Storage proof method for fast location and recovery of expired files for dynamic group users
By using the key generation protocol and RS encoding in cloud storage to generate verification tags, combined with ETM and RDI algorithms, the problem of failed file location and recovery of dynamic group users in cloud storage is solved, data availability and audit efficiency are improved, and the data is reduced. Overhead.
Patent Information
- Application Number
- CN202110965565.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-23
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2041-08-23
AI Technical Summary
The prior art cannot quickly locate and recover dynamic group user failure files in cloud storage, especially after failure of data integrity audits, resulting in inefficiency and data unavailability.
A storage proof method for fast positioning and recovery of failed files of dynamic group users is adopted, and a signature key is generated through the key generation protocol, a verification tag is generated using RS encoding, data integrity audit is implemented, and ETM and RDI algorithms are used after the audit fails. Quickly locate and restore invalid files.
It realizes rapid positioning and recovery of failed files in dynamic group user scenarios, improves data availability and audit efficiency, and reduces time and communication overhead for users and cloud providers.
Smart Images

Figure CN113821831B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of cloud storage security technology, and in particular relates to a storage certification method for fast positioning and recovery of expired files of dynamic group users. Background Art
[0002] At present, with the rapid development of digital technologies such as the Internet of Things, cloud computing, and big data, the exponential growth of data has brought severe challenges to limited local storage resources. Cloud storage solves this pressure by providing remote but low-cost storage options, which reduces the storage management and maintenance burden of users while allowing them to access their data anytime and anywhere. But this also brings security issues for data stored on cloud servers. Among them, secure storage proof is a key aspect. Ateniese et al. first proposed the concept of "provable data ownership" PDP in [Ateniese et al.: Provable data possession at untrusted stores, CCS 2007.]. Shacham et al. [Shacham et al.: Compact proofs of retrievability, ASIACRYPT 2008.] Based on Ateniese et al.'s idea of homomorphic tags, Shacham et al. used the BLS short message signature mechanism to construct homomorphic verification tags, reducing the communication overhead in the verification phase. Jules et al. [A.Juels et al.: Pors: proofs of retrievability for large files, CCS2007.] first considered how to recover damaged data and proposed a sensor-based POR scheme, which can not only perform data integrity audits but also recover faulty data to a certain extent.
[0003] Although the above schemes can easily detect the integrity of data, little attention is paid to invalid file query after audit failure. Huang et al. [M.Hwang et al.: Data error locations reported by public auditing in cloud storage service, Autom.Softw.Eng 2014] proposed a matrix-based fault file query method, which places files in the form of a matrix, batch audits the same rows and columns, and determines the failed files at the intersection of the faulty rows and faulty columns. Shin et al. [S.Shin et al.: Identification of corrupted cloud storage in batch auditing for multi-cloud environments, ICT-EurAsia / CONFENIS 2015] proposed index test-based search (ETS), which identifies the location of invalid files / blocks by comparing verification values with and without indexes. However, ETS is only applicable to the failure of individual files. Wang et al. [H.-F.Wang et al.: Batch auditing supporting fast searching invalid files in cloud storage, CHINESE JOURNAL OFCOMPUTERS 2017] proposed a batch audit method that supports fast query of invalid files. This method is characterized by being able to effectively resist "invalid file" attacks. After the data integrity audit fails, finding the location of the invalid file / block will bring a lot of computing and communication overhead, which seriously affects the efficiency of the solution. Although the above solution can locate the location of the invalid file, it cannot restore the invalid data. For users, data availability is extremely important, and invalid data cannot guarantee availability.
[0004] In cloud storage, users can further reduce storage costs by sharing data with a group of users, thereby alleviating the burden of local data storage. When a group user behaves improperly or leaves the group, the user should be revoked from the group, and the revoked user's data is managed by the legitimate user designated by the group administrator. Recently, Zhang et al. [Y. Zhang et al.: Enabling efficient user revocation in identity-based cloud storage auditing for shared big data, TDSC 2020 (EUISD)] proposed an efficient user revocation data integrity audit scheme, which effectively improved the audit efficiency of user revocation. However, the scheme does not consider data integrity audit failures. Once the data integrity audit fails, the location of the invalid file should be quickly retrieved and the invalid file should be restored. However, the above invalid file / block identification scheme is mainly designed for single-client scenarios. More precisely, the audit tags from multiple users are calculated with different private keys, which makes it difficult and inefficient to aggregate authentication tags and identify files / blocks in one challenge.
