Attribute-based Internet of Vehicles dual access control data sharing method

Abstract

The invention discloses an attribute-based Internet of Vehicles dual access control data sharing method, which comprises the following steps of: generating an attribute key by a user, and generating an agent key by using the attribute key; an owner of the data encrypts the data by using an attribute-based encryption algorithm according to a defined semi-hidden access structure, and the encryption process is divided into an offline stage and an online stage in consideration of limited computing power of a vehicle end; the cloud server filters out false information sent by the malicious user; (access control is performed on a download request sent by a user who wants to access data, so that EDOS attack is denied; only the user meeting the access strategy can decrypt the ciphertext, and the decryption stage is also divided into an outsourcing decryption stage and a user decryption stage; the malicious user is revoked, and after the malicious user is revoked, only the ciphertext related to the revocation list needs to be updated, and the whole ciphertext does not need to be updated. According to the invention, safe data sharing in the Internet of Vehicles can be realized, and wide application of the Internet of Vehicles is promoted.

Description

Technical Field

[0001] The present invention relates to vehicle networking and attribute encryption technology, and in particular to an attribute-based vehicle networking dual access control data sharing method. Background Art

[0002] Vehicle-to-vehicle networks (VANETs) are distributed self-organizing networks that improve driver safety and efficiency by transmitting vehicle status information in real time. By monitoring traffic flow in real time, VANETs enable intelligent planning, improve road traffic control capabilities, and inject new digital intelligence into urban transportation. VANETs also advance the digitalization of traffic management by combining information with vehicles to enhance the perception and computing capabilities of roadside infrastructure. While VANETs offer significant benefits for improving traffic efficiency, there are potential issues related to data privacy and security. For example, vehicle-shared data may include personal and sensitive information, such as the owner's identity and vehicle driving trajectory. In the dynamic network topology of VANETs, ​​user privacy is most vulnerable to leakage, and the leaked data and information may bring huge losses and even threaten public safety. Therefore, privacy protection is crucial in VANET security.

[0003] In order to achieve secure data sharing, the system is required to reasonably encrypt and protect the data before transmission by vehicle users, and it must also have good access control. Ciphertext Policy Attribute Encryption (CP-ABE) allows data owners to customize attribute-based access policies for their private data. Therefore, ciphertext policy attribute encryption (CP-ABE) can be better combined with cloud data sharing systems. However, it is not enough to implement access control to shared data only in the Internet of Vehicles. Some special situations may be encountered in the Internet of Vehicles environment. For example, malicious vehicles may send an unlimited number of download requests to block the resources of the cloud server, thereby hindering its ability to respond appropriately to service requests from legitimate users. In addition to access control to shared data, effective access control is also required for download requests from vehicles to ensure normal communication between vehicles.

[0004] In addition, in order to enhance the user authentication and data retrieval functions, we integrate blockchain technology into the solution. Blockchain is essentially a blockchain-stored, immutable, secure, and trusted decentralized distributed ledger. Each block has a different identity from the previous block, creating a chain that can only be appended. Blockchain uses its tamper-proof and trustworthy nature to achieve reliable data storage and strict access control through transaction verification and consensus mechanisms, ensuring data security and consistency. The immutable nature of blockchain technology can achieve secure and tamper-proof data storage, thereby ensuring data reliability and integrity. In our solution, blockchain is mainly used to securely manage user identities and data information between vehicles and RSUs. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to solve the deficiencies in the prior art and to provide a method for sharing data in a dual access control system based on attributes in an Internet of Vehicles.

