Data security sharing method based on bidirectional level matching and cooperative supervision

Abstract

The invention discloses a data security sharing method based on bidirectional level matching and cooperative supervision. According to the method, a bidirectional level matching access control mechanism is introduced, dynamic matching is carried out according to a data sensitivity level set by a data owner and an access level generated by a data user based on an access environment, and fine-grained control over access authority is achieved; and meanwhile, a cooperative supervision mechanism and a user-defined time token mechanism are combined, so that the flexibility and the supervision of the data authorization and use process are ensured. According to the method, multiple cryptographic algorithms are comprehensively adopted, sensitive data can be encrypted according to a predefined access strategy through the fast attribute-based encryption technology, and the encrypted data are packaged in a data capsule, so that the confidentiality and access limitation of the data in the transmission and use process are ensured. The user-defined number of times token mechanism supports controllable authorization of data access number of times, and effectively improves data sharing efficiency and flexibility.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of cryptographic algorithms and application systems, and particularly relates to a data security sharing method based on bidirectional level matching and collaborative supervision. BACKGROUND

[0002] In order to promote the compliance and efficient circulation of data elements, and support the construction of a national integrated data market, in 2024, the National Data Bureau released the "Trusted Data Space Development Action Plan (2024-2028)". The document emphasizes that the construction of personal trusted data space pilots will be explored steadily and cautiously. Through institutional and technological innovation, services for the authorized transfer and flow of personal data and the development and utilization of personal data will be provided. Data management agencies should strengthen supervision, guidance and standardization of the rational and effective use of personal data in accordance with the law.

[0003] Traditional data sharing modes do not fully consider the data sensitivity level and the security protection capability of data users in different use environments, which leads to the fact that in actual data circulation, sensitive data may be authorized to subjects with insufficient security capabilities, increasing the risk of data leakage, misuse and compliance. In addition, traditional data sharing modes usually only monitor and control at the data sharing end. Once the data is successfully shared to the data user, the original data provider cannot effectively manage and supervise the subsequent use of the data. This mode has obvious limitations because the flow, use and whether the data is reasonably used after sharing cannot be monitored in real time, which easily leads to misuse, leakage or even unauthorized secondary distribution of data. This not only increases the risk of data security, but also brings challenges to the compliance of data providers, making it difficult to ensure that the data is always controllable in the whole life cycle after sharing.

[0004] Therefore, the application uses a data capsule with a fine-grained authorization mechanism, combines a bilateral level access control mechanism, collaborative supervision and a self-defined number of tokens. It realizes dynamic access control of the sensitivity level of data and the access level of data users, supervision of the whole life cycle of the data capsule and efficient capsule access. SUMMARY

[0005] The purpose of the application is to design a data security sharing method based on bidirectional level matching and collaborative supervision to realize the safe and efficient sharing of personal data in complex scenarios.

[0006] A data security sharing method based on bidirectional level matching and collaborative supervision, the main function modules include: system initialization, personal trusted data space registration, data user registration, data sensitive level application, data capsule encapsulation, partial access task generation, access task generation, access task decryption, times token generation and data capsule download, data capsule decryption and collaborative supervision. The application mainly includes five types of roles: regulatory agencies, personal trusted data space, data users, personal data owners and blockchains.

[0007] In order to achieve the above purpose, the application is realized by the following technical scheme: bidirectional level matching mechanism and collaborative supervision mechanism. The bidirectional level matching mechanism is mainly aimed at the data sharing link, and is designed to match the sensitive level of the data with the data protection capability of the data user. Only in the case of successful level matching, the data user can obtain access permission, ensuring data security. The mechanism specifically includes steps 4, 5, 6 and 7. The collaborative supervision mechanism focuses on the whole life cycle management of data sharing, covering the data owner side and the data user side, and is committed to realizing the whole process controllable and transparent supervision effect, further strengthening the data compliance and security guarantee, specifically including steps 8, 9, 10 and 11. The specific steps are as follows:

[0008] Step 1. System initialization: when the data security sharing method for personal data self-controllable is established, the initialization parameters are constructed according to various standards, mainly in the initialization of the system encryption scheme;

