Lightweight equipment-oriented cloud storage data integrity verification method
By using a trusted execution environment and blockchain technology on cloud servers, combined with the Cuckoo Filter and smart contracts, the problems of data privacy protection and multiple verification for lightweight devices in cloud storage are solved. This enables real-time, multiple data integrity verifications, reduces device storage load, and ensures the reliability of verification.
Patent Information
- Application Number
- CN202410970873.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-07-19
AI Technical Summary
In edge-cloud scenarios, data privacy protection and data integrity verification for lightweight devices present challenges, especially since data cannot be verified multiple times after being stored on cloud servers, and existing solutions rely on unreliable third-party auditors.
By employing a trusted execution environment and blockchain technology, data integrity verification is performed by creating an isolated trusted zone in a cloud server. The Cuckoo Filter and blockchain are used for evidence storage and auditing to ensure that different session keys are used for each verification, and data integrity verification is performed in conjunction with smart contracts.
This enables real-time, multiple verifications of data stored on cloud servers while ensuring data privacy. It reduces the storage load on edge devices and uses blockchain to ensure the traceability and fairness of verification, thus avoiding disputes.
Smart Images

Figure CN118965449A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data security technology, and in particular to a cloud storage data integrity verification method for lightweight devices. Background Art
[0002] In the edge-cloud scenario, edge devices such as mobile phones, sensors, and drones are usually resource-constrained, but they need to continuously collect external data, which generates high computing power and storage requirements. However, the hardware configuration of these devices cannot meet the requirements, so the data needs to be transferred to a centralized cloud server for storage or computing to release the storage load. This hosting model faces many security and privacy challenges. First, the data of the edge device needs to be privately protected and should not be transmitted to the cloud server in plain text. Second, the integrity of the cloud server data must be guaranteed to prevent the cloud server from discarding part or all of the data for cost-saving reasons. In addition, lightweight edge devices usually delete local data after data hosting, but the stored data still needs to be verified multiple times. This requires a mechanism that allows users to verify the integrity of their data at any time without retaining the original data.
[0003] The first challenge can be solved by encryption. Users encrypt data before sending it. The second challenge can be achieved with digital signatures. The hash value of the data is generated by a hash function to verify the data integrity. However, a simple signature mechanism is not suitable for the third challenge, because after the end-side sends its own data, the original data is often no longer stored locally, and the signature needs to compare the digest to see if it is consistent, which requires the hash value of the original data. The cloud server only needs to save the hash value of the original data to pass the signature verification challenge. Therefore, a simple signature mechanism cannot meet the multiple arbitrary verification requirements of the end-side. The existing cloud storage data integrity verification scheme involves the role of a third-party auditor (TPA). Their scheme requires that the third-party auditor is trustworthy, which is difficult to achieve in actual situations. Summary of the invention
[0004] Purpose of the invention: The purpose of the present invention is to provide a cloud storage data integrity verification method for lightweight devices, which can realize real-time verification of cloud-side stored data while ensuring the privacy of the data, and ensure that the cloud can only honestly store complete data.
[0005] Technical solution: A cloud storage data integrity verification method for lightweight devices. First, it should ensure that the data owner and the cloud server register and deposit on the blockchain smart contract; at the same time, the data owner generates a master key locally; it includes a data storage stage and a data verification stage: the data storage stage is that the data owner sends the data securely and privately to the cloud server for storage; the data verification stage refers to the data owner performing data integrity verification on the data stored in the cloud server after the storage stage is completed; and the calculation of the cloud data integrity proof during the data storage or verification stage is performed in a trusted execution environment, and the session key for each storage or verification is different.
[0006] Furthermore, the data storage phase includes the following steps:
[0007] SA1: The data owner performs remote authentication with the trusted execution environment and negotiates a session key. The session key and the master key are used to encrypt the data to be managed and generate a corresponding data identifier. The data owner sends the encrypted data to be managed and the data identifier to the cloud server.
[0008] SA2: After the cloud server stores the encrypted managed data and data identifier, it loads the encrypted managed data and data identifier into the trusted execution environment, uses the isolated security zone provided by the trusted execution environment to decrypt the managed data and performs a hash operation on the managed data and managed data identifier. The resulting hash value is encrypted with the session key as proof of cloud storage integrity.
[0009] SA3, the trusted execution environment generates an authentication report and sends it to IAS for verification. The verification report, certificate, and signature of the verification report returned by IAS are packaged as the computational correctness proof of the encrypted cloud storage integrity proof. The encrypted cloud storage integrity proof and computational correctness proof are then submitted to the smart contract in the blockchain. If the computational correctness proof is verified, the cloud storage integrity proof is stored on the blockchain.
