Incremental efficient ciphertext retrieval method with dynamic time attribute control

Abstract

The invention provides an incremental efficient ciphertext retrieval method with dynamic time attribute control, a time server performs registration and identity authentication on a data uploader, and the authenticated data uploader has the authority of generating and uploading an encrypted index and authorizing other people to search. A data user who needs to search the ciphertext on the cloud server needs to apply for the authority of searching the uploaded ciphertext to a data uploader, and after the authorization of the data uploader is obtained, the data user receives an updating certificate broadcasted by the time server to update the authorization. According to the method, an authenticated data uploader can set an access limit of release time for the encrypted index according to specific application requirements, so that access control of a time dimension is realized; the scheme has the property of full increment, and the communication and storage overhead is reduced compared with the previous scheme; and under the condition of protecting user privacy and data privacy, the availability of the data is ensured.

Description

Technical Field

[0001] The present invention belongs to the field of computer application technology, and in particular relates to an incremental efficient ciphertext retrieval method with dynamic time attribute control. Background Art

[0002] With the rapid development of cloud service technology, users and enterprises are increasingly inclined to store data on remote cloud servers, and gradually developed many cloud service-based applications, such as data outsourcing storage backup, cloud-based data hosting and synchronization, and cloud databases. However, while enjoying the convenience of cloud services, users and enterprises are also gradually beginning to be threatened by potential threats brought by cloud services. Since data owners lose physical control of data after outsourcing it to cloud service providers, data owners can only blindly believe that cloud service providers will protect the privacy of their data. Contrary to expectations, a large number of cloud service data leaks have occurred around the world in recent years. In order to protect the privacy of outsourced data, uploading the data after encryption is a better way to protect privacy, but this will lose the availability of the data. In order to balance the confidentiality and availability of data and enjoy the benefits of cloud services, searchable encryption came into being.

[0003] Searchable encryption is a primitive whose encrypted ciphertext can be matched with a specific search trap, so that the ciphertext containing specific information can be obtained. If a general encryption scheme is used in a cloud service scenario, since the ciphertext will not leak the plaintext information, after a user uploads a large amount of data to the cloud server, if he wants to search for a keyword in the data, he can only download the data completely from the cloud server to the local computer, decrypt it, and then search the plaintext. This will inevitably bring a lot of communication overhead and computing overhead. If searchable encryption is used, such as the scenario of public key searchable encryption. The data user will generate a public-private key pair and retain the private key. During the encryption process, the data uploader can use the public key to encrypt the keywords in the data, generate a searchable encrypted index, and upload it to the cloud server together with the encrypted document. During the search process, the data user can use the private key to calculate the search trap for the keyword to be searched and upload it to the cloud server. The cloud server will use the search trap to search in the encrypted index. If the result returns a match, the cloud server will return the corresponding data ciphertext to the data user, otherwise it will return the information that does not contain the keyword in the data to the data user. Since the encrypted index will not leak the information of related keywords, the search trap will not leak the request privacy of the data user, and the data uploaded by the data uploader is also encrypted, the server will not obtain the relevant information of the data uploader and the data user, thus protecting the data privacy while ensuring the data availability.

[0004] Time is an important attribute in time-related scenarios. For example, in a sealed auction, the encrypted bids can only be decrypted after the release time. Adding time-dimension access control to the searchable encryption scheme can achieve the property that the corresponding documents can only be searched after the release time, which has very important significance and application scenarios. However, we cannot guarantee that users will search at the set release time. Therefore, the existing searchable encryption scheme with time release search control needs to record all time credentials before the system time in the cloud server so that users can search for documents with release time before the system time. This will bring a lot of additional storage overhead and computational overhead caused by traversing time credentials when searching. In addition, this method does not support efficient modification of index strategies and revocation of user search capabilities during system operation. Summary of the invention

[0005] The purpose of the present invention is to provide an incremental and efficient ciphertext retrieval method with dynamic time attribute control, aiming to improve the functionality and efficiency of existing ciphertext retrieval methods with time release search control.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] An incremental and efficient ciphertext retrieval method with dynamic time attribute control comprises the following steps:

[0008] Step 1, initialization: the time server generates public parameters, a preset time tree and a public-private key pair of the time server, and the data user generates a public-private key pair of the data user;