[0006] Technical solution: A dual access control data sharing method for Internet of Vehicles based on attributes of the present invention involves a vehicle, an attribute authority AA, a roadside unit RSU, and a cloud server CSP, and includes the following steps:

[0007] Step (1), system initialization, attribute authority AA constructs bilinear pairing e:G×G→G T , G and G T are two multiplicative cyclic groups of order p, g is a generator of the group; choose is a cryptographic hash function. AA randomly selects h,v∈G and generates a global parameter GP=(G,G T ,e,g,h,v,H);

[0008] Step (2) Initialize the attribute authority AA, that is, each AA randomly selects Set your own private key to And the corresponding public key is obtained by calculation

[0009]

[0010] Step (3), key generation phase, the attribute authority AA generates a decryption key SK and its corresponding blinding key AK for the user DU;

[0011] Step (4), ciphertext generation phase, the data owner DO undergoes two phases of offline encryption and online encryption before sending the data; the specific contents are as follows:

[0012] Step (4-1), DO input R is the revocation list, M is a l×n matrix, ρ is a mapping function, A set of vectors Z=(s,z2,...,z n ) T , s∈Z p ,(z2,...,z n )∈Z p

[0013] Step (4-2), call the linear secret sharing LSSS scheme, map a row in the matrix M to a user attribute, and calculate λ i =M i ·Z,s is used to encrypt data, (z2,...,z n) is used to hide s;

[0014] Step (4-3), for i∈[1,l], randomly select t i ∈Z p , calculate the components of the ciphertext associated with the access policy:

[0015]

[0016] Step (4-4), for j∈cover(R), calculate the ciphertext component associated with the revocation list R

[0017]

[0018] The ciphertext generated by offline encryption is finally as follows:

[0019] Step (4-5), DO input m is the plaintext data to be encrypted, after online encryption calculation The final ciphertext is generated as follows:

[0020] in is a hidden access policy that does not contain attribute values;

[0021] Step (5), ciphertext filtering stage, the cloud server CSP checks whether the ciphertext is encrypted and generated under the declared access policy, thereby filtering out false ciphertexts; the specific method is:

[0022] Considering that there may be malicious users who send false messages to CSP, that is, malicious users do not encrypt ciphertext under the defined access policy, CSP will perform a filtering operation on the received ciphertext. CSP randomly selects elements γ and calculate

[0023]

[0024] CSP Check Whether the equality holds. If not, CSP will reject the ciphertext, thus achieving the effect of ciphertext filtering;

[0025] Step (6), download request access control phase, before DU accesses data, it will send a download request to CSP, CSP checks whether the download request is legal, if not, CSP rejects the download request, thus avoiding the occurrence of EDOS attack;

[0026] Step (7), in the ciphertext decryption phase, the DU that meets the access policy will be outsourced to the RSU for decryption before the final decryption, thus reducing the decryption overhead of the DU;

[0027] Step (8), during the user revocation phase, the system will track down malicious users and add their identities to the revocation list;

[0028] Step (9), ciphertext update phase, after the user revocation phase is completed, the system will perform a ciphertext update operation to update the relevant ciphertext.

[0029] Furthermore, the specific process of the attribute authority AA generating a key for the user DU in step (3) is as follows:

[0030] Step (3-1), AA θ Assign a set of corresponding attributes S = {β1,β2,...,β θ}, and build a binary tree where leaf nodes are associated with users;

[0031] Step (3-2), calculate the intermediate variable c = Enc k (i d ), where k is the symmetric encryption key, i d is the sequence number of the leaf node associated with the user in the binary tree;

[0032] Step (3-3), (AA will construct the user into a binary tree, where the leaf nodes represent the users and each node is assigned a random value; i represents the sequence number, i d Represents the sequence number of the leaf node)AA θ Assign a random value x to each node in the binary tree i , and randomly select r,r1,r2,...,r θ ∈Z p , generate private key components related to user attributes;

[0033]

[0034] The private key component associated with the user's identity in is the value of the leaf node;

[0035] L1, L2, h r , g r , g ri , They are all private key components related to attributes. They need to be set for subsequent calculations and have no specific meaning. is the value of the binary tree leaf node; i in i∈[θ] is the same as “x i The subscript "i" in i∈[θ] does not have the same meaning. The i in i∈[θ] represents the number of AA. i The subscript i represents the serial number;

[0036] Step (3-4), algorithm outputs decryption key And send it to the user through a secure channel.