[0005] Therefore, there is an urgent need for a new storage proof method for quickly locating and recovering expired files of dynamic group users to make up for the problems existing in the existing technology.
[0006] Through the above analysis, the problems and defects of the prior art are as follows:
[0007] (1) Existing technologies cannot quickly locate invalid files / blocks and recover them after an integrity audit fails.
[0008] (2) The prior art does not apply the invalid file location and recovery solution after integrity audit failure to the dynamic group user scenario with user revocation.
[0009] The difficulty in solving the above problems and defects is: how to ensure the security and efficiency of invalid file location and recovery after data integrity audit fails in a user revocation scenario.
[0010] The significance of solving the above problems and defects is:
[0011] (1) A solution for fast location and recovery of expired files is proposed for group users, which extends the integrity audit of the POR scheme, which is outlined in the data integrity audit of the scheme of the present invention, and solves the security and efficiency limitations that have not been covered in the relevant proofs of storage schemes for group users.
[0012] (2) A specific data invalidation file fast location and recovery scheme for group users is described, called CIRG, which is safe in the security model of the present invention. CIRG uses a scheme based on efficient user revocation. In this way, CIRG has the advantage of making the dynamic group user scheme more complete.
[0013] (3) A CIRG-based prototype is implemented and evaluated in a realistic cloud environment. Experiments show that the proposed solution incurs tolerable overhead for both users and cloud providers compared to related schemes. Summary of the invention
[0014] In view of the problems existing in the prior art, the present invention provides a storage proof method for quickly locating and recovering expired files of dynamic group users.
[0015] The present invention is implemented as follows: a storage proof method for fast positioning and recovery of expired files of dynamic group users, the storage proof method for fast positioning and recovery of expired files of dynamic group users comprises the following steps:
[0016] Step 1: Generate the corresponding signing key according to the key generation protocol; generating the key is a preparatory work for generating the verification tag in step 2.
[0017] Step 2: RS encode the file and generate a verification tag for the encoded data; RS encoding is used for data recovery, and the generated verification tag is used for data integrity verification.
[0018] Step three: Implement data integrity audit; this is the main function of the solution and is used to verify the integrity and correctness of the data stored by users on the cloud.
[0019] Step 4: Locate and restore invalid files after integrity audit fails. When integrity verification fails, search and restore invalid files to ensure file integrity, which is also the innovation of the present invention.
[0020] Further, in step 1, generating a corresponding signing key according to a key generation protocol includes:
[0021] We apply the EUISD data tag generation method, use Schnorr signature to calculate the identity key and partial key, PKG generates the identity key, the group administrator generates the partial key, and the group user generates the partial key based on the identity key IDK. ID =(R ID , σ ID ) and the partial key TK ID,RN =(R RN , σ RN )(where RN = 0) calculates the private key. The group user calculates σ = (σ ID +σRN )modq, get the private key SK ID , RN Yes (R ID , R RN , σ).
[0022] Furthermore, in step 2, the RS encoding of the file and generating a verification tag for the encoded data include:
[0023] Run by a group of users, taking F as input, where F is a block {m i}, the user performs RS encoding on the data and outputs the encoded data; and uploads the encoded data to the cloud server; wherein the encoding algorithm is as follows:
[0024] (1) Divide the file F into F = m1, m2, ..., m k , according to RS erasure code, data m1, m2, ..., m k Encode; the encoded data is E = ω1, ω2, ..., ω n ; Where n = red + k, red is the redundancy, E = x * F T , x is a matrix of k linearly independent vectors of length n, x = x1, x2, ..., x k , x i ∈2 ω (finite field), n≤2 ω +1. The detailed description is as follows:
[0025]
[0026] (2) Generate a tag for the encoded data: Use the authentication private key σ and file E, that is, block {m j} is the input, and the output data block verification tag set in N is the number of auditing users.