[0010] SA4, the data owner obtains the cloud storage integrity certificate from the blockchain and decrypts it to obtain the plaintext hash; then the plaintext hash is used as input to perform the cuckoo filter's insert storage operation.
[0011] Furthermore, in step SA1, the data owner calculates the data identifier of the hosted data through a hash algorithm.
[0012] Furthermore, the data verification phase includes the following steps:
[0013] SB1, the data owner performs remote authentication with the trusted execution environment and negotiates a session key; the data identifier that needs to be verified for integrity is sent to the cloud server;
[0014] SB2, the cloud server finds the encrypted data to be hosted according to the data identifier, loads the encrypted data to be hosted and the data identifier into the trusted execution environment, uses the isolated security zone provided by the trusted execution environment to decrypt the hosted data and performs a hash operation on the hosted data and the hosted data identifier, and the obtained hash value is encrypted with the session key as the cloud storage integrity proof;
[0015] SB3, the trusted execution environment generates an authentication report and sends it to the IAS for verification. The verification report, certificate, and signature of the verification report returned by the IAS are packaged as the computational correctness proof of the encrypted cloud storage integrity proof. The cloud storage integrity proof and computational correctness proof are then submitted to the smart contract in the blockchain. If the computational correctness proof is verified, the cloud storage integrity proof is stored on the blockchain.
[0016] SB4, the data owner obtains the cloud storage integrity proof from the blockchain and decrypts it to obtain the plaintext hash; then the plaintext hash is used as the query input to execute the query operation of the cuckoo filter: if the query is successful, the integrity verification of the data in the cloud server is successful; if the query fails, the integrity verification of the data in the cloud server fails.
[0017] Compared with the prior art, the present invention has the following significant effects:
[0018] 1. The present invention uses the trusted execution environment technology to design a cloud data integrity verification scheme to ensure that the session key for each storage or verification is different, so that the cloud server cannot pass the next verification by saving the integrity verification certificate. Under the premise of ensuring the privacy of the data, the instant verification of the data stored in the cloud server is realized, ensuring that the cloud server can only honestly save complete data;
[0019] 2. During storage, the client locally saves the decrypted cloud data integrity proof and adds it to the Cuckoo filter. During verification, the decrypted cloud data integrity proof is used as input to the Cuckoo filter for search. If the search result is that the data exists, the verification is successful. In addition to high search efficiency, compared with traditional storage data structures, the Cuckoo filter usually requires less memory space to store data sets of similar size, which also reduces the storage load for client devices.
[0020] 3. Blockchain is used to audit and store integrity verification, ensuring the traceability and fairness of verification, thereby avoiding potential disputes and denials. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 It is a flowchart of the implementation process of the storage stage in the present invention;
[0022] Figure 2 It is a flowchart of the implementation process of the verification phase in the present invention;
[0023] Figure 3 It is a structural flow chart of the first execution of the storage phase in the present invention;
[0024] Figure 4 It is a structural flow chart of the verification stage in the present invention. DETAILED DESCRIPTION
[0025] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0026] Trusted Execution Environment (TEE) provides a method to solve the challenges mentioned in the prior art. By creating a trusted area in the cloud server that is inaccessible to the cloud server, the ciphertext is loaded into it for verification calculation and generates a transport bureau integrity certificate, which ensures the confidentiality of the data and prevents the cloud server from cheating during the verification process. The present invention proposes a solution to the needs of privacy protection and data integrity assurance in cloud storage. The solution combines the trusted execution environment technology and the blockchain technology to provide a reliable mechanism for real-time and multiple verification of data under the premise of satisfying data privacy protection.
[0027] In this embodiment, the participating entities involved are as follows:
[0028] 1) Data Owner (DO, i.e., end-side)
[0029] It is a lightweight end-side device in the end-cloud scenario with limited computing power and storage capabilities, and has the business need to host local data for storage.
[0030] 2) Cloud Server (CS, i.e. Cloud Side)
[0031] It is a cloud-side server in the end-cloud scenario, with huge computing power and storage capabilities, and the ability and business needs to receive end-side data for storage. In addition, the cloud server is also equipped with hardware devices that support a trusted execution environment.