[0037] Steps (3-5), AA θ The decryption key SK is used to generate the blinding key AK and the corresponding recovery key RK required for outsourced decryption. θ Random Selection calculate

[0038]

[0039] Output AA θ Send SK and AK to the user.

[0040] Furthermore, the specific process of the download request access control phase in step (6) is as follows:

[0041] Step (6-1), when DU wants to download data, it sends a download request to CSP, which contains its own attributes. CSP calls function ρ to map DU's attributes into attribute matrix M';

[0042] Step (6-2), determine whether the attribute matrix M' exists a reversible matrix. If M' does not exist a reversible matrix, it means that the DU attribute does not meet the access policy, that is, the encryption index s cannot be reconstructed, indicating that the user cannot decrypt the ciphertext, and the CSP will ignore the user's download request.

[0043] Step (6-3), if M' has a reversible matrix, it means that the DU attribute satisfies the access policy, and CSP will send the complete ciphertext C to DU.

[0044] Furthermore, the specific process of decrypting the ciphertext in step (7) is as follows:

[0045] Step (7-1), call the LSSS solution to reconstruct Σ i∈I c i λ i =s, where c i is a set of vectors satisfying ∑ i∈I c i M i =(1,0,...,0);

[0046] Step (7-2), RSU calculation

[0047]

[0048] The RSU sends the pre-decrypted ciphertext IC=(C1, B) to the DU.

[0049] Step (7-3), after DU receives the pre-decrypted ciphertext IC, it executes the final decryption algorithm to obtain the plaintext message m.

[0050]

[0051] Furthermore, the specific method of user cancellation in step (8) is:

[0052] Step (8), by calculating i d =Dec k (K2) Find the value of the leaf node in the binary tree, find the corresponding user identity u, add u to the revocation list R to obtain R'=R∪u; and prepare for the ciphertext update operation.

[0053] Furthermore, the specific process of updating the ciphertext in step (9) is as follows:

[0054] Step (9-1): When the user cancels, the value of the node in the binary tree needs to be updated, AA θ Randomly select t∈Z p , calculated For j′∈cover(R), there are two cases:

[0055] If there exists j∈cover(R) such that j=j′, then let

[0056] If there exists j∈cover(R) such that j is an ancestor of j′, define path(j′)=path(j)′{i dept(j)+1 ,...,i dept(j′)}, where i dept(j) =j,i dept(j′) = j′. k∈{depth(j),depth(j′)}, iterative calculation

[0057] Step (9-2), in the ciphertext update phase, the ciphertext component related to the access policy is not changed, only the ciphertext component related to the revocation list is updated, so the updated ciphertext is:

[0058]

[0059] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0060] (1) Considering the confidentiality of data and the security of communication, the designed encryption scheme implements dual access control. After the data owner encrypts the data according to the access policy, the data is sent to the cloud server to ensure that only users who meet the access policy can access the data. This scheme controls the user's download request to ensure that the shared data in the cloud can only be downloaded by authorized users, demonstrating the system's ability to resist EDOS attacks.

[0061] (2) Using blockchain dual-chain technology, transaction information and related authentication information and data are stored on the blockchain, achieving efficient user and data management. By deploying smart contracts on the user chain and data chain, a complete user authentication and data security management mechanism can be established to ensure user privacy and data integrity.

[0062] Beneficial effects: We achieve secure data sharing by improving traditional CP-ABE. Our scheme uses blockchain for more secure operations. In addition, the security of the cryptographic protocol is evaluated through formal security analysis and is able to resist multiple types of attacks, which is more secure than other related schemes. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a system model diagram of the present invention;

[0064] Figure 2 This is a block structure diagram of the blockchain in the present invention;

[0065] Figure 3 This is the cost diagram of the smart contract of the present invention. DETAILED DESCRIPTION

[0066] The technical solution of the present invention is described in detail below, but the protection scope of the present invention is not limited to the embodiments.