[0027] Furthermore, in step 3, the implementation of data integrity audit includes three steps: initiating a challenge, generating a proof, and verifying. TPA generates an audit challenge for the cloud server, the cloud server generates a corresponding proof to prove that it has complete cloud data, and TPA verifies data integrity based on the proof returned by the cloud server.
[0028] (1) Challenge algorithm: TPA generates audit challenge chal = {C1, C2, …, C N}Sent to the cloud, where i∈[1,N],j∈[1,s],s is a random number, s∈[1,n].
[0029] (2) Proof Algorithm: In this algorithm, the cloud server generates a corresponding proof to prove that it owns the complete cloud data. After receiving the audit query chal from the TPA, the cloud server generates a data possession proof P = (T, μ), where T = {T1, …, T N}, i∈[1,N], where
[0030] (3) Verification algorithm:
[0031] ①TPA verifies the correctness of proof P by checking whether the following equation holds:
[0032]
[0033] in If this equation holds true, it means that all files are intact and returns 1; otherwise, it means that there is at least one invalid file and returns 0, calling the search function to find the invalid file.
[0034] ② When TPA fails to perform data integrity review, TPA quickly locates the invalid file location based on ETM and RDI, and returns the searched invalid file location to the group user.
[0035] Further, in step 4, after the integrity audit fails, locating and recovering invalid files includes:
[0036] When the data integrity audit fails, TPA quickly locates the failed data file, and after querying the location of the invalid file, returns a list of invalid files to the group of users; based on the results returned by TPA, the user restores the invalid file.
[0037] (1) Search for invalid file SIF
[0038] SIF is a function composed of the ETM algorithm and the RDI algorithm. Call the ETM function to quickly find the location of an invalid file. Call the RDI function to quickly locate the locations of multiple invalid files. The rapid location of invalid files needs to take into account two types of attacks. Both a single invalid file and multiple invalid files can cause data integrity audit failures, so there are single invalid file attacks and multiple invalid file attacks.
[0039] (2) Index Estimation Method ETM
[0040] The index test quickly identifies a single invalid file from a batch audit file by calculation. The ETM is used to solve the problem of locating Schnorr signature type invalid signatures in bilinear mapping batch audits.
[0041] If a single invalid file is found by using ETM and then audited, if the audit is successful, the attack by the single invalid file is effectively resisted.
[0042] The ETM algorithm process is as follows:
[0043] ①Batch review:
[0044]
[0045] ②Calculate the signature:
[0046]
[0047] ③ Index calculation: Calculate the verification value of the equation, if Indicates that there are no invalid files. Then the L file is invalid; Indicates that there are at least two invalid files;
[0048] If no single invalid file is found, call RDI to quickly locate multiple invalid files. When calling RDI, enter a batch file, perform invalid file location, and output a location list of all invalid files.
[0049] (3) Quick search for invalid files, associated binary search, RDI
[0050] The RDI algorithm process is as follows:
[0051] ① Batch verification: calculation if Output n bits (1, 1, ..., 1) and stop;
[0052] ②Assume X [1,N] Is the root node or the left child node, calculate
[0053] ③If X [1,N] Node invalid, set BAF(X [1,N] )≠0, calculate X L The left child node of and return;
[0054] ④Calculate the checksum of the right child node: If Then set if Then set Otherwise calculate
[0055] ⑤ Recursively apply the same algorithm to the left and right subtrees, and finally output (b1, b2, ..., b N ).