[0009] Step 2: The time server assists the data uploader in generating an encrypted index and search authorization, and sends them to the cloud server and authorized data users respectively;

[0010] Step 3, the data user receives the time information from the time server and generates a search trap based on the authorization and the time information;

[0011] Step 4, search phase: the data user uploads the generated search trap to the server, and the server executes the search process. If there is a matching result, the corresponding document is returned to the data user; otherwise, the information that there is no document related to the corresponding keyword in the database is returned;

[0012] Step 5, revocation phase: When the data user's search permission needs to be revoked, the time server assists the data uploader to generate a revocation certificate, uploads it to the cloud server, terminates the data user's search process, and revokes his search permission;

[0013] Step 6, extension phase: If the time space set during initialization is exhausted, the time server generates an extension credential and broadcasts it to the cloud server and data user, who then extend the ciphertext and authorized time space.

[0014] Furthermore, the step 2 specifically includes:

[0015] Step 2.1, the data uploader sends verification information and a blinded release time information set to the time server, that is, the path of the release time in the time tree. The time server verifies the identity of the data uploader. If the verification is successful, it assists the data uploader in generating an initial authorization set related to the release time according to the release time information set requested by the data uploader;

[0016] Step 2.2: The data uploader unblinds the initial authorization set, uses it as the set that needs to be verified by the cryptographic dynamic accumulator to calculate the aggregate value, and uploads the aggregate value to the cloud server;

[0017] In step 2.3, the data user verifies the data uploader. If the verification is successful, the data uploader generates an authorization corresponding to the keyword and the relevant release time based on the de-blinded initial authorization set, and sends the authorization to the data user.

[0018] Furthermore, the step 3 specifically includes:

[0019] Step 3.1, the time server uses the public parameters to calculate the update credentials for each time period and broadcasts them to the cloud server and all data users;

[0020] In step 3.2, the data user generates a search trap after the release time and sends the search trap to the server to search the encrypted index.

[0021] Furthermore, the sizes of the revocation certificate, the extension certificate, and the update certificate are constant.

[0022] Furthermore, generating a search trap in step 3.2 specifically includes:

[0023] (1) The data user stays online, receives updated credentials at a specific time, and updates the authorization immediately after receiving the updated credentials; when a search is required, the latest authorization is used to generate a search trap;

[0024] (2) The data user goes online at a specific time and receives updated credentials, retaining only the credentials at the release time and the latest credentials received. When a search is needed, the retained credentials are used to update the authorization, and the latest authorization is used to generate a search trap.

[0025] Furthermore, the step 5 specifically includes:

[0026] Step 5.1, the data uploader sends verification information, revocation type and blinded revocation information to the time server; the time server verifies the identity of the data uploader, and if the verification is successful, it assists the data uploader in generating an initial authorization set related to the revocation information according to the revocation information requested by the data uploader;

[0027] Step 5.2: The data uploader deblinds the initial authorization set, generates corresponding revocation certificates based on the content, and uploads them to the cloud server;

[0028] In step 5.3, the cloud server uses the revocation credentials to revoke the user's search capability or change the search strategy of the encrypted index.

[0029] Compared with the prior art, the present invention has the following positive effects:

[0030] 1. The present invention provides a searchable encryption method with release time search control. The authenticated data uploader can set release time access restrictions for the encryption index according to specific application requirements to achieve access control in the time dimension.

[0031] 2. The present invention reduces the extra storage overhead caused by implementing time release search control;

[0032] 3. The present invention realizes the functionality of permission revocation, policy modification, and time and space expansion;

[0033] 4. The present invention realizes the property of full increment. While ensuring functionality, the size of the voucher is at a constant level, reducing communication overhead and the additional computing overhead brought by realizing functionality. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a system model diagram of a searchable encryption method with time dimension access control.

[0035] Figure 2 The flowchart of the program is shown in Figure 2.

[0036] Figure 3 This is the flowchart of the revocation algorithm.

[0037] Figure 4 The flowchart of the extended algorithm is shown in Figure 2. DETAILED DESCRIPTION

[0038] In order to make the above features and advantages of the present invention more clearly understood, examples are given below and detailed description is given in conjunction with the accompanying drawings.