[0009] Step 2. Personal trusted data space registration: the personal trusted data space initiates an application to the RA, and the RA generates a public key for the personal trusted data space according to the ID of the personal trusted data space. The personal trusted data space generates its own private key according to the public key;

[0010] Step 3. Data user registration: the data user initiates an application to the personal trusted data space, and the personal trusted data space generates a decryption key and an access level for the data user according to the ID, attribute set and data protection capability of the environment of the data user;

[0011] Step 4. Data sensitive level application: the data owner can apply for the sensitive level of the data to the personal trusted data space according to relevant regulations or evaluation standards;

[0012] Step 5. Data capsule encapsulation: the personal data owner packs its personal data into a data capsule, encapsulates the data capsule using the constructed access control structure, obtains the data capsule and uploads the capsule to the personal trusted data space;

[0013] Step 6. Partial access task generation: the personal data owner generates a partial access task with a self-defined access frequency according to the demand of the data user, and outputs the partial access task;

[0014] Step 7. Access task generation: the personal trusted data space generates a complete access task on the basis of the partial access task, according to the sensitivity level of the shared data and the access level of the data user, and outputs the complete access task;

[0015] Step 8. Access task decryption: the data user receives the access task and decrypts the semi-decryption result from the access task;

[0016] Step 9. Generation of number of times token and download of data capsule: the data user uses the semi-decryption result to send a number of times token request and a data capsule download request to the personal data space, and obtains the number of times token and the data capsule;

[0017] Step 10. Data capsule decryption: the data user uses the access task, the semi-decryption result and the number of times token to decrypt the data capsule, and outputs the data particle.

[0018] Step 11. Cooperative supervision: when disputes such as unauthorized access or data fraud occur, the RA cooperates with at least t personal trusted data spaces to supervise and solve the disputes.

[0019] The present application has the following advantages:

[0020] A data security sharing method based on bidirectional level matching and cooperative supervision is proposed, including data encapsulation into capsules, dynamic access strategy of data, and self-defined access number of times token.

[0021] (1) Data security. The data security sharing method based on bidirectional level matching and cooperative supervision adopts various advanced, safe and efficient cryptographic schemes, such as Feldman verifiable secret sharing technology and attribute-based encryption technology, to realize deep protection of the security of personal data. In steps 3 and 4, the system sets an access level for the data user and a sensitive level for the data, respectively; then in steps 6 and 7, the data sensitive level is embedded into the fine-grained access control process through twice task generation, so as to realize bidirectional matching of the user access level and the data sensitive level. Thus, dynamic access strategy and self-defined access conditions can be flexibly supported, and the risk of unauthorized users obtaining personal information in the data capsule can be significantly reduced.

[0022] (2) Data sharing autonomy and supervision throughout the life cycle. A data security sharing method based on two-way hierarchical matching and collaborative supervision is proposed based on a data capsule with a fine-grained authorization mechanism and an access task token mechanism, which realizes personal data sharing autonomy. Specifically, the access task token is used to realize selective data sharing, informed consent authorization, and partial update of the data capsule to revoke the access rights of the data user. At the same time, through the supervision information collected in steps 5, 6, 9 and 10, the system can continuously audit and supervise the entire life cycle of the data capsule.

[0023] The main functions include:

[0024] (1) Data user registration. After joining, the data user first needs to apply for a decryption key and an access level to the personal trusted data space, and needs to submit the applicant's ID DU and the applicant's attribute set ACS. The personal trusted data space generates a decryption key vsk DU and an access level AL DU for the applicant according to its access conditions and access environment. When the data user obtains the data capsule and the access task, the decryption key and the access level can be used to decrypt the data in the capsule.

[0025] (2) Data sensitivity level application. The data owner can apply for a data sensitivity level DSL to the personal trusted data space according to the data characteristics of the data to be shared. It is used for dynamic access control when sharing data.

[0026] (3) Data capsule encapsulation and access task generation. The data owner can encapsulate and form a whole data capsule by using the data capsule encapsulation algorithm proposed in the application. In addition, the access strategy is embedded in the data capsule encapsulation process to protect the data capsule, so that only the data user whose access conditions meet the specified access strategy can access the data capsule. And the sensitive level of the data is embedded in the task, only the data user whose security protection ability matches can decrypt the data capsule.