[0056] (4) Data recovery
[0057] After SIF quickly finds the invalid file list, it uses RS erasure code to recover the original data. Since RS erasure code has the maximum distance split (MDS) property, this means that data F can be recovered from any k valid data blocks in E. In other words, when the number of damaged data blocks is less than red, the original data F can be recovered. Assume that the number of damaged blocks in E is currently less than red. According to the invalid file list located by RDI, the user obtains k valid data blocks E′ from E. The formula for recovering data is: F T = x′ -1 *E′. The specific process is as follows:
[0058]
[0059]
[0060] Combined with all the above technical solutions, the advantages and positive effects of the present invention are as follows: the present invention combines the invalid file fast location and recovery scheme for the first time, and applies it to the multi-user data integrity audit with user revocation, which makes up for the deficiency that the invalid files cannot be quickly located and restored after the audit fails in the multi-user data integrity audit. Compared with an efficient user revocation data integrity audit scheme proposed by Zhang et al. [Y.Zhang et al.: Enabling efficient user revocation inidentity-based cloud storage auditing for shared big data, TDSC 2020], the present invention solves the problem of invalid file location and recovery after the integrity audit fails. Compared with the invalid data location scheme, the present invention provides an invalid data recovery function to recover the losses of users and cloud service providers. Compared with the scheme with data redundancy, the present invention helps to quickly query the invalid file location. In addition, the present invention is user- and cloud provider-friendly, and the time overhead and communication overhead generated by the scheme are small. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0062] Figure 1 It is a flow chart of a storage proof method for quickly locating and restoring expired files of dynamic group users provided by an embodiment of the present invention.
[0063] Figure 2 It is a structural block diagram of a storage proof system for fast location and recovery of invalid files of dynamic group users provided by an embodiment of the present invention;
[0064] In the figure: 1. Key generation module; 2. Data encoding module; 3. Integrity audit module; 4. Positioning and recovery module.
[0065] Figure 3 This is a storage proof system model diagram for fast location and recovery of failed files of dynamic group users provided by an embodiment of the present invention.
[0066] Figure 4 The figure is a flow chart of an invalid file location and recovery solution provided by an embodiment of the present invention.
[0067] Figure 5 It is a data coding diagram provided by an embodiment of the present invention.
[0068] Figure 6 It is a schematic diagram of the association dichotomy method provided by an embodiment of the present invention.
[0069] Figure 7 It is a schematic diagram of searching for invalid files provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0070] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0071] In view of the problems existing in the prior art, the present invention provides a storage proof method for quickly locating and recovering expired files of dynamic group users. The present invention is described in detail below in conjunction with the accompanying drawings.
[0072] like Figure 1 As shown, the storage proof method for fast positioning and recovery of invalid files of dynamic group users provided by the embodiment of the present invention includes the following steps:
[0073] S101, generating a corresponding signature key according to a key generation protocol;
[0074] S102, performing RS encoding on the file and generating a verification tag for the encoded data;
[0075] S103, implement data integrity audit;
[0076] S104: After the integrity audit fails, locate and recover invalid files.
[0077] like Figure 2As shown, the storage proof system for fast positioning and recovery of invalid files of dynamic group users provided by the embodiment of the present invention includes:
[0078] The key generation module 1 is used to generate a corresponding signature key according to the key generation protocol;
[0079] Data encoding module 2, used for performing RS encoding on the file and generating a verification tag for the encoded data;
[0080] Integrity audit module 3, used to implement data integrity audit;
[0081] The positioning and recovery module 4 is used to locate and recover invalid files after the integrity audit fails.
[0082] The technical solution of the present invention is further described below in conjunction with specific embodiments.
[0083] Another object of the present invention is to provide a key generation scheme: generating corresponding signature keys. The present invention uses the well-known Schnorr signature to calculate the identity key and the partial key, and then generates the signature key. In this algorithm, the PKG generates the identity key, the group administrator generates the partial key, and the group users use the identity key and the partial key to generate their private keys. The details of the key generation are shown in Table 1:
[0084] (1) When PKG receives the group ID, it selects a random number And calculate and σ ID =r ID +xH1(ID,R ID ) mod q. Then, PKG sends the identity key IDK to the group user. ID =(R ID ,σID).
[0085] (2) After receiving the identity key, the group user verifies the IDK ID If The group of users accepts the identity key IDK ID ; otherwise, reject it.
[0086] (3) First, the group administrator sets the user revocation number RN to 0 and sends it to the group users and the cloud. Then, the group administrator selects a random number And calculate and σ RN =r RN +yH2(ID,R RN , RN) mod q. The group administrator will use the partial key TK ID,RN =(R RN , σ RN) is sent to group users, where RN is initially set to 0. When a user leaves the group, RN = RN + 1, thus recursively.
[0087] (4) After receiving the partial key, the group user verifies TK ID,RN If Then the group user accepts the partial key TK ID,RN ; otherwise, reject.