[0039] Please see Figure 1The incremental efficient ciphertext retrieval method of the present invention mainly includes four participating entities: time server, cloud server, data uploader and data user. Different entities have different functions and responsibilities.

[0040] The time server provides user authentication, assists in generating encrypted indexes and authorization, and publishes update services. It authenticates data uploaders and assists authenticated data uploaders in generating encrypted indexes and authorization, generates and broadcasts update credentials, assists in generating revocation credentials, and generates extended credentials.

[0041] The cloud server provides cloud storage and computing services, stores encrypted indexes and files, searches the encrypted indexes based on the data user's request, returns the search results, and executes the revocation and extension algorithm after receiving the revocation and extension credentials.

[0042] The data uploader is the generator of private data and is a user authenticated by the time server. The user can interact with the time server to generate an encrypted index or revoke the certificate and upload it to the cloud server (using the private key of the data user and using the encryption algorithm to build an encrypted index for the keywords in the file, attaching the release time attribute to these indexes and uploading them to the cloud server with the encrypted file);

[0043] Data users are users of private data and need to apply to data uploaders for search permissions for keywords, receive update information broadcast by the time server at regular intervals, receive update credentials for updates after authorization, calculate search traps and upload them to the cloud server, and search the encrypted index in the cloud server after the corresponding release time.

[0044] In the encryption scheme, the time server registers and authenticates the data uploader. After authentication, the data uploader has the authority to generate and upload encrypted indexes and authorize others to search. Data users who need to search ciphertext on the cloud server need to apply to the data uploader for permission to search the ciphertext they uploaded. After obtaining the authorization from the data uploader, the data user receives the update certificate broadcast by the time server to update the authorization.

[0045] When the scheme is running, we must ensure the security of some private data: when the time server assists the data uploader in generating the initial authorization set, the time server cannot know the information requested by the data uploader, that is, the requested keywords and the corresponding release time information; after the data uploader uploads the encrypted index to the cloud server, the cloud server cannot obtain the plaintext information corresponding to the encrypted index; the search trap generated by the data user will not leak any information about the relevant keywords; unauthorized users cannot collude with the cloud server to generate legal authorization; authorized users cannot effectively search the encrypted index before the release time; revocation credentials and extended credentials cannot be forged, otherwise the data stored in the cloud server may be maliciously tampered with.

[0046] Aiming at the different needs of four different participating roles and their possible malicious behaviors, the present invention designs an incremental and efficient ciphertext retrieval method with dynamic time attribute control, including the following steps:

[0047] Step 1: The time server generates public parameters, a preset time tree and a public-private key pair of the time server, and the data user generates a public-private key pair of the data user.

[0048] Step 2: The time server assists the data uploader in generating an encrypted index and search authorization, and sends them to the cloud server and the authorized data user respectively. The authorization process includes the following sub-steps:

[0049] Step 2.1, the data uploader sends verification information and a blinded release time information set to the time server, that is, the path of the release time in the time tree. The time server verifies the identity of the data uploader. If the verification is successful, it will assist the data uploader in generating an initial authorization set for the release time based on the release time information set requested by the data uploader.

[0050] In step 2.2, the data uploader unblinds the initial authorization set, uses it as the set that needs to be verified by the cryptographic dynamic accumulator to calculate the aggregate value, and uploads the aggregate value to the cloud server.

[0051] In step 2.3, the data user verifies the data uploader. If the verification is successful, the data uploader generates authorization based on the de-blinded initial authorization set, that is, the member certificate corresponding to the release time in the path, and sends the authorization to the data user.

[0052] Step 3: The data user receives the time information from the time server and generates a search trap based on the authorization and the time information. The steps include:

[0053] Step 3.1: The time server uses the public parameters to calculate the update credentials for each time period and broadcasts them to the cloud server and all data users.

[0054] In step 3.2, the data user can generate a trapdoor after the release time and send it to the server to search the encrypted index. The trapdoor generation algorithm is divided into two types according to different scenarios: (1) The data user remains online for a long time, receives updated credentials at a specific time, and updates the authorization immediately after receiving the updated credentials each time. When a search is required, the latest authorization is used to generate a search trapdoor. (2) The data user goes online at a specific time and receives the updated credentials, retaining only the credentials at the release time and the latest credentials received. When a search is required, the retained credentials are used to update the authorization, and the latest authorization is used to generate a search trapdoor.