[0027] (4) Data capsule life cycle collaborative supervision. When disputes such as unauthorized access or data fraud occur, the supervisor can read the data on the blockchain after obtaining the consent of at least t personal trusted data spaces, decrypt and supervise to mediate the dispute. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 The system architecture diagram of the embodiment of the application.

[0029] Figure 2 The flowchart of the application.

[0030] Figure 3The running cost chart of each stage in the experimental analysis of the present application.

[0031] Figure 4 The storage cost chart of each stage in the experimental analysis of the present application. DETAILED DESCRIPTION

[0032] In order to make the technical means, creative features, purposes and effects of the present application easy to understand, the present application will be further described below with reference to the accompanying drawings.

[0033]

[0034]

[0035] The present application discloses a kind of data security sharing method based on two-way level matching and collaborative supervision, for supporting the safe, controllable, compliance sharing of data in multi-agent environment.Method introduces two-way level matching access control mechanism, according to the data sensitive level set by data owner and the access level generated based on access environment by data user, dynamic matching is carried out, and fine-grained control of access permission is realized;While combining collaborative supervision mechanism and self-defined number of times token mechanism, further guarantee the flexibility and supervisability of data authorization and use process.In terms of security, the present application comprehensively uses multiple cryptographic algorithms, including Feldman verifiable secret sharing (VSS) and fast attribute-based encryption (ABE) technology;Among them, fast attribute-based encryption technology can encrypt sensitive data according to predefined access policy, and encapsulate encrypted data in data capsule, to ensure its confidentiality and access restriction in transmission and use process.In addition, self-defined number of times token mechanism supports controllable authorization of data access times, effectively improves data sharing efficiency and flexibility.In the scene of use dispute or policy dispute, regulatory agency can backtrack and verify data access behavior through collaborative mechanism with trusted data space, complete compliance supervision and dispute resolution process, further enhance the auditability and credibility of system.The system architecture design of the present application is as shown in Fig. Figure 1

[0036] (1) Regulatory agency: regulatory agency is responsible for the initialization work of system, including initializing full access condition set ACU and system public parameter, completing registration for personal trusted data space, and cooperating with personal trusted data space to complete supervision when dispute occurs.

[0037] (1) Personal trusted data space: trusted agency is responsible for generating decryption key and access level according to access condition set and security protection capability of data user, authorizing data sensitive level for personal data owner, generating complete task for data user, generating number of times token for data user, and cooperating with regulatory agency to complete supervision when dispute occurs. ​

[0038] (2) Data user: After joining the system, the data user will be automatically assigned an identifier ID, and can request the trusted agency to generate a decryption key and access level according to the access condition set, ID and security protection capability of the data user. After possessing the decryption key and access level, the data user can receive and decrypt the access task, and after obtaining the number of tokens and data capsules, the data can be recovered.

[0039] (3) Personal data owner: The personal data owner can apply for a data sensitive level to the personal data space, encapsulate personal data into a data capsule, and upload the data capsule to the personal trusted data space for hosting. Then, the personal data owner issues a data user to authorize the data capsule.

[0040] (4) Personal trusted data space: The personal trusted data space is responsible for completing the registration and key generation of the data user, completing the registration and data sensitive level authorization of the data owner. At the same time, the personal trusted data space provides a large amount of space to store the data capsule of the data owner. In addition, it can generate a complete task for the data user, verify the validity of the semi-decryption result provided by the data user, and decide whether to allow the data user to access the data capsule and generate the number of tokens.

