Certificateless multi-cloud data auditing deduplication method and system based on anti-backdoor attack of blockchain

Abstract

The application discloses a kind of certificateless multi-cloud data auditing deduplication method and system based on anti-backdoor attack of block chain, belong to cloud storage security and cryptography technical field.The method includes: generating randomization main public key and being incorporated into system public parameter publishing to block chain;Based on user pseudo-identity generation user key, key generation center generates partial private key, after firewall randomization processing, it is combined with local secret value to generate complete private key by user;Based on convergent encryption generation ciphertext and deduplication label, when first uploading, after secret sharing slice of ciphertext is distributed to multiple cloud servers, when repeatedly uploading, ownership verification is acquired slice pointer;Based on data block label, audit challenge set is published through block chain smart contract, cloud server is aggregated to generate audit proof, and is stored in evidence after user verification;According to audit result, effective slice is downloaded from normal cloud server, and ciphertext is reconstructed and decrypted by secret sharing recovery algorithm.The application simultaneously realizes anti-backdoor attack, ciphertext-level deduplication, multi-cloud redundant storage and decentralized audit under the premise of not needing completely trusted third party.

Description

Technical Field

[0001] This invention relates to the intersection of cloud storage security and cryptography, specifically to a blockchain-based, certificate-free multi-cloud data auditing and deduplication method and system for resisting backdoor attacks. Background Technology

[0002] With the advent of the big data era, cloud storage technology has become the mainstream storage method due to its convenience, efficiency, and low cost. However, once users upload data to the cloud, they lose physical control over the data, making the integrity auditing of outsourced data a pressing issue. Traditional data integrity auditing solutions typically rely on third-party auditing agencies (TPAs), but TPAs ​​suffer from single points of failure, performance bottlenecks, and trust risks. In recent years, the decentralized, immutable, and traceable characteristics of blockchain technology have provided a new approach to breaking free from TPA dependence, and existing solutions utilize blockchain and smart contracts to achieve public auditing.

[0003] Another point of concern is the severe data redundancy in cloud storage; statistics show that over 80% of the data is duplicated. Data deduplication technology can significantly save storage space, but traditional encryption algorithms can cause the same data to generate different ciphertexts, making direct ciphertext-level deduplication impossible. Convergent Encryption (CE) has addressed this contradiction, but most existing deduplication solutions store data on a single cloud server, posing a single point of failure and the risk of collusion attacks. Some multi-cloud solutions use secret sharing to fragment data storage, but often neglect deduplication efficiency or the security of data ownership verification. Furthermore, existing solutions generally assume the Key Generation Center (KGC) is absolutely trustworthy, failing to consider the risk of backdoors being implanted in the KGC algorithm. Once the KGC is controlled by an attacker and a backdoor is implanted, system parameters and user private keys may be leaked, causing the entire security mechanism to fail.

[0004] Therefore, there is an urgent need for a data auditing and deduplication solution that can simultaneously solve the above problems, achieving a unified solution for resistance to backdoor attacks, certificate-free management, multi-cloud redundant storage, efficient deduplication, and decentralized auditing. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a certificateless multi-cloud data auditing and deduplication method and system based on blockchain to resist backdoor attacks. It resists backdoor risks on the key generation center side through a firewall randomization purification mechanism, integrates convergence encryption and Ramp secret sharing to achieve encrypted-level deduplication and redundant storage in a multi-cloud environment, builds a decentralized batch auditing system based on blockchain and smart contracts, and strengthens data access control by combining a dual ownership verification mechanism. Under the premise of ensuring data confidentiality and integrity, it effectively reduces the system's computing and communication overhead.

[0006] This invention is achieved through the following technical solution: Firstly, this application provides a blockchain-based, certificate-free, multi-cloud data auditing and deduplication method resistant to backdoor attacks, comprising the following steps: Step 1: Generate a randomized master public key and include it in the system's public parameters, then publish it to the blockchain; Step 2: Generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate the complete user private key. Step 3: Generate ciphertext and deduplication tags for the data to be uploaded based on the convergent encryption algorithm, and determine whether the data is duplicated by using the deduplication tags; During the initial upload, the encrypted text is secretly shared to obtain multiple data fragments. The user's private key is used to generate data block tags and ownership certificates for each data fragment. The data fragments are then distributed to multiple cloud servers for storage, and the fragment pointers returned by each cloud server are obtained. When uploading repeatedly, the user's ownership is verified through the ownership proof, and the fragment pointer is obtained after successful verification; Step 4: Based on the data block tags, publish the audit challenge set through the blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate audit proof. The user verifies the integrity of the audit proof and stores the audit results on the blockchain. Step 5: Based on the audit results, download valid data fragments from the cloud server that has passed integrity verification, reconstruct the ciphertext using the secret sharing recovery algorithm, and decrypt to obtain the original data.

