A method for auditing the integrity of multi-copy data in cloud storage based on block group design

Abstract

The present invention relates to the field of network security technology, and in particular to a cloud storage multi-copy data integrity audit method based on block group design, comprising: setting security level parameters and pseudo-random functions; the data owner divides the original data into several data components, each data component forms several copies and is distributed in a cloud storage provider (CSP); the data owner generates challenge information and sends it to the CSP, and the CSP generates a challenge set according to the pseudo-random function after accepting the challenge; based on the challenge pair and the data component in the challenge set, the partial proof of the CSP is calculated and merged into a complete proof, and a hash proof is calculated; the hash proof is exchanged and verified between the CSPs, and if the verification passes, it is judged that the complete proofs of the two CSPs that exchange the hash proofs are the same, and if the verification fails, the data owner is notified, and the questionable CSP is determined according to the partial proof and the hash value is calculated to judge the integrity of its stored data. The present invention can ensure audit efficiency and data integrity and security.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing and network security, and in particular to a cloud storage multi-copy data integrity audit method based on block group design. Background Art

[0002] In the information age, the amount of data generated by individuals and organizations every day is enormous. It is estimated that the total amount of data in the world at the beginning of 2020 is about 44ZB, and it is expected that by 2025, the amount of new data generated every day in the world will reach 463TB. Faced with such a massive increase in data, the traditional local storage model can no longer meet the needs of data retrieval and management. Therefore, cloud storage services have emerged, providing high scalability, pay-as-you-go storage methods and convenient access anytime and anywhere, which has quickly attracted a large number of users. However, although cloud service providers have invested heavily in purchasing equipment and hiring experts to ensure service quality, cloud storage security incidents caused by hardware failures, management negligence or external attacks have occurred from time to time in recent years. Some unscrupulous service providers have actively deleted data that users rarely access in order to make extra profits, and concealed the facts when data is damaged to maintain their own reputation, which has caused users to doubt the security of data outsourcing.

[0003] As an effective means to improve the security of cloud storage data, the multi-copy storage strategy ensures data availability and integrity by replicating data in multiple locations, even if a single point of failure occurs, thereby enhancing data security. Especially when multiple copies are distributed in different cloud storage service providers (Cloud Storage Provider, CSP), the ability to resist single point failures is further improved, and data security is strengthened. In a multi-copy multi-CSP environment, copies of data can be audited with verification tags. This not only eliminates the calculation and storage of homomorphic verification tags (HVT), improves the credibility of sampling audit results, but also reduces a lot of computing and storage overhead compared to traditional models.

[0004] However, existing HVT-based multi-copy auditing schemes have some shortcomings that affect their practicality. For example, the generation and verification of HVT requires complex mathematical operations, such as exponential operations in the discrete logarithm problem (DLP) environment. Each time a label is generated, at least one exponential operation must be performed, and the number of operations is usually the same as the number of data blocks after segmentation. Although label generation is usually performed only once when data is outsourced, its huge amount of computation is still unacceptable. In addition, these schemes often use bilinear pairing to implement data integrity verification. Although the pairing operation is simpler than label generation and only needs to be calculated once for each audit, its computational overhead is still quite large compared to ordinary operations such as multiplication. Furthermore, users need to store a large amount of additional auxiliary data, namely labels, which brings significant and unreasonable storage burden to users.

[0005] In addition, the collaborative audit model requires each CSP to send its audit messages to all other CSPs, which is In a distributed system with 10 nodes, of network traffic. Although the current Internet bandwidth is capable of handling large amounts of data, with the increase in the number of CSPs and the increase in concurrent audit tasks, the exponential growth of network traffic is destined to become an inevitable and worrying problem. Therefore, how to reduce network overhead while ensuring data integrity and audit efficiency is a key challenge faced in multi-copy and multi-CSP scenarios. This requires the development of a new audit scheme that can overcome the computational and storage overhead of the traditional HVT model and solve the network transmission bottleneck in collaborative auditing, so as to meet the growing demand for cloud storage and users' higher expectations for data security. To this end, the present invention proposes a cloud storage multi-copy data integrity audit method based on block group design. Summary of the invention

