zero trust concept and attribute encryption in the cloud environment. Attribute encryption ensures the security of data during transmission. In addition, by monitoring and measuring the behavior of the user and the server in real time, the access control center measures the trust level of untrusted users and can dynamically adjust the data user's resource access permissions based on the measurement results. When the trust level drops to untrusted or the user's behavior is attacked, the data user is blocked from continuing to access the data file, thereby solving the problem of data leakage caused by abnormal user behavior.

[0188] (4) This technical solution addresses the problem of untrusted third parties in existing attribute encryption schemes. In existing attribute encryption schemes, cloud service providers and authorization centers are considered semi-trusted institutions. On the one hand, they execute user commands according to regulations; on the other hand, they may perform actions beyond regulations for high profits, such as leaking user privacy. The root authorization center (CA) is a trusted institution, while each attribute authorization center (AA) may try to obtain user information outside its managed area for profit. Moreover, cloud servers are semi-trusted, and they may collude with each other to leak user privacy. In addition, users may also obtain data files outside their permissions through collusive attacks, which will inevitably lead to data privacy leaks. This technical solution introduces the zero-trust security concept, which monitors cloud servers and authorization centers in real time through an access control center. If the returned behavior measurement results are abnormal, communication is immediately blocked to ensure data security.

[0189] (5) This technical solution supports dynamic updates of access policies and proxy re-encryption. When abnormal data user behavior is detected, the data owner needs to update the access policy for the data user behavior, including deletion and update. In order to simplify the computational burden, the data owner does not need to re-encrypt the file upload. Instead, the cloud server is notified to update the access policy of the data file directly based on the trust measurement result of the access control center. This part of the work is handled by the cloud server. The data owner only needs to upload the new access policy to the cloud server, which greatly reduces the computational burden of the data owner.

[0190] This zero-trust-based, highly efficient, revocable data security access control device includes:

[0191] A processor is configured to execute computer-executable instructions;

[0192] The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the aforementioned zero-trust-based, efficient, revocable data security access control method.

[0193] The processor for efficient and revocable data security access control based on zero trust is configured to execute computer-executable instructions, which, when executed by the processor, implement the steps of the efficient and revocable data security access control method based on zero trust described above.

[0194] The computer-readable storage medium contains a computer program that can be executed by a processor to implement the steps of the above-described zero-trust-based efficient revocable data security access control method.

[0195] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0196] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution device.

[0197] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0198] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0199] In the description of this specification, references to terms such as "an embodiment," "some embodiments," "example," "specific example," or "embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0200] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

[0201] This invention employs a zero-trust-based, highly efficient, revocable data security access control method, apparatus, processor, and computer-readable storage medium. It utilizes a finer-grained and more secure access control mechanism combining zero-trust principles and attribute encryption, applied in a medical cloud. Attribute encryption ensures data security during transmission. Furthermore, by real-time monitoring and trust measurement of user and server behavior, the trust level of untrusted users is measured using a behavior measurement module, and data user resource access permissions are dynamically adjusted based on the measurement results. When the trust level drops to untrusted, further access to data files by the data user is blocked, thus resolving data leakage caused by abnormal user behavior. Compared to existing attribute encryption technologies, this solution significantly reduces the computational burden on data users and further reduces the computational burden on data owners by employing outsourced encryption. In addition, multiple authorization centers are used to avoid single points of failure and complex key distribution issues.

[0202] In this specification, the invention has been described with reference to specific embodiments thereof. However, it will be apparent that various modifications and variations can be made without departing from the spirit and scope of the invention. Therefore, the specification and drawings should be considered illustrative rather than restrictive.

Claims

1. A highly efficient and revocable data security access control method based on zero trust, characterized in that, The pre-configured data owner establishes connections with the cloud server, the authorization center, and the access control center, respectively. The access control center establishes a connection with the proxy server. The method includes the following steps: (1) The data owner encrypts the data file to be accessed and uploads the encrypted data file to the cloud server for storage; (2) The data user sends an access request for the required data file to the proxy server; (3) The proxy server sends the user behavior of the data user, the trust value measurement request of the cloud server and the authorization center to the access control center according to the access request. (4) The access control center performs trust value calculations according to the trust value measurement request to obtain the corresponding trust measurement results. If all the trust measurement results are not abnormal, then step (5) is executed. If only the trust measurement result of the user behavior is abnormal, then step (6) is executed. Otherwise, step (7) is executed. (5) The authorization center generates the private key required by the data user and the proxy key required by the cloud server and sends them to the data user and the cloud server respectively. The cloud server decrypts the data file according to the proxy key to obtain partially decrypted ciphertext and sends it to the data user. Then, the data user decrypts the partially decrypted ciphertext according to the private key to form plaintext, thereby realizing secure access to the data file. (6) The data owner sets access restrictions on the data user; (7) The proxy server immediately terminates the data user's access.