[0088] (5) Then, the group users use the identity key IDK ID =(R ID , σ ID ) and the partial key TK ID,RN =(R RN , σ RN )(where RN = 0) calculates the private key. The group user calculates σ = (σ ID +σ RN )mod q, get the private key SK ID,RN Yes (R ID , R RN , σ).
[0089] Table 1 Signature key generation process
[0090]
[0091] Another object of the present invention is to provide a data encoding scheme: the algorithm is run by a group of users. It takes F as input, F is a block {m i}, the user performs RS encoding on the data and then uploads the encoded data to the cloud server. The encoding algorithm details are as follows Figure 5 As shown:
[0092] Run by a group of users, taking F as input, where F is a block {m i}, output the encoded and obfuscated data; the user performs RS encoding on the data and uploads the encoded data to the cloud server; wherein the encoding algorithm is as follows:
[0093] 1) Divide the file F into F = m1, m2, ..., m k , according to RS erasure code, data m1, m2, ..., m k Encode; the encoded data is E = ω1, ω2, ..., ω n ; Where n = red + k, red is the redundancy, E = x * F T , x is a matrix of k linearly independent vectors of length n, x = x1, x2, ..., x k , x i ∈2 ω (finite field), n≤2 ω+1. The detailed description is as follows:
[0094]
[0095] 2) Then, the present invention generates a label for the encoded data. It takes the authentication private key σ and file E as input and outputs a data block verification label set in FID is the file identifier, i is the i-th user, J is the data block index, u is a random value, u∈G1, and RN is the number of times the user has revoked. Group users calculate the file tag tag = FID||RN||R ID ||R RN ||T, where τ = SSig ssk (FID||RN||R ID ||R RN ), where ssk is the file tag private key.
[0096] Another object of the present invention is to provide an integrity audit scheme: the integrity audit algorithm includes three steps: initiating a challenge, generating a proof, and verifying. TPA generates an audit challenge for the cloud server, the cloud server generates a corresponding proof to prove that it has complete cloud data, and TPA verifies the data integrity based on the proof returned by the cloud server. The integrity audit scheme is shown in Table 2:
[0097] 1) Challenge algorithm: TPA generates the following audit challenge chal:
[0098] ① Randomly select a set I containing N·s elements, where I={I1,…,I S ,…,N1,…,N s}, where s∈[1,n], N represents the number of audited users.
[0099] ②For each i s ∈I, i∈[1,N] generates a random value
[0100] ③ Set the audit query chal = {C1, C2, ..., C N}Sent to the cloud, where i∈[1,N],j∈[1,s].
[0101] 2) Proof Algorithm: In this algorithm, the cloud server generates a corresponding proof to prove that it owns the complete cloud data. After receiving the audit query chal from the TPA, the cloud server generates a data ownership proof as shown below:
[0102] ① Calculate T = {T1, ..., T N},in μ={μ1,…,μ N}, i∈[1,N], where
[0103] ② Calculate P, where P = (T, μ).
[0104] ③Send P and the file label together to TPA as evidence.
[0105] 3) Verification algorithm:
[0106] ①TPA first retrieves the file tag and verifies the validity of the file tag by checking whether τ is a valid signature through the ID. If so, TPA receives the data returned by the server and verifies whether the server has correctly stored the client's files in the form of batches. TPA verifies the correctness of proof P by checking whether the following equation holds:
[0107] ②
[0108] ③Among them If this equation holds true, it means that all files are intact and returns 1; otherwise, it means that there is at least one invalid file and returns 0, calling the search function to find the invalid file.
[0109] ④When TPA fails to perform data integrity review, TPA will quickly locate the invalid file location based on ETM and RDI, and return the searched invalid file location to the group user.
[0110] Table 2 Data integrity verification protocol
[0111]
[0112] Another object of the present invention is to provide an invalid file location and recovery solution: once the data integrity audit fails, TPA quickly locates the failed data file, and after finding the location of the invalid file, returns the invalid file list to the group of users. Based on the result returned by TPA, the user can restore the invalid file.