[0041] As shown in Figure 1 and 2 , a data security sharing method based on bidirectional level matching and cooperative supervision includes the following steps:

[0042] Step 1 is specifically as follows:

[0043] Phase 1: When the system is initialized, first, the regulatory agency RA constructs a bilinear group (G1, G2, G T ,e), where G1 and G2 are two p-order cyclic additive groups on an elliptic curve, G T is a p-order cyclic multiplicative group, and e is a Weil pairing: G1xG2→G T , i.e. Q∈G2 and a, b∈Z p , e(P a ,Q b )=e(P,Q) ab ; then randomly select a generator g1 in the G1 group and a generator g2 in the G2 group; then, select nine hash functions, h3:Z p →Z p ,H1:{0,1} * →G1,H2:{0,1} * →G2,H3:G T →{0,1}l ,H4:G T →G T , H6:G2→G2; Indicates that there are two elements on the G2 group; τ represents the number of elements, n1 represents an integer, and the parameter l represents the data particle length;

[0044] RA randomly selects two random numbers And use it as the master private key msk, save msk locally, calculate And initialize all possible access condition sets ACU; finally, output and publish system public parameters

[0045]

[0046] Phase 2: Access level initialization; First, randomly select q∈Z p , calculate the trusted data space access level parameter D P =H1(q) β ,Q P =H2(q); Assume that the access level is divided into 1-N levels, N is the highest access level; randomly select N access level parameters Calculate the level N parameter Q N =H6(Q P ); for i = N-1 to 1, calculate the level i parameter, Q i =H6(Q i+1 ); Calculate aggregate access level parameters Output access level parameter ALP=({Q i} i∈[N] ,{D i} i∈[N] ,D P ,Q P ,ALQ,ALV);

[0047] Finally, initialize a pair of ECC algorithm public and private keys And use Feldman's Verifiable Secret Sharing algorithm to split it into pieces and send it to all trusted spaces;

[0048] Step 2:

[0049] Randomly select random parameters in trusted data space Calculate the intermediate parameter γ=h1(ID PDS ,σ), Output (ψ,γ), ψ PDS ,ID PDS Sent to the regulator RA; the regulator randomly selects a random parameter r∈Z p ,Store r,ID PDS ,ψ PDS , calculate the intermediate parameter ε=h2(a,r) and the private key of the trusted data space Send(pk PDS ,ε,ALP) to the trusted data space; the trusted data space receives (pk PDS ,ε), combined with γ, calculate the intermediate parameter α PDS =γε,sk PDS =α PDS ; Public and private keys in trusted space Finally, initialize the blockchain public and private key pair

[0050] Step 3:

[0051] When a data user registers, he first sends his access condition set ACS to the trusted data space. After receiving it, the trusted data space generates a verification key vsk for the ACS. DU The specific process is as follows:

[0052] First, using the random parameters r∈Z selected during registration p ,For each access condition c∈ACS, calculate the first parameter vsk of the verification key 1,c =H1(c) r , the second parameter of the verification key The third parameter of the verification key The fourth parameter of the verification key is vsk4=H1(ID DU ) r , the fifth parameter of the verification key The sixth parameter of the verification key Data user's verification key vsk DU =(ACS,{vsk 1,c} c∈ACS , vsk2, vsk3, vsk4, vsk5, vsk6); calculate the access level of the data owner Trusted data space will AL DU Upload blockchain evidence, blockchain transactions are Store intermediate parameters H1(ID DU ) r With ID DU The corresponding relationship, output vsk DU ,ALDU to the data user;

[0053] Step 4 is specifically as follows:

[0054] The data owner sends data description information {DDS i , i} i∈[n] to the trusted data space, and the trusted data space performs data sensitivity level authorization: DSLReq(mpk, ALP, {DDS i , i} i∈[n] )→({DSL i} i∈[n] ): for i∈[n], n represents the total number of data; calculate intermediate parameters DDS i is data description information, DSL i is data sensitivity level, and the corresponding data is DATA i ; upload a blockchain transaction to the blockchain for storage; and send the requested data sensitivity level ({DSL i} i∈[n] ) to the data owner.

[0055] Step 5 is specifically as follows:

[0056] The data owner runs to encapsulate the data into a capsule and store it in the trusted data space in advance; the maximum length of a single data is l; select a random parameter a∈{0,1} l , Calculate the intermediate parameter P1=a, the capsule identifier Randomly select a parameter y∈Z p , and calculate the first parameter of the data capsule Intermediate parameter The second parameter of the data capsule Access control policy M is (n1×n2); select a random vector For j∈[τ] calculate the third parameter of the data capsule for i in[n1] calculate the fourth parameter of the data capsule Data capsule local parameter Data capsule Send DCI, DC, {DDS i}i∈[n] Give trusted data space, local storage DCI,{DATA i} i∈[n] Corresponding relationship; Calculate the first supervision record of the capsule R1 = ({DDS i} i∈[n] ||C2), intermediate parameters

[0057] Capsule verification parameters Upload blockchain transactions On to the blockchain.