[0007] Preferably, in step 1, the key generation center generates an original master public key, and the firewall deployed at the output end of the key generation center selects randomization parameters and performs randomization operations on the original master public key to generate the randomized master public key; the firewall privately stores the randomization parameters.

[0008] Preferably, in step 2, the user selects a secret random number and combines it with their real identity to generate a pseudo identity; the key generation center generates a partial private key based on the pseudo identity; the firewall performs randomization processing on the partial private key using locally stored randomization parameters; the user combines the randomized partial private key with a local secret value to generate a complete user private key, and calculates the user public key based on the local secret value and system public parameters.

[0009] Preferably, the secret sharing in step 3 adopts the Ramp secret sharing scheme, which sets the total number of shares n, the recovery threshold k, and the security threshold r, and divides the ciphertext into kr information fragments and r random fragments, which are encoded into n data fragments; when an attacker obtains no more than r fragments, he cannot obtain any valid information from the ciphertext.

[0010] Preferably, the ownership verification in step 3 when uploading repeatedly includes: The first layer of verification involves the cloud server comparing the ownership proof submitted by the user with the ownership proof stored during the initial upload to verify the consistency of the ownership proof. The second layer of verification involves the cloud server randomly selecting a data block index, the user returning the Merkle authentication path and response value for the corresponding data block, and the cloud server reconstructing the Merkle root and comparing it with the root hash stored on the blockchain. After both layers of verification pass, the shard pointer is obtained.

[0011] Preferably, in step 4, within the same audit period, the challenge set of the same data block is reused by multiple users who store the data; the users verify the audit proof based on the bilinear pairing equation.

[0012] Preferably, in step 5, the user verifies the integrity of each downloaded fragment by using data block tags; marks the cloud server corresponding to the fragment that fails verification as a malicious node and publishes it to the blockchain; downloads fragments from other cloud servers to replace the original fragments until a number of valid fragments that meet the recovery threshold are collected.

[0013] Preferably, in step 3, after the initial upload is completed, the transaction smart contract jointly signed by the user and multiple cloud servers is broadcast to the blockchain for evidence storage. The transaction smart contract records the storage location of the data fragments, the data block tags, and the ownership proof.

[0014] Secondly, this application provides a blockchain-based certificateless multi-cloud data auditing and deduplication system resistant to backdoor attacks, comprising: The initialization module is used to generate a randomized master public key and incorporate it into the system's public parameters for publication on the blockchain; The key generation module is used to generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate a complete user private key. The data upload module is used to generate ciphertext and deduplication tags for the data to be uploaded based on a convergent encryption algorithm, and to determine whether the data is duplicated by using the deduplication tags. During the first upload, the ciphertext is secretly shared to obtain multiple data fragments. Data block tags and ownership certificates are generated for each data fragment using the user's private key. Each data fragment is distributed to multiple cloud servers for storage, and fragment pointers returned by each cloud server are obtained. During repeated uploads, the user's ownership is verified by the ownership certificate. After successful verification, the fragment pointers are obtained. The data audit module is used to publish an audit challenge set based on data block tags through a blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate an audit certificate. Users verify the integrity of the audit certificate and store the audit results on the blockchain. The data download module is used to download valid data fragments from the cloud server that has passed integrity verification based on the audit results, reconstruct the ciphertext through the secret sharing recovery algorithm, and decrypt to obtain the original data.

[0015] Thirdly, this application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks as described above.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: This application provides a blockchain-based certificateless multi-cloud data auditing and deduplication method to resist backdoor attacks. By deploying a firewall at the output of the key generation center and performing randomization perturbation on the master public key before uploading it to the blockchain, even if a backdoor is implanted in the key generation center's algorithm, attackers cannot extract valid information from the public parameters, thus cutting off the backdoor attack path at the source. Then, the firewall's randomization protection is extended to the user key generation stage. Part of the private key is purified and combined with the user's local secret value, preventing the key generation center from obtaining the complete private key. This eliminates key escrow risks and builds end-to-end backdoor defense capabilities. Secondly, convergent encryption ensures consistent ciphertext generation from the same plaintext, supporting ciphertext-level deduplication to compress storage redundancy. Combined with secret sharing, ciphertext is fragmented and stored on multiple independent cloud servers. Security thresholds ensure that a small number of servers colluding cannot obtain ciphertext information, overcoming the single point of failure bottleneck of single cloud server storage. Dual ownership verification confirms that the user indeed holds the original data from two dimensions: content hash and Merkle structure, effectively preventing unauthorized attackers from illegally obtaining data by stealing fragment pointers. Furthermore, blockchain smart contracts replace traditional third-party auditing institutions, achieving decentralization of the entire audit process and eliminating single points of failure and trust risks. The on-chain reuse mechanism of the challenge set allows multiple users to share the same challenge result, significantly reducing audit computation and communication overhead. Finally, leveraging the threshold recovery characteristic of secret sharing, sufficient fragments can be downloaded from a legitimate audited server to reconstruct the ciphertext, providing inherent fault tolerance. Through the synergistic combination of these technical features, this method simultaneously solves the technical challenges of backdoor attacks, single points of failure, deduplication efficiency, and centralized auditing without relying on a fully trusted third party.