[0006] The purpose of the present invention is to provide a cloud storage multi-copy data integrity audit method based on block group design, which can reduce network traffic, improve audit efficiency, and ensure data integrity and security.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] The cloud storage multi-copy data integrity audit method based on block group design includes:

[0009] Set security level parameters and pseudo-random functions;

[0010] The data owner divides the original data into several data components, each of which forms several copies, and the copies are distributed in the cloud storage provider;

[0011] The data owner generates challenge information and sends it to the cloud storage provider. After the cloud storage provider accepts the challenge, it generates a challenge set according to a pseudo-random function;

[0012] Calculate a partial proof of the cloud storage provider based on the challenge pair in the challenge set and the data component and combine them into a complete proof, and calculate a hash proof of the cloud storage provider based on the complete proof;

[0013] The hash certificate is exchanged and verified between cloud storage providers. If the verification is successful, it is determined that the complete certificates of the two cloud storage providers that exchanged the hash certificates are the same. If the verification is not successful, a report is sent to the data owner.

[0014] After receiving the report, the data owner determines the tampered copy based on the partial proof, identifies the suspected cloud storage provider, and determines the integrity of the stored data by calculating the hash value of the complete proof of the suspected cloud storage provider.

[0015] Optionally, the pseudo-random function is and ,in, for The set of natural numbers of order, namely , is a large number randomly selected according to the security level parameters, is a set of integers The upper boundary of

[0016] The challenge information is ,in, Represents a randomly drawn number, Indicates the number of data blocks that are evenly divided when each data component is sampled. Indicates the number of data blocks to be challenged;

[0017] The challenge pairs in the challenge set are , ,in, For the The selected data block, ; For the The corresponding data block of the random number selected is , .

[0018] Optionally, the method for calculating the partial proof of the cloud storage provider based on the challenge pair in the challenge set and the data component is:

[0019]

[0020] in, Part of the certification for cloud storage providers, , is an auxiliary collection used to distribute replicas, through - Design acquisition, The data components are The first data blocks.

[0021] Optionally, the method of merging the partial proofs into a complete proof is:

[0022]

[0023] in, For the Full certification of the cloud storage provider; For the A partial set of attestations from cloud storage providers; is an intermediate result set, a set part of; To collect Remove numbers from The resulting auxiliary set.

[0024] Optionally, the method for calculating the hash proof is:

[0025]

[0026] in, for A hash certificate from a cloud storage provider, is a hash function.

[0027] Optionally, exchanging and verifying the hash certificate between cloud storage providers includes:

[0028] The hash proof is exchanged between cloud storage providers using a round-robin scheduling method;

[0029] After the exchange is completed, the cloud storage provider calculates the verification hash proof through the hash proofs received from other cloud storage providers. If the verification hash proof is the same as the hash proof, it is determined that the complete proofs of the two cloud storage providers that exchanged hash proofs are the same; if the verification hash proof is different from the hash proof, a report is sent to the data owner.

[0030] Optionally, determining the integrity of the stored data of the cloud storage provider in question by calculating the hash value of the integrity certificate of the cloud storage provider in question includes:

[0031] Storage integrity is determined by comparing the hash value of the integrity certificate of the questionable cloud storage provider with the hash value of the local storage, wherein the hash value of the local storage is calculated when the data owner divides the original data into several data components.

[0032] The beneficial effects of the present invention are:

[0033] The present invention provides an efficient, flexible and secure cloud storage data integrity auditing solution, which is suitable for complex environments with multiple copies and multiple CSPs. It not only significantly reduces network communication overhead and removes the burden of label calculation and storage, but also provides a mechanism for detecting and responding to data corruption and CSP misbehavior, thereby ensuring the integrity and security of cloud storage data, providing users with reliable data auditing services, and has significant social and economic benefits and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. 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 paying creative work.