[0032] 3) Trusted Execution Environment (TEE)
[0033] Although TEE is not an independent entity, in order to highlight its isolation from the cloud server, TEE is logically separated from the cloud server for explanation. TEE is the core code used to perform data integrity verification. Due to its confidentiality, even the cloud server (TEE's external host) cannot access the code and data in the isolated space it opens up, and can only obtain the execution result of the integrity check. The present invention uses Intel's SGX (Software Guard Extensions) as the TEE instance used.
[0034] 4) Blockchain (BC)
[0035] As a distributed database, blockchain provides a decentralized storage trust root for the data integrity verification results generated by TEE. At the same time, based on the programmability of the smart contracts it supports, it automatically verifies, audits and arbitrates the results.
[0036] This avoids potential transaction disputes between the two parties.
[0037] The present invention uses a trusted execution environment to calculate the cloud data integrity proof and the session key for each storage or verification is different. Since the cloud data integrity proof is encrypted with the session key, the cloud server cannot pass the next verification by saving the cloud data integrity proof. The cloud data integrity verification scheme based on the trusted execution environment realizes the instant verification of the data stored on the cloud server while ensuring the privacy of the data, ensuring that the cloud server can only honestly save complete data.
[0038] Considering the computing power and storage capacity issues of lightweight end-side devices, as terminal devices such as sensors need to generate data continuously, as the amount of data grows, the retrieval efficiency of signature verification of specified data is relatively low. In order to provide a fast and efficient verification mechanism that is suitable for lightweight devices, a lightweight and efficient end-side verification scheme based on a cuckoo filter is proposed. During storage, the end-side locally saves the hash value of the managed data and the managed data identifier and adds it to the cuckoo filter. During verification, the decrypted hash value is used as input and entered into the cuckoo filter for search. If the result is returned, it proves that the verification is successful. In addition to the higher search efficiency, compared with traditional storage data structures, cuckoo filters usually require less memory space to store data sets of similar size, which also reduces the storage load for end-side devices.
[0039] The blockchain is used to audit and store integrity verification, ensuring the traceability and fairness of verification, thereby avoiding potential disputes and denials.
[0040] The cloud storage data integrity verification method of the present invention is mainly divided into two stages: storage stage and verification stage. The storage stage is the process in which the data owner sends the data to the cloud server securely and privately for storage; and the verification stage refers to the process in which the data owner performs data integrity verification on the data stored in the cloud server after the storage stage is completed. The specific implementation process is as follows:
[0041] 1. Preparation
[0042] Before the data owner and the cloud server interact, several preparations need to be done. First, the smart contract and SGX application Enclave need to be deployed and started. DO and CS register and deposit on the smart contract. At the same time, DO generates the master key msk locally.
[0043] (II) Storage stage
[0044] like Figure 1 As shown in the figure, the implementation process of the storage phase is as follows:
[0045] Step A1: DO and SGX perform remote authentication and negotiate a session key ssk.
[0046] DO calculates the data identifier d based on the data to be managed id :
[0047] d id =Hash(data)
[0048] Wherein, Hash() represents a hash algorithm, and the SHA256 hash algorithm is adopted in the present invention.
[0049] Encrypt the managed data data with the master key msk to obtain the encrypted managed data, denoted as ct data :
[0050] ct data =Enc msk (data)
[0051] Among them, Enc msk () indicates encryption with the master key msk. The present invention adopts the symmetric encryption algorithm AES for encryption. The session key ssk is used to encrypt msk to obtain the encrypted master key, which is recorded as ct msk :
[0052] ct msk =Enc ssk (msk)
[0053] Among them, Enc ssk () indicates encryption with session key ssk, and the present invention adopts symmetric encryption algorithm AES for encryption.
[0054] If the data is stored for the first time, the data owner DO needs to id ,ct data ,ct msk ) is sent to the cloud server CS; if the data has been stored before, (d id ,ct data ) is sent to the cloud server CS, such as Figure 3 shown.
[0055] Step A2: After receiving the data from the data owner, CS will id ,ct data ) is stored.
[0056] The implementation process is as follows:
[0057] Step A21, if it is the first time to perform storage, the cloud server will (d id ,ct data ,ct msk ) is loaded into SGX, SGX uses ssk to decrypt msk and obtain the master key msk:
[0058] msk=Dec ssk (ct msk )
[0059] Among them, Dec ssk () indicates decryption using session key ssk.
[0060] And by utilizing the Sealing mechanism of SGX, the master key msk used to encrypt data is sealed locally on the execution host, and only the SGX trusted zone can decrypt it, thus avoiding the need to transmit the master key again.