[0017] This application also proposes a blockchain-based certificateless multi-cloud data auditing and deduplication system resistant to backdoor attacks, an electronic device, and a computer storage medium, which possess all the advantages of the aforementioned blockchain-based certificateless multi-cloud data auditing and deduplication method resistant to backdoor attacks. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the system model of the present invention; Figure 2 This is a flowchart of the certificateless multi-cloud data auditing and deduplication method based on blockchain to resist backdoor attacks, as described in this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] See Figure 1 The system entities include the Key Generation Center (KGC) and the firewall. Group of users (GU), multiple cloud servers (CS), and blockchain smart contracts.

[0023] 1. Key Generation Center (KGC): Used to generate system public parameters, master private key and master public key, generate partial private keys for users and is responsible for deploying blockchain and smart contracts; 2. Firewall ( ): Deployed at the KGC output end, it is used to randomize the master public key and part of the private key, output the purified parameters, and resist backdoor attacks; 3. User (GU): Used to generate pseudo-identities, secret values, public-private key pairs, perform convergent encryption, secret sharing sharding, data block label generation, Merkle tree construction, dual ownership verification, and audit challenges and verification; 4. Cloud Server (CS): Multiple independent cloud servers are used to store encrypted fragments and data tags, and respond to data upload verification, ownership challenges and audit proof generation; 5. Blockchain and Smart Contracts: Used to publish system parameters, transaction smart contracts (TSCs), audit challenge sets, and audit results, providing a decentralized and transparent audit and arbitration mechanism.

[0024] A blockchain-based, certificate-free, multi-cloud data auditing and deduplication method resistant to backdoor attacks includes the following steps: Step 1: Generate a randomized master public key and include it in the system's public parameters, then publish it to the blockchain; Step 2: Generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate the complete user private key. Step 3: Generate ciphertext and deduplication tags for the data to be uploaded based on the convergent encryption algorithm, and determine whether the data is duplicated by using the deduplication tags; During the initial upload, the encrypted text is secretly shared to obtain multiple data fragments. The user's private key is used to generate data block tags and ownership certificates for each data fragment. The data fragments are then distributed to multiple cloud servers for storage, and the fragment pointers returned by each cloud server are obtained. When uploading repeatedly, the user's ownership is verified through the ownership proof, and the fragment pointer is obtained after successful verification; Step 4: Based on the data block tags, publish the audit challenge set through the blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate audit proof. The user verifies the integrity of the audit proof and stores the audit results on the blockchain. Step 5: Based on the audit results, download valid data fragments from the cloud server that has passed integrity verification, reconstruct the ciphertext using the secret sharing recovery algorithm, and decrypt to obtain the original data.

[0025] This method involves introducing a firewall ( This invention randomizes system parameters and a portion of the private key output by the key generation center, making it impossible for attackers to distinguish whether a backdoor exists, thus achieving a certificateless framework resistant to backdoor attacks. It integrates convergent encryption and Ramp secret sharing, storing ciphertext fragments across multiple cloud servers, supporting ciphertext deduplication while resisting single points of failure and collusion attacks. Decentralized batch auditing is implemented based on blockchain and smart contracts, reducing overhead through challenge set reuse on the chain. A dual verification mechanism of data ownership proof and Merkle tree is employed to prevent users without ownership from obtaining fragment pointers. Security proofs and performance analysis show that this invention possesses security features such as resistance to backdoor attacks, unforgeable tags, unforgeable audit proofs, and unobtainable fragment pointers. Computational and communication overhead is within acceptable limits, making it suitable for cloud storage scenarios with high security requirements.

[0026] Example 1 See Figure 2 This embodiment provides a blockchain-based certificateless multi-cloud data auditing and deduplication method to resist backdoor attacks. The overall process is divided into five stages: system initialization, key generation, data upload, data auditing, and data download, covering the entire process of key management, data storage, integrity verification, and data recovery. Step 1: Using the security parameter κ as input, the Key Generation Center (KGC) builds a bilinear group cryptographic foundation system and generates the original master public and private keys and hash function clusters. Then, the firewall deployed at the output end performs randomization perturbation on the master public key to eliminate backdoor risks. Finally, it generates the security-hardened system public parameters and publishes them to the blockchain. This common parameter serves as a unified operational benchmark for the entire system, providing a mathematical foundation for all subsequent cryptographic operations, including key generation, data block label generation, and audit verification.

[0027] The specific implementation method of step 1 will be described in detail below.

[0028] S1.1: The key generation center inputs the security parameter κ, selects two bilinear cyclic groups of prime order p, defines the corresponding bilinear mapping, and selects the generator g of the cyclic group; it then randomly selects the master private key. The master public key is calculated. Five cryptographic hash functions were selected. , , , , Combine to generate initial system common parameters .