[0035] Figure 1 It is a flow chart of a cloud storage multi-copy data integrity audit method based on block group design according to an embodiment of the present invention;

[0036] Figure 2 A schematic diagram of data storage in a 6-CSP system according to an embodiment of the present invention;

[0037] Figure 3 The communication route between CSPs in the hash proof exchange phase in the 6-CSP system of the embodiment of the present invention;

[0038] Figure 4 For the embodiment of the present invention - Flowchart of the designed construction algorithm. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] This embodiment provides a cloud storage multi-copy data integrity audit method based on block group design, such as Figure 1 As shown, including:

[0042] Set security level parameters and pseudo-random functions;

[0043] The data owner divides the original data into several data components, each of which forms several copies, which are distributed among cloud storage providers (CSPs);

[0044] The data owner generates a challenge message and sends it to the CSP. After the CSP accepts the challenge, it generates a challenge set based on a pseudo-random function.

[0045] Calculate partial proofs of CSP based on challenge pairs and data components in the challenge set and merge them into a complete proof, and calculate a hash proof of CSP based on the complete proof;

[0046] The hash proof is exchanged and verified between CSPs. If the verification is successful, the complete proofs of the two CSPs that exchanged the hash proofs are judged to be the same. If the verification fails, a report is sent to the data owner.

[0047] After the data owner receives the report, he determines the tampered copy based on the partial proof, determines the questionable CSP, and determines the integrity of its stored data by calculating the hash value of the complete proof of the questionable CSP.

[0048] The following is combined with Figure 2-Figure 3 The 6-CSP system shown in the figure is used to explain this embodiment in detail, and specifically includes the following contents:

[0049] Step 1: System preparation phase. Select a large number that matches the security level. , and specify two pseudo-random functions, namely and , which are used to generate pseudo-random numbers and pseudo-random permutations respectively, and these functions and parameters are open to the public.

[0050] Step 2: Setup phase. In the case of In addition, a pseudo-random function and a pseudo-random permutation It is also specified that all of the above are publicly available.

[0051] Step 3: Storage phase. The data owner (DO) selects a prime number based on actual needs. , which determines the Distributed among CSPs copies, of which Then, DO will outsource the files Divide Components , each component The copies will be distributed to Different CSPs ensure that each CSP holds one of the copies. The combination and distribution of copies from different components follows -Design. To protect privacy, DO can encrypt all copies using symmetric encryption techniques such as AES before uploading. In addition, users need to calculate hash values ​​for each component for later arbitration. Since too many components will not be divided in the designed scheme, this local storage overhead can be ignored.

[0052] design:

[0053] As auxiliary information, construct an incident structure ,in , , , yes No. elements, for The elements, .

[0054] for :

[0055]

[0056] Then for :

[0057]

[0058] in, is a prime number selected by the data owner based on actual needs. Decided in Distributed among CSPs copies.

[0059] Then, -design By exchanging the blocks, each block All contain nodes , thus from The calculation details are as follows: Figure 4 shown.

[0060] like Figure 2As shown, this embodiment specifically adopts the (7,3,1)-design. The data owner divides the original data into 7 independent data components, and then stores these data components in the form of multiple copies evenly on 7 different CSPs according to the rules of block group design. The purpose of this is to enhance the security and reliability of data by using data redundancy, while reducing the risk of single node failure. In the (7,3,1)-design, the distribution method of data components ensures that at least one of any three CSPs stores a copy of each data component, thereby achieving redundant protection and efficient auditing of data. This design method balances the relationship between data redundancy, network overhead, and audit efficiency.

[0061] Step 4: Challenge phase. The audit task is generated by DO. The challenge was launched. Represents a randomly drawn number, Indicates the number of data blocks that are evenly divided when each data component is sampled. Indicates the number of data blocks to be challenged. After accepting the challenge, each CSP will generate a set of ,in , .

[0062] middle, is a selected function, , which accepts two inputs, namely An element in the integer group of order , and the set An element of and , the output is a set An element of ;

[0063] middle, is a selected function, , which accepts two inputs, namely An element in the integer group of order , and the set An element of and , the output is An element in the integer group of order is .