[0113] 1) Search for invalid files (SIF)
[0114] SIF is a function composed of the ETM algorithm and the RDI algorithm. The rapid location of invalid files needs to consider single invalid file attacks, that is, cloud server attacks provide a single invalid file or server failures damage files, resulting in batch audit failures, so the invalid file search function is called to increase audit overhead. Therefore, before batch audits, the ETM method is used to audit single invalid files to prevent malicious servers from deliberately causing file damage or server failures from causing single invalid file attacks. If a single invalid file is not found, it means that there are multiple invalid files, and RDI is called to quickly locate the locations of multiple invalid files.
[0115] Next, the details of the ETM and RDI algorithms of the present invention are introduced.
[0116] 2) Exponential Estimation Method (ETM)
[0117] Index testing can quickly identify single invalid files from a batch of audit files by means of calculation.
[0118] Definition 1: Batch audit value "P = Q", the calculation function of the verification value is ETM(·), ETM(·) = P / Q
[0119] Definition 2: Checksum IFN ETM For the check value of i, the calculation function of the check value is ETM(·)
[0120] For user i, the index test value is as follows:
[0121]
[0122] Verification value of the document evidence index The calculation is as follows:
[0123]
[0124] in X is the audited file.
[0125] If a single invalid file is found by using ETM and then audited, if the audit is successful, the attack by the single invalid file is effectively resisted.
[0126] The details of the ETM algorithm are shown in Table 3.
[0127] If no single invalid file is found, call RDI to quickly locate multiple invalid files. When calling RDI, enter a batch file, perform invalid file location, and output a location list of all invalid files.
[0128] Table 3 Details of the ETM algorithm
[0129]
[0130] 3) Quick search for invalid files (Relational Dichotomy, RDI)
[0131] In order to make full use of the intermediate results in the query process and reduce the number of batch audits, a fast search method RDI based on elliptic curve signature is proposed to establish the correlation between audits and transform non-correlated audits into correlated audits.
[0132] On the basis of the efficient revocation scheme of group users, a batch file auditing scheme based on discrete logarithm is constructed, which can quickly locate the location of invalid files.
[0133] Definition 3: Batch audit value "P=Q", the calculation function of the verification value is RDI(·), RDI(·)=PQ.
[0134] Using the bilinear mapping property, the audit value comparison method: "P = Q" is converted into a 0 value judgment: IFN RDI =PQ,
[0135] in Verify value IFN RDI is the value of the above formula, and the batch audit value is expressed as if The batch review is passed, indicating that the data is complete; if There is at least one invalid file. For N users' files, the RDI batch audit formula is:
[0136]
[0137] where θ and v are the same as above.
[0138] The product of the verification values of multiple consecutive file evidence pairs (1→N) is expressed as According to the above formula, set X [1,N] The verification value of the document evidence pair is
[0139] The value 0 judgment method establishes the correlation between the parent node and the left and right child nodes. Figure 6 As shown, the verification value of the parent node in the binary search tree is equal to the sum of the verification values of the left and right child nodes.
[0140] like Figure 7 As shown, gray represents invalid files. The location of invalid files can be quickly located based on the correlation dichotomy. According to the calculation method of the left and right subtrees, the values of the left and right subtrees are associated, and the value of the other subtree can be derived from the values of the left and right subtrees. In the formula below, by comparing the checksum values of the parent node and the left child node, the validity of the right child node can be directly determined without batch review.
[0141]
[0142] The verification value of each node is calculated based on the batch audit function BAF(X) based on the value 0. The algorithm outputs an n-bit string (b1, b2, ..., b N ). If and only if file i is valid, b i =1 has a very high probability.
[0143] The details of the RDI algorithm are shown in Table 4.
[0144] According to the above analysis, the RDI method based on value 0 eliminates the batch review process of the right child node, thereby reducing the number of batch reviews.