[0067] The present invention realizes flexible access control while supporting secure data sharing. The LSSS scheme is adopted to divide the attributes into two parts, namely, the attribute value and the attribute name, wherein the attribute value is encrypted to realize policy hiding. The data owner DO formulates the access policy, and only the users who meet the access policy can access and decrypt the ciphertext. The CSP filters out the false messages sent by the malicious users. The CSP performs fine-grained access control on the download request, and rejects the occurrence of EDOS attack. The system supports the user revocation operation, and the identity information of the malicious user is added to the revocation list to reject the access request of the malicious user. Meanwhile, the blockchain is adopted to securely manage the user identity and data information between the vehicle and the RSU, and the specific content is to write the user identity into the user chain, and write the ciphertext address and the ciphertext hash into the data chain.

[0068] like Figure 1As shown, the present invention involves the following participating entities, namely, attribute authority (AA), data owner (DO), roadside unit (RSU), data user (DU), cloud server provider (CSP) and blockchain network.

[0069] The attribute authority is responsible for generating global parameters within its management domain and generating attribute keys for DUs; data owners formulate access policies and upload ciphertexts to CSP; RSU acts as a communication medium and executes outsourced decryption algorithms; when the user's attributes meet the access policy of the ciphertext, the user can access the data and use the decryption algorithm to decrypt the file using the key to obtain the plaintext. If the data owner deletes any user, the user cannot access the group of data; the cloud service provider stores the encrypted files uploaded by the data owner and provides ciphertext filtering and download request access control functions. It manages access to stored data and provides related services.

[0070] The attribute-based Internet of Vehicles dual access control data sharing method of the present invention comprises the following steps:

[0071] Step (1), system initialization, attribute authority AA constructs bilinear pairing e:G×G→G T , G and G T are two multiplicative cyclic groups of order p, g is a generator of the group; choose is a cryptographic hash function. AA randomly selects h,v∈G and generates a global parameter GP=(G,G T ,e,g,h,v,H);

[0072] Step (2) Initialize the attribute authority AA, that is, each AA randomly selects Set your own private key to And the corresponding public key is obtained by calculation

[0073]

[0074] Step (3), key generation phase, the attribute authority AA generates a decryption key SK and its corresponding blinding key AK for the user DU;

[0075] Step (4), ciphertext generation phase, the data owner DO undergoes two phases of offline encryption and online encryption before sending the data; the specific contents are as follows:

[0076] Step (4-1), DO input R is the revocation list, M is a l×n matrix, ρ is a mapping function, A set of vectors Z=(s,z2,...,z n ) T , s∈Zp ,(z2,...,z n )∈Z p

[0077] Step (4-2), call the linear secret sharing (LSSS) scheme, map a row in the matrix M to a user attribute, and calculate λ i =M i ·Z,s is used to encrypt data, (z2,...,z n ) is used to hide s;

[0078] Step (4-3), for i∈[1,l], randomly select t i ∈Z p , calculate the components of the ciphertext associated with the access policy:

[0079]

[0080] Step (4-4), for j∈cover(R), calculate the ciphertext component associated with the revocation list R

[0081]

[0082] The ciphertext generated by offline encryption is finally as follows:

[0083] Step (4-5), DO input m is the plaintext data to be encrypted, after online encryption calculation The final ciphertext is generated as follows:

[0084] in is a hidden access policy that does not contain attribute values;

[0085] Step (5), ciphertext filtering stage, the cloud server CSP checks whether the ciphertext is encrypted and generated under the declared access policy, thereby filtering out false ciphertexts; the specific method is:

[0086] The cloud server CSP performs a filtering operation on the received ciphertext. CSP randomly selects elements γ and calculate

[0087]

[0088]

[0089] CSP Check Whether the equality holds. If not, CSP will reject the ciphertext.

[0090] Step (6), download request access control phase: before accessing data, DU sends a download request to the cloud server CSP. CSP checks whether the download request is legal. If not, CSP rejects the download request.