[0058] Step 6:

[0059] The data owner runs Indicates the data to be shared by the capsule {DAT A i} i∈[n] Subscript subset of ; Parsing capsule local parameters First, use the data owner's private key sk DO Calculate the first parameter of the partial task Calculate intermediate parameters of some tasks and Randomly select z∈Z p , calculate the second parameter of the partial task The third parameter of some tasks For w∈DI randomly select r w ∈Z p , calculate the intermediate parameters of some tasks

[0060] Indicates the data to be encapsulated in some tasks w The first parameter, Indicates the data to be encapsulated in some tasks w The second parameter of The fourth parameter representing the partial task;

[0061] Then, calculate the partial task ID: T ID =∏ w∈DI r w , set the access limit atl; output some tasks Send DCI, T DO , C, {DSL w} w∈DI to the trusted data space; set the second piece of supervision data Upload to the blockchain; C represents the access frequency control parameter, represents the first parameter of access frequency control, represents the second parameter of access frequency control; Timestamp represents the current timestamp; represents the input of the blockchain transaction, represents the public key used by RA.

[0062] Step 7 is specifically as follows:

[0063] The trusted data space executes PDS TaskIssue (mpk, skPDS, IDDU, DCI, DI, {DSL w} w∈DI , AL DU , C, ALP, T DO )→(T)

[0064] Extracts part of the parameters from the partial task T DO Randomly samples parameters for w∈DI Calculates four parameters of the partial task Sets the partial task Randomly samples r t ∈Z p , t1 = h3(r t ), For i = 2 to atl, calculates t i = h3(t i-1 ), obtains {t i} i∈[atl] , at = 0, calculates four parameters of the access frequency control parameter Finally outputs the complete task Complete access control parameter Locally stores T ID , C, sends DCI, T, DC to the data user.

[0065] Step 8 is specifically as follows:

[0066] The data user executes Calculates and outputs intermediate parameters ​Send DCI, T ID , To the trusted data space.

[0067] Step 9 is specifically as follows:

[0068] The trusted data space executes

[0069] First, judge If satisfied, calculate the temporary variable of the third and fourth access control parameters Update the number of times accessed at = at + 1, Send the number of times accessed tokens To the data user, set the third regulatory data Update the third and fourth parameters in the access control parameter C Not updated; upload blockchain transactions To the blockchain for storage.

[0070] Step 10 is specifically as follows:

[0071] The data user executes

[0072] 1) Data capsule integrity verification phase to parse data capsule Calculate intermediate parameters If Return ⊥;

[0073] 2) Attribute-based decryption phase: restore intermediate parameters If vsk DU The ACS meets the access control policy, which can decrypt the correct

[0074] 3) Data recovery phase: calculate intermediate parameters Reconstruct intermediate parameters For w ∈ DI, calculate intermediate parameters

[0075]

[0076] Finally, restore the data particles Set the fourth regulatory data Upload blockchain transactions To the blockchain for storage.

[0077] Step 11 specific method as follows:

[0078] After obtaining at least t secret sub-shares (obtaining at least t trusted data space consent), the regulatory authority RA recovers the regulatory key using the Feldman's Verifiable Secret Sharing algorithm According to the corresponding capsule DCI, task ID: T ID , query the capsule related blockchain record, extract four pieces of regulatory data R1, R2, R3, R4, and execute Using Decrypt the intermediate parameters for w in[DI]get After recovering the data, the regulatory authority recovers the data.