[0029] S1.2: The firewall receives the initial system public parameters and master public key output by the key generation center. Randomly select randomization parameters Based on this parameter, the master public key Perform randomization operations to obtain the randomized master public key. Firewall The randomization parameter t is kept private and not disclosed to any third-party entity within the system.

[0030] S1.3: Firewall Will contain randomized master public key The complete system public parameters are published to the blockchain smart contract and made public to the entire system; all participating entities can read these public parameters from the blockchain for use in various subsequent cryptographic operations.

[0031] By adding a firewall module to the output link of the key generation center, the core master public key is randomized and perturbed, making the randomized parameters statistically indistinguishable from the original parameters. Even if a backdoor is implanted in the algorithm of the key generation center, attackers cannot extract valid information related to the backdoor from the publicly available system parameters, thus fundamentally resisting backdoor attacks on the key generation center side. Simultaneously, the system's public parameters are published to the blockchain, leveraging the blockchain's immutability to ensure the public consistency of parameters, avoiding the risk of parameter tampering caused by centralized distribution, and achieving trusted synchronization of parameters across the entire system.

[0032] In some embodiments, the aforementioned bilinear group can be replaced by an asymmetric bilinear group to further improve computational efficiency under the same security level; the aforementioned cryptographic hash function can adopt the national cryptographic SM3 hash algorithm to meet the compliance requirements of domestic cryptographic applications.

[0033] In some embodiments, the firewall can be implemented using a separate hardware security module, with the randomization parameter t generated and stored entirely within the hardware security module, further reducing the risk of randomization parameter leakage.

[0034] Step 2: Using user identity information, system public parameters, key generation center master private key and firewall randomization parameters as input, the user first hides their real identity with a fake identity and initiates a key application. After the key generation center generates a partial private key and the firewall performs randomization purification, the user combines it with their local exclusive secret value to generate a complete user public-private key pair. This public-private key pair is used for subsequent data tag signing, ownership proof, audit verification, and other operations, enabling key distribution with both identity privacy protection and backdoor defense capabilities in a certificate-free architecture.

[0035] The specific implementation method of step 2 will be described in detail below.

[0036] S2.1: The user randomly selects a secret random number. It calculates and generates a fake identity by combining its own real identity information. , to use false identities The key is sent to the Key Generation Center (KGC); the user's real identity is not directly exposed to the outside world, but only participates in the entire process of system interaction through a pseudo-identity.

[0037] S2.2: The Key Generation Center (KGC) receives the pseudo-identity submitted by the user. Then, use its own master private key to calculate the partial private key corresponding to the fake identity. Send part of the private key to the firewall .

[0038] S2.3: Firewall Using the privately stored randomization parameter t, the received partial private key is... Perform randomization operations to generate a randomized partial private key. and return it to the corresponding user. .

[0039] S2.4: User With local secret value As the second private key component, it is combined with the randomized partial private key. The complete user private key is calculated. Simultaneously, based on the user's private secret value and system public parameters, the corresponding user's public key is calculated. User private key User public key stored locally and privately It can be publicly used for various verification operations.

[0040] Employing a certificateless key generation mechanism, the user's private key is composed of a partial private key generated by the key generation center and a unique secret value generated locally by the user. The key generation center cannot obtain the user's complete private key, eliminating the key escrow risk inherent in traditional identity-based encryption. Secondary randomization of the partial private key via a firewall further blocks the path for information leakage through a partial private key via a backdoor at the key generation center, ensuring that even if a backdoor exists at the key generation center, it is impossible to recover the user's complete key from a partial private key. Simultaneously, a pseudo-identity mechanism is employed to prevent the user's real identity from being leaked in public interactions, balancing identity authentication and privacy protection.

[0041] In some embodiments, the aforementioned pseudo-identity can be generated in conjunction with an expiration date field, giving the generated key an expiration date and automatically invalidating it upon expiration, thereby further enhancing the security of key management.

[0042] In some embodiments, a multi-authorization architecture can be adopted, in which multiple independent key generation centers manage partial private keys for different attribute domains. Users need to aggregate partial private keys from multiple authorization centers to generate a complete key, further mitigating the risk of single-point key leakage.

[0043] Step 3: Taking the original data F to be uploaded, system public parameters, and user private key as input, first, ciphertext is generated based on convergent encryption to complete deduplication preprocessing, and then the deduplication label is used to determine whether the data has been stored. For the first uploaded data, Ramp secret sharing sharding is used to distribute it to multiple cloud servers for storage. For duplicate uploaded data, a dual verification mechanism is used to confirm user ownership before allocating shard pointers. Finally, the transaction smart contract is completed and stored on the blockchain. This achieves ciphertext-level data deduplication, multi-cloud sharded redundant storage, and data ownership confirmation, providing a storage foundation and ownership basis for data auditing and downloading.

[0044] The specific implementation method of step 3 will be described in detail below.