[0055] Step 4, search method, the data user uploads the generated search trap to the server, the server executes the search process, and if there is a matching result, the corresponding document is returned to the data user; otherwise, the information that there is no corresponding keyword-related document in the database is returned.

[0056] Step 5: When this solution is running, some data users may lose trust, and the data uploader may want to revoke the data users' permission to search for certain keywords. Therefore, the solution provides a revocation method that can make the file unable to be indexed or revoked to a later time. The time server assists the data uploader in generating a revocation certificate and uploads it to the cloud server. The generation process includes the following sub-steps:

[0057] Step 5.1, the data uploader sends verification information, revocation type, and blinded revocation information to the time server, that is, the path between the previous release time and the release time you want to set on the time tree. The time server verifies the identity of the data uploader. If the verification is successful, it will assist the data uploader in generating an initial authorization set for revocation information based on the revocation information requested by the data uploader.

[0058] In step 5.2, the data uploader deblinds the initial authorization set, generates corresponding revocation credentials based on the content, and uploads them to the cloud server.

[0059] In step 5.3, the cloud server uses the revocation credential to execute the revocation algorithm to revoke the user's search ability or change the search strategy of the encrypted index.

[0060] Step 6, extension method, the time server generates an extension credential when the time space is exhausted, broadcasts it to the cloud server and data users, and the cloud server and data users execute the extension algorithm to extend the ciphertext and the authorized time space.

[0061] The present invention draws on the idea of ​​cryptographic dynamic accumulators: in terms of search, the search strategy (i.e., all time after the release time) is aggregated, reducing the additional storage overhead caused by the introduction of time dimension search control; in terms of functionality, since the scheme retains the computability of the encrypted index, the data uploader can calculate the encrypted index with the voucher and change the search strategy of the encrypted index to achieve the purpose of strategy change, revocation, and time and space expansion. We use a full binary tree of post-order traversal to represent the linear time space, which can represent the system time while retaining the minimum voucher, reduce the number of elements that may need to be proved for membership, and further reduce the storage overhead and the computational overhead during the search; at the same time, the root node of the binary tree can be set as keyword information, and the keyword information can be bound to the access strategy to prevent combined attacks.

[0062] In terms of security, the updated credentials, revoked credentials, extended credentials, and authorizations in the scheme are all impossible to forge. In terms of efficiency, the scheme achieves the property of full increment, that is, the size of the credentials is constant, and only entities holding authorizations or encrypted indexes can perform update, revocation, extension, and search operations. For the search process, the updated credentials ensure that the authorizations held by the user can be searched at any time after the release time, so that the time server does not need to save authorizations for all time periods, reducing the storage and computing overhead of the time server; for the revocation and extension functions, the data uploader does not need to recalculate the encrypted index after the policy is changed, and the time server does not need to regenerate the system after the time space is exhausted, reducing the computing overhead required to implement the revocation and extension functions.

[0063] Example

[0064] An incremental and efficient ciphertext retrieval method with dynamic time attribute control. During initialization, the time server generates a public parameter pp according to the security parameter λ, and the time tree T and its public-private key pair (pk ts ,sk ts ), the data user generates his public-private key pair (pk u ,sk u ). After registering and authenticating with the time server, the data uploader can generate an encrypted index CT and corresponding authorization Aut about specific keywords and their release time after interacting with the time server, and upload the ciphertext to the cloud server. The data user applies to the data uploader for search permission for specific keywords, and the data uploader sends the corresponding authorization to the data user. The time server uses its own private key sk ts Calculate and update voucher tkn upd, and broadcast it to the cloud server and users. The information broadcast by the time server is not related to the user's private information and is not generated for a specific user, so it can be broadcast publicly, and any user can update their authorization by updating the certificate. Data users receive specific update certificates to update authorization, and use the keyword W to be searched, the corresponding authorization Aut and their own private key sk when searching. u Generate search trapdoor Tr W The cloud server receives the search trap and matches it with the encrypted index CT. If the corresponding keywords are the same and the current time is after the corresponding release time, the corresponding document is returned. In the revocation phase, the data uploader determines the keywords to be revoked and the time to be revoked, and interacts with the time server to generate the corresponding revocation certificate tkn re Upload to the cloud server, the cloud server executes the revocation algorithm to change the corresponding encryption index. In the extension phase, the time server determines the time period that needs to be extended. When the time period is reached, the time server uses its own private key sk ts Generate extended credentials.