[0079] Embodiment: The algorithm efficiency comparison test data of the application are as follows:

[0080] The experimental environment is Windows 10, Intel(R)Core(TM)i7-7700HQ CPU@2.80Ghz, 16gbRAM. The experimental test uses Python3.6.9, and depends on PBC-0.5.14 and Charm-0.50 and other cryptographic libraries. The test uses the MNT-224 curve. In the parameter setting, this paper sets l=128. Since the NHS+ scheme and the SMZ+ scheme encrypt and share data, this paper sets n=1. In addition, the NHS+ scheme and the SMZ+ scheme do not support attribute reuse, so this paper sets tau=1. In practical applications, |ACU|>>|ACS|, so the application sets |ACU|=100, |S|=n1=10. According to the results of the theoretical analysis, the experimental analysis of the data security sharing method based on bidirectional hierarchical matching and cooperative supervision is as shown in Figure 3 and Figure 4 .

[0081] In Figure 3 , the application respectively compares and analyzes the running time overhead of the initialization, DO key generation, DU key generation, encryption, preparation, decryption six stages, and compares the performance of the same stage with several current mainstream schemes. According to the benchmark test results of the commonly used cryptographic operators, the average time consumption of the exponential operation on G1, G2 and G_T groups is 0.92ms, 6.71ms and 1.45ms respectively, and the average time consumption of the bilinear pairing operation is 4.9ms. The time consumption of the hash operation is only 1 / 100-1 / 10 of the time consumption of the exponential operation, which can be ignored, so it is not listed separately in the statistics.

[0082] In the initialization phase, BLM-DCSS has slightly higher time cost than the comparison schemes; however, initialization is only performed once when the system is deployed, and its cost is amortized over the entire system life cycle, so it is completely acceptable.

[0083] In the DO key generation phase, the time cost of BLM-DCSS remains constant with respect to the number of access conditions m, and is only slightly higher than the previous generation LZR+. This additional time cost is exchanged for more functions (such as time window control and revocation), and the cost performance is outstanding.

[0084] In the DU key generation phase, the time cost of this phase is still approximately constant, and the curve slope is very small: only 4-5 ms higher than LZR+. Given that the DU end only needs to generate a key once before receiving a task, this increase can be considered as exchanging new features such as traceability and bidirectional hierarchical matching at a very low cost, and it is completely acceptable.

[0085] In the encryption phase, we take the logarithm of the ordinate for ease of display. As can be seen from the figure, when m\ge 20, the growth rate of the overhead of BLM-DCSS is the lowest, and the overall time cost is also optimal - thanks to its moving the heavy computation linearly related to m to the offline phase, only retaining constant-time lightweight operations.

[0086] In the preparation phase, the running time of BLM-DCSS does not increase with m, and is a horizontal straight line; although the constant term is slightly higher than LZR+, it is still much better than NHS+ and SMZ+, and this phase is a background offline process, which has no significant impact on real-time performance.

[0087] In the decryption phase, the decryption end also maintains constant time, and is only about 20 ms higher than LZR+. In actual application scenarios, this difference is minimal, but it exchanges more fine-grained revocation and auditing capabilities.

[0088] In summary, BLM-DCSS effectively moves or offloads the heavy computation linearly related to the number of access conditions m to the offline phase, so that the time cost of the online path (encryption and decryption) is almost not affected by m; at the same time, by introducing bidirectional hierarchical matching, revocation, and traceability in constant-level overhead, the performance and function are balanced.

[0089] As shown in Figure 4 The storage overhead of the key components is evaluated, including the master public key (MPK), the data owner and data user private keys (KeyDO and KeyDU), and the ciphertext (CT). At the same time, the storage cost of the auxiliary materials generated in the data sharing preparation phase is compared: DReq of NHS+, tk and re-encryption ciphertext of SMZ+, task of LZR+, and DOTask and PDSTask in BLM-DCSS of this paper. The group element size, single data item size, and access policy size are used as measurement parameters.

[0090] The results show that the MPK size of SMZ+, LZR+ and BLM-DCSS remains constant, while NHS+ grows linearly with the number of access control units. In terms of DO private key, the size of LZR+ and BLM-DCSS remains constant, while SMZ+ grows linearly with the number of access control policies, and other schemes do not involve this component. The DU private key size grows linearly with the number of access control policies in all schemes. For ciphertext, all schemes except SMZ+ grow linearly with the number of data blocks; SMZ+ grows linearly with the number of access control policies. In the preparation phase, the storage of BLM-DCSS and LZR+ remains constant, and the storage of the rest of the schemes grows linearly with the number of shared data items.