[0061] Next, use the decrypted msk to decrypt ct data , restore the data to be hosted:
[0062] data=Dec msk (ct data )
[0063] Among them, Dec msk () indicates decryption using the master key msk, and the present invention uses the symmetric encryption algorithm AES for encryption.
[0064] Step A22: If it is not the first time to perform storage, the cloud server will (d id ,ct data) is loaded into SGX, and the master key msk used to encrypt data is decapsulated from the execution host using SGX's Sealing mechanism. Next, the decapsulated msk is used to ct data Decrypt and restore the data to be hosted:
[0065] data=Dec msk (ct data )
[0066] Among them, Dec msk () indicates decryption using the master key msk, and the present invention uses the symmetric encryption algorithm AES for encryption.
[0067] Step A23, then judge d id Is it equal to Hash(data)?
[0068] If they are not equal, a storage error is returned;
[0069] If they are equal, continue to execute hash data =Hash(data||d id ),ct hash =Enc ssk (hash data );
[0070] Among them, ct hash Indicates that the session key ssk encrypts data and d id The result after the hash value is used as the cloud data integrity proof. id The hash value of the session key ssk and the encrypted data are processed using the SHA256 hash algorithm and the AES symmetric encryption algorithm respectively.
[0071] Step A3, SGX performs calculations to generate an authentication report σ tee , certification report σ tee With ct hash Output together.
[0072] The certification report tee It can be sent to Intel's third-party service IAS (International Accreditation Service) for verification and return the verification report vr, certificate ca, and signature of the verification report σ IAS , will verify the report vr, certificate ca, signature σ IAS Packing is done to obtain the proof of computational correctness π:
[0073] π=(vr,ca,σ IAS )
[0074] Among them, vr contains the pair σ tee The verification results of validity and related information of SGX, σ IAS is the IAS to σ tee And the signature of the verification result, ca contains the public key of the signature.
[0075] Then π and ct hash Submit to the smart contract in the blockchain, where π is used as ct hash The computational correctness of π is proved. Since the signature σ contained in π IAS There is no need for trusted hardware, nor is there a need to contact the IAS again to verify it, so the smart contract can parse the signature public key from the certificate CA and verify the σ IAS After the signature verification is successful, the smart contract parses the verification report vr and compares the information in it with the trusted configuration. After the comparison is completed, the smart contract will receive ct hash To keep evidence.
[0076] Step A4, DO can obtain ct from the blockchain hash , and decrypt it with the session key ssk to get the plaintext hash hash data =Dec ssk (ct hash ) and store it in its local Cuckoo filter for subsequent verification.
[0077] (III) Verification phase
[0078] like Figure 2 As shown in the figure, the implementation process of the verification phase is as follows:
[0079] Step B1, DO and SGX perform remote authentication and negotiate a session key ssk′.
[0080] The data identifier of the managed data that DO will verify the integrity of. id Sent to the cloud server CS, the cloud server CS identifies the data according to d id , find the corresponding data (d id ,ct data ).
[0081] Step B2, CS will (d id ,ct data ) is loaded into SGX, and the master key msk used to encrypt data is decapsulated from the execution host locally using the SGX encapsulation mechanism.
[0082] Next, use the decapsulated master key msk to decrypt ct data , thus recovering data data = Dec msk (ctdata ).
[0083] Then determine the data identifier d id Is it equal to Hash(data):
[0084] If they are not equal, a validation error is returned;
[0085] If they are equal, continue to execute hash data =Hash(data||d id ), ct′ hash =Enc ssk′ (hash data );
[0086] Among them, ct′ hash Indicates that the session key ssk′ encrypts data and d id The result is the hash value of the cloud data and serves as proof of cloud data integrity.
[0087] Step B3, SGX performs calculations to generate an authentication report σ′ tee , certification report σ′ tee and ct′ hash Output together.
[0088] The authentication report is sent to Intel's third-party service IAS for verification and returns the verification report vr′, certificate ca′, and signature σ′ for the verification report IAS , the verification report vr′, certificate ca′, signature σ′ IAS Packing is done to obtain the proof of calculation correctness π′:
[0089] π′=(vr′,ca′,σ′ IAS )
[0090] Then π′ and ct′ hash Submit to the smart contract in the blockchain, where π′ is used as ct′ hash The computational correctness of is proved. Since the signature σ′ contained in π′ IAS There is no need for trusted hardware, nor is there any need to contact the IAS again to verify it, so the smart contract can parse the signature public key from the certificate ca′ and verify σ′ IAS After the signature verification is successful, the smart contract parses the verification report vr′ and compares the information in it with the trusted configuration. After the comparison is completed, the smart contract will receive ct′ hash To keep evidence.