[0065] The following is the algorithm flow Figure 2 Further detailed description of the incremental time-release based searchable encryption algorithm:

[0066] 1. System initialization, the time server generates public parameters pp, time tree T, and generates the public and private key pair of the time server (pk ts ,sk ts ), the data user generates the public-private key pair (pk u ,sk u ).

[0067] 2. After the data uploader registers and authenticates his identity at the time server, he will interact with the time server as follows:

[0068] a. The data uploader sets the release time t for keyword W r , get t r All nodes on the path in the time tree. Calculate the blinded values ​​for all nodes and send them to the time server.

[0069] b. The time server uses its own private key sk ts Each blinded value is signed and the result is sent to the data uploader.

[0070] c. The data uploader receives the information and unblinds it, calculating its hash value H.

[0071] 3. The data uploader uses the data user’s public key and the information obtained from the interaction with the time server to calculate the encryption index CT and authorization Aut, uploads the encryption index to the cloud server, sends the authorization to the requesting data user, and retains the information c about the encryption index random number and the data user’s public key.

[0072] 4. The time server uses the private key sk ts Calculate the specific information of each period and calculate its hash value H t Broadcast to cloud servers and data users.

[0073] 5. Data users receive H t And calculate it with Aut to get the latest authorized Aut t .

[0074] 6. Data user uses keyword W and private key sk u And the latest authorized Aut t Calculate the search trapdoor Tr W And upload it to the cloud server.

[0075] 7. Cloud server receives search trap Tr W , run the search algorithm, obtain relevant documents containing keywords, and return the results to the data user.

[0076] The following is the algorithm flow Figure 3 and Figure 4 Describe the revocation algorithm and extension algorithm in the scheme:

[0077] 8. The data uploader can choose two types of revocation according to different needs. If the index needs to be revoked to a certain time in the future, the keyword W to be revoked and the previous release time t are determined. r The time to cancel If the index needs to be completely revoked, a random number x is generated, which is blinded and passed to the time server. The time server signs it and sends it to the data uploader. After the data uploader deblinds the returned information, the retention information c related to the keyword W and document D that needs to be revoked calculates the revocation certificate and uploads it to the cloud server. The cloud server executes the revocation algorithm and changes the information of the relevant encrypted index.

[0078] 9. After the current time reaches the right child node of the root node of the time tree, the time server uses its private key sk ts Calculate the extended credentials and broadcast them to the cloud server and data users. The cloud server aggregates the extended credentials into the ciphertext to complete the ciphertext update. The user uses the extended credentials as the update credentials and updates the authorization according to the update phase method.

[0079] In the above scheme, the time server registers and authenticates the data uploader, restricting the data uploader to generate a valid encrypted index and granting the data uploader the right to add time dimension access control and authorized user search permissions to the encrypted index.

[0080] This embodiment provides an incremental efficient ciphertext retrieval device, comprising: a memory, a processor, and a computer program stored in the memory, wherein the processor implements the steps of the incremental efficient ciphertext retrieval method when executing the computer program.

[0081] This embodiment provides a computer storage medium, wherein the computer storage medium stores an executable program, and the executable program is executed by a processor to implement the steps of the incremental efficient ciphertext retrieval method.

[0082] The present invention proposes an incremental and efficient ciphertext retrieval method with dynamic time attribute control. The data uploader can encrypt the file and set the keyword search release time according to his own needs, that is, the file related to the keyword can only be searched after this time, and the encrypted file is outsourced to the cloud server that provides cloud storage and computing services. Data users who are authorized to search for a certain keyword are allowed to search for the keyword on the cloud server after the corresponding search release time. While ensuring the confidentiality and availability of the data, the access rights of the data user are controlled in the time dimension.