2. The efficient and revocable data security access control method based on zero trust according to claim 1, characterized in that, Step (1) specifically includes the following steps: (1.1) The authorization center sends the first public key information to the cloud server and the first public key information and the second public key information to the data owner; (1.2) The cloud server generates attribute encrypted ciphertext based on the first public key information and sends it to the data owner; (1.3) The data owner encrypts the data file to be accessed based on the first public key information, the second public key information and the attribute encryption ciphertext to obtain the encrypted data file, and uploads it to the cloud server for storage.

3. The efficient and revocable data security access control method based on zero trust according to claim 2, characterized in that, The authorization center includes a root authorization center CA and multiple attribute authorization centers AA connected to the root authorization center CA. Therefore, step (1.1) specifically includes the following steps: (1.1.1) The root authorization center (CA) performs global data initialization by inputting a system security parameter. and attribute domain Output global common parameters User identification and Authorization Center Logo and the authorization center identifier Send them to each of the aforementioned attribute authorization centers (AA): ； (1.1.2) Each of the attribute authorization centers AA inputs the authorization center identifier. Output a pair of public keys and private key And generate an attribute public key for each attribute managed by the attribute authorization center AA. : ； (1.1.3) The authorization center will use the global public parameters and the public keys of each attribute described The first public key information is sent to the cloud server; In step (1.2), the cloud server inputs the global public parameters. and attribute public key and output the encrypted ciphertext of the attribute, wherein, This indicates that the attribute is encrypted with ciphertext: ； (1.1.4) The authorization center will identify the authorization center. and the public key of each of the attribute authorization centers AA The second public key information is sent to the data owner; In step (1.3), the data owner inputs the global public parameters. The public key of the relevant attribute authorization center AA The attribute public key The attribute encryption ciphertext Plaintext M and access structure Output the encrypted data file CT: 。 4. The efficient and revocable data security access control method based on zero trust according to claim 1, characterized in that, The access control center includes an SDN controller, which is connected to the Trust Metrics Proxy (TMP) and the Access Decision Proxy (ADA). The Trust Metrics Proxy (TMP) is connected to the Trust Metrics Database (TMD), which stores real-time monitoring data of the behavior of each data user, as well as the measurement data of the cloud server and the authorization center. The trust value calculation in step (4) specifically includes the following steps: (4.1) The SDN controller forwards the trust value measurement request to the trust measurement agent TMP; (4.2) The Trust Measurement Agent (TMP) submits a request for behavioral data related to the current access request to the Trust Measurement Database (TMD). (4.3) The Trust Measurement Database (TMD) queries the behavioral data of the data user who made the access request, the measurement data of the cloud server and the authorization center, and returns it to the Trust Measurement Agent (TMP); (4.4) The Trust Measurement Agent (TMP) will simultaneously send the Trust Measurement result obtained by processing the behavioral data and the measurement data to the SDN controller and the Access Decision Agent (ADA). (4.5) The access decision agent ADA returns the decision result to the SDN controller; (4.6) The SDN controller obtains the user behavior, the authorization center and the final trust measurement result of the cloud server based on the decision result.

5. The efficient and revocable data security access control method based on zero trust according to claim 3, characterized in that, Step (5) specifically includes the following steps: (5.1) Each attribute authorization center (AA) inputs global common parameters. User identification A set of user attributes Attribute Authorization Center AA's private key and attribute public key and output the proxy key. and user private key : ； (5.2) The attribute authorization center AA will output the user's private key. Send the proxy key to the data user. Send to the cloud server; (5.3) The cloud server input global common parameters Complete encrypted text Proxy keys for data users and attribute public key Output the decrypted ciphertext : ； (5.4) The user inputs complete encrypted data. Partially decrypted ciphertext and user private key Output plaintext : 。 6. The efficient and revocable data security access control method based on zero trust according to claim 1, characterized in that, In step (6), the data owner setting access restrictions on the data user includes the data owner revoking the data user's attributes or updating the data user's access policy.

7. The efficient and revocable data security access control method based on zero trust according to claim 6, characterized in that, Step (6) specifically includes the following steps: (6.1) If the trust measurement result of the user behavior indicates an attack, the cloud server needs to delete the corresponding proxy key of the data user. Complete the operation of revoking the data user attributes; if the trust measurement result of the user behavior does not indicate an attack but only a change in the trust measurement value, then execute steps (6.2) and (6.3) to complete the operation of updating the data user access policy; (6.2) When an access policy update event occurs, the cloud server first performs a Transform to output ciphertext without attribute information. ; (6.3) The data owner inputs a new access structure. Public Key Key for each attribute And some decrypted ciphertext Finally, the new encrypted ciphertext is output.

8. A highly efficient and revocable data security access control device based on zero trust, characterized in that, The device includes: A processor is configured to execute computer-executable instructions; A memory that stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the zero-trust-based, efficient, revocable data security access control method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, It stores a computer program that can be executed by a processor to implement the steps of the zero-trust-based efficient revocable data security access control method as described in any one of claims 1 to 7.

Images