[0064] Step 5: Partial Proof Generation Phase: Each CSP will calculate a partial proof (PP) based on the stored data components.

[0065]

[0066] in, , is an auxiliary collection used to distribute replicas, through -Design Acquisition The data component conduct The first data blocks.

[0067] Step 6: Complete proof merging phase. It will select some from the PP it generates , indicating that this is from Then, the CSPs will exchange PFs. After the exchange is completed, each All the PFs at hand can be combined into a complete proof (CP) as follows:

[0068] in, For the i CSPs, For the j CSPs, since each CSP in this embodiment is equal, i The first CSP represents any one of the CSPs, and the j The first CSP represents the difference from the i Any remaining CSP of CSPs, For the A complete proof of a CSP; For the A set of partial proofs of a CSP; is an intermediate result set, a set part of; To collect Remove numbers from The resulting auxiliary set.

[0069] Step 7: Hash Proof Exchange Phase. After obtaining the complete proof, each CSP calculates its hash proof (HashedProof, HP), that is ,in is its index. Then, a chairperson-CSP is selected according to the round-robin scheduling method, and each of the remaining non-chairperson-CSPs exchanges its HP with the chairperson-CSP. The exchange process is as follows Figure 3 For consistency, continue to use Figure 2The 6-CSP system used in . Since each exchange is the same, only the information exchanged between the chairman-CSP and one of the remaining CSPs is marked in the figure. This process ensures that the complete proof of the data can be exchanged between all participating CSPs. At the same time, by reducing the broadcast communication overhead, the network transmission burden of the entire system is significantly reduced. Under the framework of the (7,3,1)-design, the data owner and the seven CSPs form a complete system. The DO is responsible for initiating audit tasks and receiving reports from the CSP, while the CSPs generate partial proofs based on the challenges issued by the DO and complete the integration and verification of evidence through mutual communication exchanges.

[0070] Step 8: Verification phase. Each CSP computes a hash value by concatenating its CP with the sender’s index, e.g. from Received from of Post-calculation .Then examine Is it true? If the equation is true, ,otherwise The situation will be reported to the DO, triggering arbitration.

[0071] Step 9, arbitration phase. After receiving the report from the CSP, the DO can request all CSPs to send the partial proofs PP they hold, and then determine which copies of the data components have been tampered with based on the majority principle, and determine the CSPs where these copies are located. If the DO has doubts about the honesty of the CSP, it can ask the CSP to provide complete data, calculate the hash value by itself and compare it with the locally stored value to confirm the integrity of the data. The entire audit process ensures the integrity and security of the data in a multi-copy, multi-cloud storage provider environment, while optimizing the communication and computing costs during the audit process.

[0072] Experimental verification:

[0073] In order to evaluate the proposed cloud storage multi-copy data integrity audit method based on block group design, the experiment was conducted on the Ubuntu Desktop 22.04 operating system, using an Intel Core i7-6500U processor with a main frequency of 2.50GHz and 4GB of memory. All simulation experiments were written in C language and used the GCC 11.2.0 version compiler. SHA-256 was used as the hash function, and the GNU Multiple Precision Arithmetic (GMP) library was used to process large number calculations. For comparison scheme 1 DSAS: Y. Su, Y. Li, B. Yang and Y. Ding, "Decentralized Self-Auditing Scheme With Errors Localization for Multi-Cloud Storage," in IEEETransactions on Dependable and Secure Computing, vol. 19, no. 4, pp. 2838-2850, 1 July-Aug. 2022, doi: 10.1109 / TDSC.2021.3075984. And comparison scheme two CIA: T. Li, J. Chu and L. Hu, "CIA: A Collaborative Integrity Auditing Scheme for CloudData With Multi-Replica on Multi-Cloud Storage Providers," in IEEETransactions on Parallel and Distributed Systems, vol. 34, no. 1, pp. 154-162, 1 Jan. 2023, doi: 10.1109 / TPDS.2022.3216614. uses the Pairing-Based Cryptography (PBC) library combined with GMP implementation.