[0091] Step (7), in the ciphertext decryption phase, the DU that meets the access policy will be outsourced to the RSU for decryption before the final decryption;

[0092] Step (8), user revocation phase, malicious users will be tracked and their identities will be added to the revocation list;

[0093] Step (9), the ciphertext update phase, after the user revocation phase is completed, the ciphertext update operation is performed to update the relevant ciphertext.

[0094] Embodiment Step (3) The specific process of the attribute authority AA generating a key for the user DU is as follows:

[0095] Step (3-1), AA θ Assign a set of corresponding attributes S = {β1,β2,...,β θ}, and build a binary tree where leaf nodes are associated with users;

[0096] Step (3-2), calculate the intermediate variable c = Enc k (i d ), where k is the symmetric encryption key, i d is the sequence number of the leaf node associated with the user in the binary tree;

[0097] Step (3-3), AA θ Assign a random value x to each node in the binary tree i , i represents the serial number, i d Represents the sequence number of the leaf node, and randomly selects r,r1,r2,...,r θ ∈Z p , generate private key components related to user attributes;

[0098]

[0099] The private key component associated with the user's identity in is the value of the leaf node;

[0100] Step (3-4), algorithm outputs decryption key And send it to the user through a secure channel;

[0101] Steps (3-5), AA θUse the decryption key SK to generate the blinding key AK and the corresponding recovery key RK required for outsourced decryption; AA θ Random Selection calculate:

[0102]

[0103] Output

[0104] Finally, AA θ Send SK and AK to the user.

[0105] The specific process of the download request access control stage in step (6) of the embodiment is as follows:

[0106] Step (6-1), when DU wants to download data, it sends a download request to CSP, which contains its own attributes. CSP calls function ρ to map DU's attributes into attribute matrix M';

[0107] Step (6-2), determine whether the attribute matrix M' is a reversible matrix. If M' does not have a reversible matrix, it means that the DU attribute does not meet the access policy, that is, the encryption index s cannot be reconstructed, indicating that the user cannot decrypt the ciphertext, and the CSP ignores the user's download request;

[0108] Step (6-3), if M' has a reversible matrix, it means that the DU attribute satisfies the access policy, and CSP will send the complete ciphertext C to DU.

[0109] The specific process of decrypting the ciphertext in step (7) of the embodiment is as follows:

[0110] Step (7-1), call the LSSS solution to reconstruct ∑ i∈I c i λ i =s, where c i is a set of vectors satisfying Σ i∈I c i M i =(1,0,...,0);

[0111] Step (7-2), RSU calculation

[0112]

[0113] RSU sends the pre-decrypted ciphertext IC=(C1,B) to DU;

[0114] Step (7-3), after receiving the pre-decrypted ciphertext IC, DU executes the final decryption algorithm to obtain the plaintext message m;

[0115]

[0116] The specific method of user cancellation in step (8) of the embodiment is:

[0117] By calculating i d =Dec k (K2) Find the value of the leaf node in the binary tree, find the corresponding user identity u, add u to the revocation list R to obtain R'=R∪u; and prepare for the ciphertext update operation.

[0118] The specific process of updating the ciphertext in step (9) of the embodiment is as follows:

[0119] Step (9-1): When the user cancels, the value of the node in the binary tree needs to be updated, AA θ Randomly select t∈Z p , calculated For j′∈cover(R), there are two cases:

[0120] If there exists j∈cover(R) such that j=j′, then let

[0121] If there exists j∈cover(R) such that j is an ancestor of j′, define path(j′)=path(j)∪{i dept(j)+1 ,...,i dept(j′)},

[0122] where i dept(j) =j,i dept(j′) = j′, k∈{depth(j),depth(j′)}, iterative calculation

[0123] Step (9-2), in the ciphertext update phase, the ciphertext component related to the access policy has not changed, only the ciphertext component related to the revocation list has been updated, so the updated ciphertext is:

[0124]

[0125] In order to facilitate understanding of the technical solution of this embodiment, the meaning of the relevant variables is first explained, and the specific content is shown in Table 1.