[0145] Table 4 Details of the RDI algorithm
[0146]
[0147] 4) Data recovery
[0148] After SIF quickly finds the invalid file list, it uses RS erasure code to restore the original data. Since RS erasure code has the maximum distance divisibility (MDS) property, this means that data F can be recovered from any k valid data blocks in E. In other words, when the number of damaged data blocks is less than red, the original data F can be restored. Assume that the number of damaged blocks in E is currently less than red. According to the invalid file list located by RDI, the user obtains k valid data blocks E′ from E. The formula for recovering data is: FT=x′ -1 *E′. The specific process is as follows:
[0149]
[0150]
[0151] Efficiency analysis of multi-user and multi-file auditing:
[0152] In this paper, we apply the data label generation method of the EUISD article by Zhang et al. Therefore, when the RN of the TPA audit data is equal, batch auditing can be performed. Because RN equality means that the label keys of the batch files are the same, the batch files can be audited at the same time. For multi-user multi-file audits, TPA performs data aggregation audits on multi-user multi-files through bilinear mapping, which improves the efficiency of data audits. When performing multi-user task audits, TPA adopts a multi-user file aggregation audit method, which can audit multiple users and multiple files at the same time. This will greatly improve the efficiency of TPA audits. This will also improve the efficiency of cloud servers in responding to challenges and enhance user experience. In addition, after a multi-user multi-file audit, if there are invalid files, TPA can promptly query the location of the user's invalid files and promptly respond to the user's effective recovery of the invalid files in the next step.
[0153] Solution security proof part:
[0154] Security and correctness of batch auditing: Due to the characteristics of the label key adopted in this article, all users in the group use this key. Therefore, batch auditing can audit different files of multiple users at the same time. To prove the security of batch auditing, consider the following batch auditing. According to the integrity audit formula, for the sake of ease of description, we set the number of storage files of the user to 1. The data integrity audit of user A, user B, and user C is used as an example to illustrate. First, check whether the RN of the audit data labels of users A, B, and C are equal. If RN is equal, the three user label keys are the same, so they can be audited at the same time. The audit formula is as follows:
[0155]
[0156]
[0157] in, Because the left side is equal to the right side. The cloud server is able to store user data correctly. If the malicious server does not store correct or invalid user data, the left side must be made equal to the right side in order to pass the audit. Since the malicious server cannot obtain the private key, it cannot generate the correct storage proof, let alone forge the correct data signature. In other words, if the left side of the above formula is equal to the right side, it will pass the audit only if and only if the files of A, B, and C pass the audit respectively. Therefore, the batch audit security of the scheme is guaranteed. If the three RNs of the user are not equal, the label keys are not the same. Therefore, A, B, and C are audited for data integrity respectively. The security analysis is the same as when RN is equal.
[0158] Detectability: In the proposed scheme, if a file stored in the cloud contains n blocks, including d bad (deleted or modified) blocks, and the number of challenged blocks is s, the probability of detecting data corruption is at least
[0159] Evidence: The present invention adopts a multi-user batch auditing scheme, but the detection probability of files is the same as the probability framework of the scheme of Ateniese et al. In this protocol, multi-user data auditing only increases the number of audited files, and the detection probability calculation method is similar. The detection probability formula of the present invention is as follows: P N represents the probability that if the server provider deletes d blocks of the file, then the group of users will detect the misbehavior of the server provider after asking for proof of s blocks.
[0160] Data recoverability: The possibility of data recovery studied in this paper depends on the relationship between the number of damaged data dd and the redundancy red. When dd≤red, the damaged data can be recovered. According to the list of invalid files located by RDI, the user obtains k valid data blocks E' from E. The formula for data recovery is: F T = x′ -1 *E′. When dd>red, the damaged data is recovered. Because at this time, F T = x′ -1 *E′ cannot satisfy matrix calculation. Therefore, this paper can effectively recover the damaged data.
[0161] In summary, the solution of the present invention can quickly locate the invalid data of group users and effectively restore them in cloud storage applications of dynamic group users. The present invention also proves the security of the solution. The blind signature and convergent encryption ideas and identity verification in the solution enhance the security and innovation of the solution. This invention makes the application of dynamic group users in cloud storage more secure and complete.
[0162] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.