Claims

1. A data security sharing method based on two-way grade matching and collaborative supervision, characterized in that: The specific steps are as follows: Step 1. System initialization: Construct initialization parameters according to various standards; Step 2. Trusted Data Space Registration: The trusted data space submits an application to the regulator (RA). The RA generates a public key for the trusted data space based on its ID, and the trusted data space generates its own private key based on the public key. Step 3. Data user registration: The data user submits an application to the trusted data space. The trusted data space generates a decryption key and access level for the data user based on the data user's ID, attribute set, and the data protection capabilities of the environment in which they are located. Step 4. Data Sensitivity Level Application: The data owner applies for the data sensitivity level from the trusted data space based on relevant laws and regulations or assessment standards. Step 5. Data capsule encapsulation: The data owner packages their data into a data capsule, uses the constructed access control structure to encapsulate the data capsule, obtains the data capsule, and uploads the capsule to the trusted data space; Step 6. Partial access task generation: The data owner generates a partial access task with a customized number of accesses based on the needs of the data user and outputs the partial access task; Step 7. Access task generation: Based on the partial access tasks, the trusted data space generates a complete access task according to the sensitivity level of the shared data and the access level of the data user, and outputs the complete access task; Step 8. Access task decryption: After receiving the access task, the data user decrypts the semi-decrypted result from the access task; Step 9. Generate a count token and download a data capsule: The data user uses the semi-decrypted result to send a count token request and a data capsule download request to the data space to obtain the count token and data capsule; Step 10. Data capsule decryption: The data user uses the access task, the semi-decryption result and the times token to decrypt the data capsule and output the data granule; Step 11. Collaborative supervision: When unauthorized access or data fraud disputes occur, the supervisor RA collaborates with at least t trusted data spaces to supervise and resolve the disputes.

2. A data security sharing method based on two-way level matching and collaborative supervision according to claim 1, characterized in that: Step 3: When a data user registers, he first sends his access condition set ACS to the trusted data space. After receiving it, the trusted data space generates a verification key vsk for the ACS. DU The specific process is as follows: First, using the random parameters r∈Z selected during registration p ,For each access condition c∈ACS, calculate the first parameter vsk of the verification key 1,c =H1(c) r , the second parameter of the verification key The third parameter of the verification key The fourth parameter of the verification key is vsk4=H1(ID DU ) r , the fifth parameter of the verification key The sixth parameter of the verification key Data user's verification key vsk DU =(ACS,{vsk 1,c } c∈ACS , vsk2, vsk3, vsk4, vsk5, vsk6); calculate the access level of the data owner Trusted data space will AL DU Upload blockchain evidence, blockchain transactions are Store intermediate parameters H1(ID DU ) r With ID DU The corresponding relationship, output vsk DU ,AL DU To the data user.

3. A data security sharing method based on two-way level matching and collaborative supervision according to claim 2, characterized in that: Step 4: The data owner sends data description information {DDS i ,L i } i∈[n] To the trusted data space, the trusted data space performs data sensitivity level authorization: DSLReq(mpk,ALP,{DDS i ,L i } i∈[n] )→({DSL i } i∈[n] ): for i∈[n], n represents the total number of data; calculate the intermediate parameters DDS i For data description information, DSL i For data sensitivity level, the corresponding data is DATA i ;Upload blockchain transactions Store evidence on the blockchain; send the requested data sensitivity level ({DSL i } i∈[n] ) to the data owner.

4. A data security sharing method based on bidirectional level matching and collaborative supervision according to claim 3, characterized in that: Step 5: Data owner runs Encapsulate data into capsules and store them in the trusted data space in advance; the maximum length of a single data is l; select random parameters a∈{0,1} l , Calculate the intermediate parameter P1 = a, capsule identifier Randomly select parameters y∈Z p , calculate the first parameter of the data capsule Intermediate parameters The second parameter of the data capsule Access control policy M is (n1×n2); select a random vector For j∈[τ] calculate the third parameter of the data capsule for iin[n1], calculate the fourth parameter of the data capsule H1(π(i)) y′[ρ(i)] ,Data capsule local parameters Data Capsule Send DCI, DC, {DDS i } i∈[n] Give trusted data space, local storage DCI,{DATA i } i∈[n] Correspondence; The first regulatory record of the calculation capsule R1=({DDS i } i∈[n] ||C2), intermediate parameters Capsule verification parameters Upload blockchain transactions On to the blockchain.