[0126] Table 1

[0127]

[0128]

[0129] Example:

[0130] The blockchain structure used in this embodiment is as follows: Figure 2As shown, a block in the blockchain consists of a block header and a block body, wherein the block header contains the hash of the previous block and the root of a Merkle tree, and a large amount of complex data is recorded in the Merkle tree.

[0131] The functional implementation of the technical solution of the present invention is compared, and the results are shown in Table 2. The technical solution of the present invention supports more functions suitable for the Internet of Vehicles.

[0132] Table 2 Functional implementation analysis

[0133]

[11]

[18]

[24] [2] Ours Access structure LSSS LSSS LSSS LSSS LSSS Traceability × × × × √ Recovation level × Attribute User User,Attribute User Update × Ciphertext,Key Key Ciphertext,Key Ciphertext Based-Blockchain × × × × √ Outsources decryption √ √ × √ √ Request control × × × × √ Ciphertext check × × × × √ Forward security √ √ √ √ √ Backward security √ × √ √ √

[0134] The computational overhead of the technical solution of the present invention is compared, and the results are shown in Table 3. The technical solution of the present invention still has a relatively good computational overhead under relatively secure privacy protection.

[0135] Table 3 Computational cost analysis

[0136]

[0137] Where E is the exponential operation time on the group, P is the bilinear pairing time, M is the multiplication operation time on the group, s is the number of user attributes, u is the number of user attributes managed by an authority, H is the hash operation time, and t is the number of revoked attributes.

[0138] like Figure 3 As shown, the cost overhead of the smart contract of the technical solution of the present invention shows that the present invention has reasonable gas overhead, which is feasible for the Internet of Vehicles.

Claims

1. A dual access control data sharing method for Internet of Vehicles based on attributes, characterized in that: The following steps are involved: Step (1), system initialization, attribute authority AA constructs bilinear pairing e:G×G→G T , G and G T are two multiplicative cyclic groups of order p, and g is a generator of the group; choose H: is a cryptographic hash function. AA randomly selects h,v∈G and generates a global parameter GP=(G,G T ,e,g,h,v,H); Step (2) Initialize the attribute authority AA, that is, each AA randomly selects Set your own private key to And the corresponding public key is obtained by calculation Step (3), key generation phase, the attribute authority AA generates a decryption key SK and its corresponding blinding key AK for the user DU; Step (4), ciphertext generation phase, the data owner DO undergoes two phases of offline encryption and online encryption before sending the data; the specific contents are as follows: Step (4-1), DO input R is the revocation list, M is a l×n matrix, ρ is a mapping function, A set of vectors Z=(s,z2,...,z n ) T , s∈Z p ,(z2,...,z n )∈Z p Step (4-2), call the linear secret sharing LSSS scheme, map a row in the matrix M to a user attribute, and calculate λ i =M i ·Z,s is used to encrypt data, (z2,...,z n ) is used to hide s; Step (4-3), for i∈[1,l], randomly select t i ∈Z p , calculate the components of the ciphertext associated with the access policy: Step (4-4), for j∈cover(R), calculate the ciphertext component associated with the revocation list R The ciphertext generated by offline encryption is finally as follows: Step (4-5), DO input m is the plaintext data to be encrypted, after online encryption calculation The final ciphertext is generated as follows: in is a hidden access policy that does not contain attribute values; Step (5), ciphertext filtering stage, the cloud server CSP checks whether the ciphertext is encrypted and generated under the declared access policy, thereby filtering out false ciphertexts; the specific method is: The cloud server CSP performs a filtering operation on the received ciphertext. CSP randomly selects elements γ and calculate CSP Check Whether the equality holds. If not, CSP will reject the ciphertext. Step (6), download request access control phase: before accessing data, DU sends a download request to the cloud server CSP. CSP checks whether the download request is legal. If not, CSP rejects the download request. Step (7), in the ciphertext decryption phase, the DU that meets the access policy will be outsourced to the RSU for decryption before the final decryption; Step (8), user revocation phase, malicious users will be tracked and their identities will be added to the revocation list; Step (9), the ciphertext update phase, after the user revocation phase is completed, the ciphertext update operation is performed to update the relevant ciphertext.