Claims
1. A storage proof method for fast location and recovery of expired files of dynamic group users, characterized in that: The system model of the storage proof method for fast positioning and recovery of expired files of dynamic group users consists of five entities: group users, group administrators, cloud, private key generator PKG and trusted third-party auditor TPA; specifically includes the following steps: Step 1: Generate the corresponding signing key according to the key generation protocol; Step 2: RS encode the file and generate a verification tag for the encoded data; Step 3: Conduct data integrity audit; Step 4: Locate and recover invalid files after integrity audit fails; In step 4, after the integrity audit fails, the invalid file is located, including: When the data integrity audit fails, TPA quickly locates the failed data file, and after finding the location of the invalid file, returns a list of invalid files to the group of users. Based on the results returned by TPA, the user restores the invalid file. (1) Search for invalid file SIF SIF is a function composed of ETM algorithm and RDI algorithm. The ETM function is called to quickly find the location of an invalid file, and the RDI function is called to quickly locate the locations of multiple invalid files. (2) Index Estimation Method ETM Index testing uses calculations to quickly identify single invalid files from a batch of audited files; The ETM algorithm process is as follows: ①Batch review: ②Calculate the signature: ③ Index calculation: Calculate the verification value of the equation, if Indicates that there are no invalid files; Then the L file is invalid; Indicates that there are at least two invalid files; If no single invalid file is found, RDI is called to quickly locate multiple invalid files; when RDI is called, a batch file is input to perform invalid file location and output a location list of all invalid files; (3) Rapid search for invalid files, associated dichotomy, RDI: The RDI algorithm process is as follows: ① Batch verification: calculation if Output n bits (1, 1, ..., 1) and stop; ②Assume X [1,N ] is the root node or the left child node, calculate ③If X [1,N] Node invalid, set BAF(X [1,N] )≠0, calculate X L The left child node of and return; ④Calculate the checksum of the right child node: If Then set if Then set Otherwise calculate ⑤ Recursively apply the same algorithm to the left and right subtrees, and finally output (b1, b2, ..., b N ).
2. The storage proof method for fast location and recovery of expired files of dynamic group users as claimed in claim 1, characterized in that: In step 1, generating a corresponding signing key according to a key generation protocol includes: Schnorr signature is used to calculate the identity key and partial key. PKG generates the identity key, the group administrator generates the partial key, and the group users generate the partial key based on the identity key IDK. ID =(R ID , σ ID ) and the partial key TK ID,RN =(R RN , σ RN ), RN = 0 to calculate the private key; the group user calculates σ = (σ ID +σ RN )modq, get the private key SK ID,RN Yes (R ID , R RN , σ).
3. The storage certification method for fast location and recovery of expired files of dynamic group users as claimed in claim 1, characterized in that: In step 2, the RS encoding of the file and generating a verification tag for the encoded data include: Run by a group of users, taking F as input, where F is a block {m i }, output the encoded data; the user performs RS encoding on the data, and the encoded data E = ω1, ω2, ..., ω n , uploaded to the cloud server; then generate a tag for the encoded data to authenticate the private key σ and file E, that is, block {m j } is the input, and the output data block verification tag set in N is the number of auditing users.
4. The storage proof method for fast location and recovery of expired files of dynamic group users as claimed in claim 1, characterized in that: In step 3, the implementation of data integrity audit includes three steps: initiating a challenge, generating a proof, and verifying. TPA generates an audit query chal = {C1, C2, ..., C N }, the cloud server generates the corresponding proof P = (T, μ) to prove that it has the complete cloud data. TPA verifies the data integrity based on the proof returned by the cloud server; TPA verifies the correctness of the proof P by checking whether the following equation holds: in If this equation holds true, it means all files are intact, and returns 1; Otherwise, it means there is at least one invalid file, and 0 is returned, and the search function is called to find the invalid file; When TPA fails to perform a data integrity audit, TPA quickly locates the invalid file location based on ETM and RDI, and returns the searched invalid file location to the group user.
5. The storage proof method for fast location and recovery of expired files of dynamic group users as claimed in claim 1, characterized in that: In step 4, after the integrity audit fails, invalid file recovery is performed, including: After SIF quickly finds the invalid file list, it uses RS erasure code to restore the invalid data; the user performs subsequent operations based on the audit results returned by TPA; if the returned audit is successful, it is normal; if the invalid file list is returned, the user can effectively restore the invalid file based on the location of the invalid file.