[0074] In the experiment, the parameter a.param was configured according to the PBC library parameters, and each experiment was repeated 10 times.

[0075] In the computational cost test of proof generation and verification, this method has an advantage over other schemes because it does not use operations with high computational cost, such as multiplication and exponential operations. When the number of replicas increases, the communication cost of this scheme, DSAS and CIA in the proof phase is also compared. Bit and bit conditions, where The number of chunks each replica is divided into, The number of data blocks challenged for the audit phase, is the size of the multiplication cycle group element selected in the algorithm, is the size of the integer group element selected in the algorithm, is the output space size of the hash function used in the algorithm. The communication overhead of a single CSP in this method is about 1280 bits when the number of CSPs is 57. However, the communication overhead of a single CSP in DSAS and CIA exceeds 5000 bits when the number of CSPs is 13.

[0076] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A cloud storage multi-copy data integrity audit method based on block group design, characterized in that: include: Set security level parameters and pseudo-random functions; The data owner divides the original data into several data components, each of which forms several copies, and the copies are distributed in the cloud storage provider; The data owner generates challenge information and sends it to the cloud storage provider. After the cloud storage provider accepts the challenge, it generates a challenge set according to a pseudo-random function; Calculate a partial proof of the cloud storage provider based on the challenge pair in the challenge set and the data component and combine them into a complete proof, and calculate a hash proof of the cloud storage provider based on the complete proof; The hash certificate is exchanged and verified between cloud storage providers. If the verification is successful, it is determined that the complete certificates of the two cloud storage providers that exchanged the hash certificates are the same. If the verification is not successful, a report is sent to the data owner. After receiving the report, the data owner determines the tampered copy based on the partial proof, identifies the suspected cloud storage provider, and determines the integrity of the stored data by calculating the hash value of the complete proof of the suspected cloud storage provider; Wherein, the pseudo-random function is and ,in, for The set of natural numbers of order, namely , is a large number randomly selected according to the security level parameters, is a set of integers The upper boundary of The challenge information is ,in, Represents a randomly drawn number, Indicates the number of data blocks that are evenly divided when each data component is sampled. Indicates the number of data blocks to be challenged; The challenge pairs in the challenge set are , ,in, For the The selected data block, ; For the The corresponding data block of the random number selected is , ; The method for calculating the partial proof of the cloud storage provider based on the challenge pair in the challenge set and the data component is: in, Part of the certification for cloud storage providers, , is an auxiliary collection used to distribute replicas, through - Design acquisition, The data components are The first data blocks.

2. The cloud storage multi-copy data integrity audit method based on block group design according to claim 1 is characterized in that: The method of combining the partial proofs into a complete proof is: in, For the Full certification of the cloud storage provider; For the A partial set of attestations from cloud storage providers; is an intermediate result set, a set part of; To collect Remove numbers from The resulting auxiliary set.

3. The cloud storage multi-copy data integrity audit method based on block group design according to claim 2 is characterized in that: The method for calculating the hash proof is: in, for A hash certificate from a cloud storage provider, is a hash function.

4. The cloud storage multi-copy data integrity audit method based on block group design according to claim 1 is characterized in that: Exchanging and verifying the hash certificate between cloud storage providers involves: The hash proof is exchanged between cloud storage providers using a round-robin scheduling method; After the exchange is completed, the cloud storage provider calculates the verification hash proof through the hash proofs received from other cloud storage providers. If the verification hash proof is the same as the hash proof, it is determined that the complete proofs of the two cloud storage providers that exchanged hash proofs are the same; if the verification hash proof is different from the hash proof, a report is sent to the data owner.

5. The cloud storage multi-copy data integrity audit method based on block group design according to claim 1 is characterized in that: Determining the integrity of the stored data by calculating the hash value of the complete proof of the cloud storage provider in question includes: Storage integrity is determined by comparing the hash value of the integrity certificate of the questionable cloud storage provider with the hash value of the local storage, wherein the hash value of the local storage is calculated when the data owner divides the original data into several data components.

Images