[0091] Step B4, DO can obtain ct′ from the blockchain hash , and decrypted with ssk′ to get the plaintext hash hash′data =Dec ssk′ (ct′ hash ). Then hash data ′ is used as the query input to execute the query operation of the cuckoo filter. If the query succeeds, the integrity verification of the data in the cloud server succeeds. If the query fails, the integrity verification of the data in the cloud server fails.
Claims
1. A cloud storage data integrity verification method for lightweight devices, first of all, ensure that the data owner and the cloud server register and deposit on the blockchain smart contract; at the same time, the data owner generates a master key locally, characterized in that, It includes the data storage phase and the data verification phase: the data storage phase is when the data owner sends the data securely and privately to the cloud server for storage; The data verification phase refers to the data integrity verification performed by the data owner on the data stored in the cloud server after the storage phase is completed; and the calculation of the cloud data integrity proof during the data storage or verification phase is performed in a trusted execution environment, and the session key for each storage or verification is different.
2. The cloud storage data integrity verification method for lightweight devices according to claim 1 is characterized in that: The data storage phase includes the following steps: SA1: The data owner performs remote authentication with the trusted execution environment and negotiates a session key. The session key and the master key are used to encrypt the data to be managed and generate a corresponding data identifier. The data owner sends the encrypted data to be managed and the data identifier to the cloud server. SA2: After the cloud server stores the encrypted managed data and data identifier, it loads the encrypted managed data and data identifier into the trusted execution environment, uses the isolated security zone provided by the trusted execution environment to decrypt the managed data and performs a hash operation on the managed data and managed data identifier. The resulting hash value is encrypted with the session key as proof of cloud storage integrity. SA3, the trusted execution environment generates an authentication report and sends it to IAS for verification. The verification report, certificate, and signature of the verification report returned by IAS are packaged as the computational correctness proof of the encrypted cloud storage integrity proof. The encrypted cloud storage integrity proof and computational correctness proof are then submitted to the smart contract in the blockchain. If the computational correctness proof is verified, the cloud storage integrity proof is stored on the blockchain. SA4, the data owner obtains the cloud storage integrity certificate from the blockchain and decrypts it to obtain the plaintext hash; Then the insert storage operation of the cuckoo filter is performed with the plaintext hash as input.
3. The cloud storage data integrity verification method for lightweight devices according to claim 2 is characterized in that: In step SA1, the data owner calculates the data identifier of the hosted data through a hash algorithm.
4. The cloud storage data integrity verification method for lightweight devices according to claim 1, characterized in that: The data verification phase includes the following steps: SB1, the data owner performs remote authentication with the trusted execution environment and negotiates a session key; the data identifier that needs to be verified for integrity is sent to the cloud server; SB2, the cloud server finds the encrypted data to be hosted according to the data identifier, loads the encrypted data to be hosted and the data identifier into the trusted execution environment, uses the isolated security zone provided by the trusted execution environment to decrypt the hosted data and performs a hash operation on the hosted data and the hosted data identifier, and the obtained hash value is encrypted with the session key as the cloud storage integrity proof; SB3, the trusted execution environment generates an authentication report and sends the report to the IAS for verification, and packages the verification report, certificate, and signature of the verification report returned by the IAS as the computational correctness proof of the encrypted cloud storage integrity proof; Then the cloud storage integrity proof and the computational correctness proof are submitted to the smart contract in the blockchain. If the computational correctness proof is verified, the cloud storage integrity proof is stored on the blockchain. SB4, the data owner obtains the cloud storage integrity certificate from the blockchain and decrypts it to obtain the plaintext hash. The plaintext hash is then used as the query input to perform the cuckoo filter query operation: if the query is successful, the integrity verification of the data on the cloud server is successful. If the query fails, the integrity verification of the data on the cloud server fails.
Citation Information
Patent Citations
Cloud storage big data integrity verification method for mobile terminal device in internet of things
CN109104440A
Data integrity verification method supporting data privacy protection in cloud storage
CN110365469A
Privacy-protecting outsourcing Internet-of-things data based on blockchain and backup integrity checking method thereof
CN112054897A
Untrusted third-party cloud storage ciphertext deduplication method and system based on SGX
CN115484031A
Public verifiable outsourcing data management method without trusted third party
CN116992467A