5. A data security sharing method based on two-way level matching and collaborative supervision according to claim 4, characterized in that: Step 6: The data owner runs Indicates the data to be shared by the capsule {DATA i } i∈[n] Subscript subset of ; Parsing capsule local parameters First, use the data owner's private key sk DO Calculate the first parameter of the partial task Calculate intermediate parameters of some tasks and Randomly select z∈Z p , calculate the second parameter of the partial task The third parameter of some tasks For w∈DI randomly select r w ∈Z p , calculate the intermediate parameters of some tasks Indicates the data to be encapsulated in some tasks w The first parameter, Indicates the data to be encapsulated in some tasks w The second parameter of The fourth parameter representing the partial task; Then, calculate the partial task ID: T ID =Π w∈DI r w , set the access limit atl; output some tasks Send DCI, T DO ,C,{DSL w } w∈DI Provide trusted data space; set up the second regulatory data Upload Store the certificate on the blockchain; C represents the access control parameter. The first parameter indicating the number of visits. Indicates the second parameter for access count control; Timestamp indicates the current timestamp; Represents the input of a blockchain transaction, Indicates the public key used by RA.

6. A data security sharing method based on two-way level matching and collaborative supervision according to claim 5, characterized in that: Step 7: Trusted data space executes PDSTaskIssue(mpk,sk PDS ,ID DU ,DCI,DI,{DSL w } w∈DI ,AL DU ,C,ALP,T DO )→(T) from part of task T DO Extract some parameters T ID , for w∈DI, randomly sampled parameters Calculate 4 parameters of some tasks Setting up partial tasks Random sampling r t ∈Z p ,t1=h3(r t ),For i=2to atl calculate t i =h3(t i-1 ), get {t i } i∈[atl] ,at=0, calculate the four parameters of the access control parameters Final output of the complete task Complete access control parameters Local Storage ID ,C, sends DCI,T,DC to data users.

7. A data security sharing method based on two-way level matching and collaborative supervision according to claim 6, characterized in that: Step 8: Data user execution Calculate and output intermediate parameters Send DCI,T ID , Provide trusted data space.

8. A data security sharing method based on two-way level matching and collaborative supervision according to claim 7, characterized in that: Step 9: Trusted Data Space Execution First judge If satisfied, calculate the temporary variables of the third and fourth access control parameters Update the number of visits at=at+1, Send visit token For data users, set up the third regulatory data Update the third and fourth parameters in access control parameter C Do not update; upload blockchain transactions Store evidence on the blockchain.

9. A data security sharing method based on two-way level matching and collaborative supervision according to claim 8, characterized in that: Step 10: The specific method is as follows: Data user execution Data capsule integrity verification phase: parsing data capsules Calculate intermediate parameters if return ⊥; Attribute-based decryption phase: recovering intermediate parameters If vsk DU The ACS in the system satisfies the access control policy and can decrypt the correct Data recovery phase: calculating intermediate parameters Reconstructing intermediate parameters For w∈DI, calculate the intermediate parameters Finally, restore the data granules Set the fourth regulatory data Upload blockchain transactions Store evidence on the blockchain.

10. A data security sharing method based on two-way level matching and collaborative supervision according to claim 9, characterized in that: Step 11: After obtaining at least t secret sub-shares (and obtaining consent from at least t trusted data spaces), the regulatory authority RA uses Feldman's Verifiable Secret Sharing algorithm to recover the regulatory key. According to the corresponding capsule DCI, task ID: T ID , query the blockchain records related to the capsule, extract four regulatory data R1, R2, R3, R4, and execute use Decrypt the intermediate parameters for w in[DI] get Supervision is carried out after the data is restored.

Images