2. The attribute-based dual access control data sharing method for Internet of Vehicles according to claim 1 is characterized in that: The specific process of the attribute authority AA generating a key for the user DU in step (3) is as follows: Step (3-1), AA θ Assign a set of corresponding attributes S = {β1,β2,...,β θ }, and build a binary tree where leaf nodes are associated with users; Step (3-2), calculate the intermediate variable c = Enc k (i d ), where k is the symmetric encryption key, i d is the sequence number of the leaf node associated with the user in the binary tree; Step (3-3), AA θ Assign a random value x to each node in the binary tree i , i represents the serial number, i d Represents the sequence number of the leaf node, and randomly selects r,r1,r2,...,r θ ∈Z p , generate private key components related to user attributes; The private key component associated with the user's identity in is the value of the leaf node; Step (3-4), algorithm outputs decryption key And send it to the user through a secure channel; Steps (3-5), AA θ Use the decryption key SK to generate the blinding key AK and the corresponding recovery key RK required for outsourced decryption; AA θ Random Selection calculate: Output RK = τ; Finally, AA θ Send SK and AK to the user.

3. The attribute-based Internet of Vehicles dual access control data sharing method according to claim 1, characterized in that: The specific process of the download request access control phase in step (6) is as follows: Step (6-1), when DU wants to download data, it sends a download request to CSP, which contains its own attributes. CSP calls function ρ to map DU's attributes into attribute matrix M'; Step (6-2), determine whether the attribute matrix M' is a reversible matrix. If M' does not have a reversible matrix, it means that the DU attribute does not meet the access policy, that is, the encryption index s cannot be reconstructed, indicating that the user cannot decrypt the ciphertext, and the CSP ignores the user's download request; Step (6-3), if M' has a reversible matrix, it means that the DU attribute satisfies the access policy, and CSP will send the complete ciphertext C to DU.

4. The attribute-based Internet of Vehicles dual access control data sharing method according to claim 1, characterized in that: The specific process of decrypting the ciphertext in step (7) is as follows: Step (7-1), call the LSSS solution to reconstruct ∑ i∈I c i λ i =s, where c i is a set of vectors satisfying Σ i∈I c i M i =(1,0,...,0); Step (7-2), RSU calculation RSU sends the pre-decrypted ciphertext IC=(C1,B) to DU; Step (7-3), after receiving the pre-decrypted ciphertext IC, DU executes the final decryption algorithm to obtain the plaintext message m; 5. The attribute-based dual access control data sharing method for Internet of Vehicles according to claim 1, characterized in that: The specific method of user cancellation in step (8) is: By calculating i d =Dec k (K2) Find the value of the leaf node in the binary tree, find the corresponding user identity u, add u to the revocation list R to obtain R'=R∪u; and prepare for the ciphertext update operation.

6. The attribute-based Internet of Vehicles dual access control data sharing method according to claim 1, characterized in that: The specific process of updating the ciphertext in step (9) is as follows: Step (9-1): When the user cancels, the value of the node in the binary tree needs to be updated, AA θ Randomly select t∈Z p , calculated For j′∈cover(R), there are two cases: If there exists j∈cover(R) such that j=j′, then let If there exists j∈cover(R) such that j is an ancestor of j′, then define path(j′)=path(j)∪{i dept(j)+1 ,...,i dept(j′) }, where i dept(j) =j,i dept(j′) = j′, k∈{depth(j),depth(j′)}, iterative calculation Step (9-2), in the ciphertext update phase, the ciphertext component related to the access policy has not changed, only the ciphertext component related to the revocation list has been updated, so the updated ciphertext is:

Images