[0045] S3.1: User Divide the data F to be uploaded into fixed-size data blocks. Construct a Merkle tree and calculate the Merkle tree root hash value. Read the global salt value and Argon2id algorithm parameters from the blockchain smart contract, and calculate the convergence key based on the data content. The original data is encrypted using the convergent key to obtain the ciphertext. ; Calculate deduplication tags based on ciphertext and global salt value .

[0046] S3.2: The user sends the deduplication tag T to the cloud server. The cloud server queries the tag record stored locally to determine whether the deduplication tag already exists. If it does not exist, it is determined to be the first upload, and S3.3 is executed. If it already exists, it is determined to be a duplicate upload, and S3.4 is executed.

[0047] S3.3: Deduplicated tag T first upload (1) User The Ramp secret sharing scheme is used for the ciphertext C, with a total number of shares n, a recovery threshold k, and a security threshold r. The Share algorithm divides the ciphertext into kr information fragments, mixes in r random fragments, and encodes them into n data fragments. When an attacker obtains no more than r fragments, they cannot obtain any valid information from the ciphertext.

[0048] (2) The user calculates the corresponding data block label for each data fragment based on their own private key. At the same time, a data ownership certificate is generated. With file tags .

[0049] (3) The user divides each data into fragments, corresponding data block labels, and data ownership certificates. The transaction smart contract scripts to be signed are uploaded to the corresponding cloud server CS.

[0050] (4) Each cloud server verifies the validity of the data block label. After successful verification, it stores the corresponding shard and returns the shard pointer. Subsequently, all parties jointly sign the transaction smart contract. And broadcast it to the blockchain to complete the evidence storage.

[0051] S3.4: Deduplication Tag T Duplicate Upload Process (1) First-level ownership verification: The user submits proof of data ownership to the cloud server. The cloud server compares it with the proof stored during the initial upload. Verify consistency; if the verification fails, access is directly denied.

[0052] (2) Second-level ownership verification: After the first-level verification is passed, the cloud server CS randomly selects the data block index set S and the random challenge number. Challenge information is generated and sent to the user. ;user The system returns the Merkle authentication path and response value for the corresponding data block. The cloud server reconstructs the Merkle root based on the returned information and verifies whether it is consistent with the root hash Root stored in the smart contract on the blockchain.

[0053] (3) After both layers of verification are passed, the cloud server returns the corresponding data shard pointer to the user, and both parties jointly sign the transaction smart contract and broadcast it to the blockchain to complete the notarization.

[0054] By employing a convergent encryption algorithm, identical original data generates identical ciphertext, thus supporting ciphertext-level data deduplication and significantly saving cloud storage space. Combined with the Ramp secret sharing scheme, ciphertext fragments are stored across multiple independent cloud servers, avoiding single points of failure. Furthermore, a security threshold ensures that even if fewer than the threshold number of cloud servers collude, they cannot obtain any valid information from the ciphertext. Even if more than the recovery threshold number of cloud servers collude to recover the ciphertext, the original data cannot be decrypted without the convergent key, forming a two-layer security protection. A dual ownership verification mechanism confirms that the user indeed possesses complete original data from both content hash and structure verification dimensions, effectively preventing attackers from illegally gaining data access rights by stealing fragment pointers or hash values. On-chain notarization of transaction smart contracts enables public traceability of data ownership, providing a credible basis for subsequent audits and ownership disputes.

[0055] In some embodiments, the above-mentioned convergence encryption algorithm can be replaced by a message locking encryption scheme to further improve key security in weak entropy data scenarios; the above-mentioned deduplication tags can be generated in combination with user attributes to achieve flexible switching between fine-grained cross-user deduplication and single-user deduplication.

[0056] In some embodiments, the Ramp secret sharing scheme described above can be replaced with the Shamir secret sharing scheme to adapt to application scenarios with higher security requirements and a tolerance for higher computational overhead.

[0057] In some embodiments, the aforementioned dual ownership verification can be supplemented with a zero-knowledge proof step, allowing users to complete ownership proof without disclosing the original data, thereby further protecting data privacy.

[0058] Step 4: Taking the audit requirements, system public parameters, user private keys, and shards and tags stored on the cloud server as input, publish the audit challenge set based on the blockchain smart contract and realize the reuse of challenge results by multiple users. The cloud server aggregates the corresponding data blocks to generate audit proofs. Users complete batch integrity verification through bilinear pairing equations. Finally, the audit results are stored on the blockchain.

[0059] The specific implementation method of step 4 will be described in detail below.

[0060] S4.1 users Based on the audit requirements, a challenge set is generated by randomly selecting the index of the data blocks to be audited. The challenge set is published to the blockchain smart contract; within the same audit period, the same data block is challenged only once, and other users who store the data can directly reuse the challenge set without having to generate it again.

[0061] S4.2: Various cloud servers After receiving the challenge information, the corresponding data block labels are aggregated based on the data block index in the challenge set. With data factors Calculate and generate aggregate audit proof And return to the user who initiated the audit. .