[0083] The present invention allows data uploaders to add dynamic time dimension access control to encrypted indexes, that is, data users can only effectively search encrypted indexes after a specified time point, and the method supports data uploaders to modify the time access policy corresponding to the generated encrypted indexes. This solution can ensure the availability of data while protecting the privacy of uploaded data and the privacy of user requests.

Claims

1. An incremental and efficient ciphertext retrieval method with dynamic time attribute control, characterized in that: Includes steps: Step 1, initialization: the time server generates public parameters, a preset time tree and a public-private key pair of the time server, and the data user generates a public-private key pair of the data user; Step 2: The time server assists the data uploader in generating an encrypted index and search authorization, and sends them to the cloud server and authorized data users respectively; Step 3, the data user receives the time information from the time server and generates a search trap based on the authorization and the time information; Step 4, search phase: the data user uploads the generated search trap to the server, and the server executes the search process. If there is a matching result, the corresponding document is returned to the data user; otherwise, the information that there is no document related to the corresponding keyword in the database is returned; Step 5, revocation phase: When the data user's search permission needs to be revoked, the time server assists the data uploader to generate a revocation certificate, uploads it to the cloud server, terminates the data user's search process, and revokes his search permission; Step 6, extension phase: If the time space set during initialization is exhausted, the time server generates an extension credential and broadcasts it to the cloud server and data user, who then extend the ciphertext and authorized time space.

2. According to claim 1, an incremental efficient ciphertext retrieval method with dynamic time attribute control is characterized in that: The step 2 specifically includes: Step 2.1, the data uploader sends verification information and a blinded release time information set to the time server, that is, the path of the release time in the time tree. The time server verifies the identity of the data uploader. If the verification is successful, it assists the data uploader in generating an initial authorization set related to the release time according to the release time information set requested by the data uploader; Step 2.2: The data uploader unblinds the initial authorization set, uses it as the set that needs to be verified by the cryptographic dynamic accumulator to calculate the aggregate value, and uploads the aggregate value to the cloud server; In step 2.3, the data user verifies the data uploader. If the verification is successful, the data uploader generates an authorization corresponding to the keyword and the relevant release time based on the de-blinded initial authorization set, and sends the authorization to the data user.

3. The incremental efficient ciphertext retrieval method with dynamic time attribute control according to claim 1, characterized in that: The step 3 specifically includes: Step 3.1, the time server uses the public parameters to calculate the update credentials for each time period and broadcasts them to the cloud server and all data users; In step 3.2, the data user generates a search trap after the release time and sends the search trap to the server to search the encrypted index.

4. The incremental efficient ciphertext retrieval method with dynamic time attribute control according to claim 1, characterized in that: The sizes of the revocation certificate, the extension certificate, and the update certificate are constant.

5. The incremental efficient ciphertext retrieval method with dynamic time attribute control according to claim 1, characterized in that: Generating a search trap in step 3.2 specifically includes: (1) The data user stays online, receives updated credentials at a specific time, and updates the authorization immediately after receiving the updated credentials; when a search is required, the latest authorization is used to generate a search trap; (2) The data user goes online at a specific time and receives updated credentials, retaining only the credentials at the release time and the latest credentials received. When a search is needed, the retained credentials are used to update the authorization, and the latest authorization is used to generate a search trap.

6. The incremental efficient ciphertext retrieval method with dynamic time attribute control according to claim 1, characterized in that: The step 5 specifically includes: Step 5.1, the data uploader sends verification information, revocation type and blinded revocation information to the time server; the time server verifies the identity of the data uploader, and if the verification is successful, it assists the data uploader in generating an initial authorization set related to the revocation information according to the revocation information requested by the data uploader; Step 5.2: The data uploader deblinds the initial authorization set, generates corresponding revocation certificates based on the content, and uploads them to the cloud server; In step 5.3, the cloud server uses the revocation credentials to revoke the user's search capability or change the search strategy of the encrypted index.

7. An incremental and efficient ciphertext retrieval device, characterized in that: include: A memory, a processor and a computer program stored in the memory, wherein the processor implements the steps of the incremental efficient ciphertext retrieval method according to any one of claims 1 to 6 when executing the computer program.

8. A computer storage medium, characterized in that: The computer storage medium stores an executable program, and the executable program is executed by a processor to implement the steps of the incremental efficient ciphertext retrieval method according to any one of claims 1 to 6.

Images