[0062] S4.3: User After receiving audit verification reports from each cloud server, the aggregated tag is calculated based on system public parameters and the user's public key. and data aggregation factors Verify whether the following bilinear pairing equation holds:

[0063] If the equation is true, the data stored on the corresponding cloud server is determined to be complete, and (result=1) is output; otherwise, the data is determined to have been tampered with, and (result=0) is output. The user publishes the final audit results to the blockchain smart contract for evidence storage.

[0064] S4.4: If the cloud server disagrees with the audit results, it can submit an arbitration application to the blockchain smart contract. The smart contract will automatically review the audit results based on the challenge set, tags and proofs stored on the chain. If the review determines that the audit results are incorrect, the audit results will be corrected and the user who maliciously initiated the audit will be punished accordingly. If the review determines that the data has indeed been tampered with, the corresponding cloud server will be marked as an untrusted node.

[0065] By leveraging blockchain and smart contracts to replace traditional third-party auditing firms, a fully decentralized integrity audit is achieved, eliminating single points of failure and trust risks inherent in third-party auditing institutions. Employing a batch aggregation verification mechanism, a single challenge can complete the integrity verification of multiple data blocks. Combined with an on-chain reuse mechanism for challenge sets, this significantly reduces the computational and communication overhead of auditing in multi-user scenarios. Based on a certificate-free signature-based bilinear verification equation, and under the assumption of computational Diffie-Hellman difficulty, it possesses unforgeability, ensuring the correctness and reliability of audit results. The audit results, stored on-chain, are fully traceable, and an automatic arbitration mechanism further guarantees the openness, transparency, and fairness of the audit process.

[0066] In some embodiments, the above auditing scheme can support dynamic data auditing, synchronously updating data block labels and Merkle tree structures for data insertion, deletion, and modification operations, adapting to cloud storage scenarios with frequent dynamic updates.

[0067] In some embodiments, zero-knowledge proof technology can be introduced to optimize the audit proof generation process, so that the cloud server does not disclose any plaintext related information when generating proof, thereby further enhancing the privacy of the audit process.

[0068] Step 5: Using the fragment pointer, system public parameters, user private key, and convergence key as input, download data fragments from multiple cloud servers and verify their integrity one by one. Simultaneously mark malicious nodes and supplement valid fragments. After collecting valid fragments that meet the recovery threshold, reconstruct the complete ciphertext through the secret sharing recovery algorithm. Finally, decrypt the original data using the convergence key. Fault-tolerant download is achieved by utilizing the threshold recovery feature, and reliable data recovery is completed while ensuring end-to-end confidentiality of the data.

[0069] The specific implementation method of step 5 will be described in detail below.

[0070] S5.1: User Select at least a number of fragments that meet the recovery threshold k from the fragment pointers in local storage, and initiate a data download request to the corresponding cloud server CS. Obtain the corresponding data shards.

[0071] S5.2: For each downloaded data fragment, the user verifies its integrity using the corresponding data block label;

[0072]

[0073] If a shard fails verification, the corresponding cloud server is marked as a malicious node, the malicious node information is published to the blockchain, and additional valid shards are downloaded from other cloud servers to replace it, until a sufficient number of valid shards are collected.

[0074] S5.3: After the user collects a number of valid fragments that meet the recovery threshold, the Ramp secret-shared Recover algorithm is invoked to reconstruct the complete ciphertext C from the fragments; the locally stored convergence key is used. Perform a decryption operation on the ciphertext to restore the original data F.

[0075] Based on the threshold recovery feature of Ramp secret sharing, the complete ciphertext can be restored as long as a sufficient number of valid fragments are obtained. It has inherent fault tolerance and can withstand the downtime, data tampering, or offline failure of a small number of cloud servers. Through fragment-by-fragment tag verification, the cloud server storing malicious or corrupted data can be accurately located, enabling the traceability and marking of malicious nodes. The entire decryption operation is performed locally on the user's computer, and the convergence key is held independently by the user. The cloud server cannot obtain the original plaintext, ensuring end-to-end confidentiality of the data.

[0076] In some embodiments, the above-mentioned segmented download can adopt a parallel download method, downloading segments from multiple cloud servers simultaneously to improve the overall data download speed.

[0077] In some embodiments, the user's convergence key and private key can be stored in a hardware security module, and the entire decryption operation is performed within the hardware security module to avoid the risk of key leakage and further improve data security.

[0078] In some embodiments, for large-volume data, an incremental recovery mechanism can be used to prioritize the recovery of key segments of the data, thereby improving the access response speed of large files.

[0079] The blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks described in this embodiment achieves the following technical effects: First, it possesses reliable resistance to backdoor attacks. By deploying a firewall on the output link of the key generation center, and performing randomization processing on the master public key during the system initialization phase and on a portion of the user's private key during the key generation phase, the randomized output is statistically indistinguishable from the original output. Even if a backdoor is implanted in the algorithm of the key generation center, attackers cannot extract any valid information from the public parameters and the distributed partial private keys, thus constructing a certificateless cryptographic system with backdoor defense capabilities.

[0080] Second, it achieves deep integration of multi-cloud redundant storage and secure deduplication of encrypted data. A convergent encryption algorithm ensures consistent ciphertext generation for identical original data, supporting ciphertext-level data deduplication and effectively reducing cloud storage overhead. Combined with the Ramp secret sharing scheme, ciphertext is fragmented and stored on multiple independent cloud servers, avoiding the risk of single-point-of-failure on a single cloud server. Simultaneously, a two-layer security defense is constructed through security thresholds and recovery thresholds: malicious cloud servers colluding within the security threshold cannot obtain valid ciphertext information; even if more than the recovery threshold number of cloud servers colluding to recover the ciphertext, the original plaintext cannot be decrypted without the convergent key, comprehensively ensuring data security in a multi-cloud storage environment.

[0081] Third, it enables decentralized and efficient batch auditing. By relying on blockchain and smart contracts to replace traditional third-party auditing institutions, it automates the entire process of audit challenge issuance, proof verification, and result storage, eliminating single points of failure and trust risks inherent in centralized auditing. Through an on-chain reuse mechanism, the same data block only needs to be challenged once within the same audit period, and other users storing the data can directly reuse the challenge results, significantly reducing the computational and communication overhead of the audit process in multi-user scenarios. At the same time, an on-chain arbitration mechanism ensures the fairness and traceability of the audit process.

[0082] Fourth, a dual-protection data ownership verification mechanism is constructed. For scenarios involving repeated uploads, a dual verification mode combining data ownership proof verification and Merkle tree random challenge verification is adopted. From two dimensions—content hash matching and data structure verification—it confirms that the requesting user indeed possesses the complete original data. Attackers cannot bypass the first layer of verification by forging data ownership proofs, nor can they pass the Merkle root verification without possessing the original data, effectively preventing the security risk of unauthorized users illegally obtaining fragment pointers.

[0083] Fifth, it enables lightweight, certificate-free key management. It avoids the complexity of certificate management in traditional public key infrastructures and the key escrow issues of identity-based cryptography. A user's complete private key consists of a locally generated exclusive secret value and a partial private key issued by the key generation center. The key generation center cannot obtain the user's complete private key; furthermore, the partial private key needs to be randomized by a firewall before redistribution. This reduces the system's key management burden and eliminates the security risks of key escrow at the source.

[0084] Example 2 A blockchain-based certificateless multi-cloud data auditing and deduplication system resistant to backdoor attacks, comprising: The initialization module is used to generate a randomized master public key and incorporate it into the system's public parameters for publication on the blockchain; The key generation module is used to generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate a complete user private key. The data upload module is used to generate ciphertext and deduplication tags for the data to be uploaded based on a convergent encryption algorithm, and to determine whether the data is duplicated by using the deduplication tags. During the first upload, the ciphertext is secretly shared to obtain multiple data fragments. Data block tags and ownership certificates are generated for each data fragment using the user's private key. Each data fragment is distributed to multiple cloud servers for storage, and fragment pointers returned by each cloud server are obtained. During repeated uploads, the user's ownership is verified by the ownership certificate. After successful verification, the fragment pointers are obtained. The data audit module is used to publish an audit challenge set based on data block tags through a blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate an audit certificate. Users verify the integrity of the audit certificate and store the audit results on the blockchain. The data download module is used to download valid data fragments from the cloud server that has passed integrity verification based on the audit results, reconstruct the ciphertext through the secret sharing recovery algorithm, and decrypt to obtain the original data.

[0085] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0086] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0087] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks as described in any of the above embodiments.

[0088] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection (including Wi-Fi, Bluetooth, Bluetooth Low Energy, and IEEE 802.11s-based communication technology).

[0089] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks as described in any of the above embodiments.

[0090] For descriptions of relevant parts of the blockchain-based certificateless multi-cloud data auditing and deduplication system, electronic device, and computer-readable storage medium provided in this application's embodiments, please refer to the detailed descriptions of the corresponding parts in the blockchain-based certificateless multi-cloud data auditing and deduplication method provided in this application's embodiments; they will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0091] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A blockchain-based, certificate-free multi-cloud data auditing and deduplication method resistant to backdoor attacks, characterized in that, Includes the following steps: Step 1: Generate a randomized master public key and include it in the system's public parameters, then publish it to the blockchain; Step 2: Generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate the complete user private key. Step 3: Generate ciphertext and deduplication tags for the data to be uploaded based on the convergent encryption algorithm, and determine whether the data is duplicated by using the deduplication tags; During the initial upload, the encrypted text is secretly shared to obtain multiple data fragments. The user's private key is used to generate data block tags and ownership certificates for each data fragment. The data fragments are then distributed to multiple cloud servers for storage, and the fragment pointers returned by each cloud server are obtained. When uploading repeatedly, the user's ownership is verified through the ownership proof, and the fragment pointer is obtained after successful verification; Step 4: Based on the data block tags, publish the audit challenge set through the blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate audit proof. The user verifies the integrity of the audit proof and stores the audit results on the blockchain. Step 5: Based on the audit results, download valid data fragments from the cloud server that has passed integrity verification, reconstruct the ciphertext using the secret sharing recovery algorithm, and decrypt to obtain the original data.

2. The blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks according to claim 1, characterized in that, In step 1, the key generation center generates the original master public key, and the firewall deployed at the output end of the key generation center selects randomization parameters and performs randomization operation on the original master public key to generate the randomized master public key; The firewall privately stores the randomization parameters.

3. The blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks according to claim 1, characterized in that, In step 2, the user selects a secret random number and combines it with their real identity to generate a pseudo-identity; the key generation center generates a partial private key based on the pseudo-identity; the firewall performs randomization processing on the partial private key using locally stored randomization parameters. The user combines the randomized portion of the private key with the local secret value to generate the complete user private key, and calculates the user public key based on the local secret value and system public parameters.

4. The method for deduplication and certificate-free multi-cloud data auditing based on blockchain to resist backdoor attacks, as described in claim 1, is characterized in that... The secret sharing described in step 3 adopts the Ramp secret sharing scheme, which sets the total number of shares n, the recovery threshold k, and the security threshold r. The ciphertext is divided into kr information fragments and r random fragments, which are encoded into n data fragments. When an attacker obtains no more than r fragments, he cannot obtain any valid information from the ciphertext.

5. A blockchain-based, certificate-free multi-cloud data auditing and deduplication method for resisting backdoor attacks, as described in claim 1, is characterized in that... Ownership verification during repeated uploads in step 3 includes: The first layer of verification involves the cloud server comparing the ownership proof submitted by the user with the ownership proof stored during the initial upload to verify the consistency of the ownership proof. The second layer of verification involves the cloud server randomly selecting a data block index, the user returning the Merkle authentication path and response value for the corresponding data block, and the cloud server reconstructing the Merkle root and comparing it with the root hash stored on the blockchain. After both layers of verification pass, the shard pointer is obtained.

6. The method for deduplication and certificate-free multi-cloud data auditing based on blockchain to resist backdoor attacks, as described in claim 1, is characterized in that... In step 4, within the same audit period, the challenge set of the same data block is reused by multiple users who store the data; the users verify the audit proof based on the bilinear pairing equation.

7. The blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks according to claim 1, characterized in that, In step 5, the user verifies the integrity of each downloaded fragment by using data block tags; marks the cloud server corresponding to the fragment that fails verification as a malicious node and publishes it to the blockchain; and downloads fragments from other cloud servers to replace them until a number of valid fragments that meet the recovery threshold are collected.

8. A blockchain-based, certificate-free multi-cloud data auditing and deduplication method for resisting backdoor attacks, as described in claim 1, is characterized in that... In step 3, after the initial upload is completed, the transaction smart contract jointly signed by the user and multiple cloud servers is broadcast to the blockchain for evidence storage. The transaction smart contract records the storage location of the data fragments, the data block tags, and the ownership proof.

9. A blockchain-based certificateless multi-cloud data auditing and deduplication system resistant to backdoor attacks, characterized in that, include: The initialization module is used to generate a randomized master public key and incorporate it into the system's public parameters for publication on the blockchain; The key generation module is used to generate a user key based on the user's pseudo identity. The key generation center generates a partial private key corresponding to the pseudo identity based on the master private key. The firewall randomizes the partial private key and returns it to the user. The user combines the randomized partial private key with the local secret value to generate a complete user private key. The data upload module is used to generate ciphertext and deduplication tags for the data to be uploaded based on a convergent encryption algorithm, and to determine whether the data is duplicated by using the deduplication tags. During the first upload, the ciphertext is secretly shared to obtain multiple data fragments. Data block tags and ownership certificates are generated for each data fragment using the user's private key. Each data fragment is distributed to multiple cloud servers for storage, and fragment pointers returned by each cloud server are obtained. During repeated uploads, the user's ownership is verified by the ownership certificate. After successful verification, the fragment pointers are obtained. The data audit module is used to publish an audit challenge set based on data block tags through a blockchain smart contract. Multiple cloud servers aggregate the corresponding data block tags and challenge factors according to the data block index in the challenge set to generate an audit certificate. Users verify the integrity of the audit certificate and store the audit results on the blockchain. The data download module is used to download valid data fragments from the cloud server that has passed integrity verification based on the audit results, reconstruct the ciphertext through the secret sharing recovery algorithm, and decrypt to obtain the original data.

10. An electronic device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the blockchain-based certificateless multi-cloud data auditing and deduplication method for resisting backdoor attacks as described in any one of claims 1 to 8.

Images