Method for verifying data integrity and locating errors of multi-cloud storage in low-altitude unmanned aerial vehicle network

Abstract

The application discloses a kind of low altitude unmanned aerial vehicle network in multi-cloud storage data integrity verification and error positioning method, data integrity verification is in low altitude unmanned aerial vehicle network, after generating user private key and public key by comprehensive supervision service platform registration, stored data is sent to multiple cloud servers to carry out multi-cloud storage data;The low altitude unmanned aerial vehicle data stored on multiple cloud servers is executed integrity verification, including: the first smart contract on the blockchain generates audit challenge according to chain state and issues;Each cloud server generates response proof according to the stored encoding data block and issues to chain;Verification contract listens to and aggregates all response proofs, batch verification is carried out, and the result is published in audit log.The method constructs a decentralized audit framework and does not rely on third-party auditors.If verification fails, a positioning mechanism can be triggered to accurately locate damaged data.To protect identity privacy, a key exchange protocol is embedded to achieve conditional anonymity.The supervision platform can balance anonymity and accountability.

Description

Technical Field

[0001] This invention relates to the field of distributed cloud storage technology, and in particular to a method for verifying the integrity of multi-cloud storage data and locating errors in low-altitude unmanned aerial vehicle (UAV) networks. Background Technology

[0002] With the acceleration of digitalization and the booming development of the low-altitude economy, Unmanned Aerial Vehicles (UAVs), with their high mobility and high-precision sensors such as visible light cameras, LiDAR, and multispectral / infrared imaging, have become core equipment for low-altitude information sensing and acquisition, widely used in military reconnaissance, geographic mapping, emergency rescue, environmental monitoring, and personal aerial photography. During a single long-duration mission, UAVs can generate terabytes of high-resolution, full-element, multimodal mission data, placing enormous pressure on local storage. Traditional solutions relying on expanding ground servers or building their own data centers suffer from high costs, complex operation and maintenance, and the potential for information silos, making it difficult to support cross-domain collaboration and centralized management. In contrast, cloud storage, with its high scalability, elastic resource scheduling, and professional data management capabilities, provides UAVs with an efficient and persistent solution for massive data storage, supporting remote access, mission backtracking, and subsequent intelligent analysis, and has become a key infrastructure for UAV mission data storage, management, and sharing.

[0003] However, the cloud environment itself carries inherent risks: cloud servers may suffer data corruption due to hardware failure, software bugs, or human error; in high-confrontation scenarios, cloud service providers (CSPs) may also conceal data anomalies or even maliciously tamper with or delete critical data, compromising the authenticity and integrity of the data, driven by profit motives (including reputation and cost considerations). Therefore, building a secure and reliable cloud storage mechanism and achieving effective integrity verification of UAV mission data has become a core requirement for ensuring the safety of low-altitude missions.

[0004] Data integrity verification mechanisms typically introduce a third-party auditor (TPA) to check the integrity of cloud data on behalf of the user (drone device) without downloading the full data. While existing research has proposed numerous data integrity verification schemes, most are based on a single-cloud storage architecture, posing a risk of permanent data loss due to a single point of failure. To address this, a multi-cloud, multi-replica integrity verification mechanism has been proposed: multiple copies of the same data file are generated, and these copies are distributed to different cloud servers through a centralized cloud broker. During integrity verification, the TPA sends a challenge message to the cloud broker, which forwards it to each cloud server and aggregates the returned response proofs.

[0005] While this mechanism improves the reliability and durability of data storage, storing multiple identical copies results in significant waste of storage resources. During the auditing process, repeatedly transmitting multiple proofs greatly increases network bandwidth overhead and reduces overall system efficiency. Furthermore, in terms of security, existing solutions struggle to effectively defend against collusion attacks between cloud servers. Because each server stores the same copy and the audit challenges are identical, dishonest cloud agents can selectively forward challenges and copy and paste valid proofs from other servers to deceive the TPA into performing false verifications. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for verifying the integrity of multi-cloud storage data and locating errors in low-altitude unmanned aerial vehicle (UAV) networks. Based on Reed-Solomon erasure coding, identity-based cryptography, homomorphic hash functions, binary search algorithms, uniform (K,N) sets on finite sets, and blockchain technology, the core of this invention is to construct an architecture for encoding-driven distributed data storage, integrity verification, and accurate location and effective recovery of damaged data.

[0007] The objective of this invention is achieved through the following technical solution:

[0008] Firstly, this application discloses a method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle (UAV) network.

[0009] In a low-altitude drone network, drones register through a comprehensive regulatory service platform to generate user private and public keys, and then send their stored data to multiple cloud servers for multi-cloud data storage. Integrity verification is then performed on the low-altitude drone data stored on these cloud servers, including: a first smart contract deployed on the blockchain generates an audit challenge for multiple cloud servers and publishes the audit challenge as a first transaction to the blockchain; multiple cloud servers storing target data encoded data blocks each execute a response proof generation phase, with each cloud server monitoring the blockchain and, upon receiving the audit challenge from the first transaction, generating a response proof based on its stored encoded data blocks and publishing its response proof as a second transaction to the blockchain; and a verification contract deployed on the blockchain executes a response proof verification phase, monitoring and acquiring multiple response proofs from all cloud servers within a preset time period, performing batch verification of the aggregated multiple response proofs according to a predetermined verification algorithm to determine the integrity of the target data, and publishing an audit log containing the verification results as a third transaction to the blockchain.

[0010] Secondly, this application also discloses a method for locating errors in multi-cloud storage data in a low-altitude unmanned aerial vehicle (UAV) network. This method is based on the aforementioned method for verifying the integrity of multi-cloud storage data in a low-altitude UAV network. When the verification result determined in the response proof verification stage is "fail," the following steps are performed: For low-altitude UAV data stored on multiple cloud servers in the low-altitude UAV network, a location contract deployed on the blockchain is used to locate the faulty cloud servers. The set of multiple cloud servers undergoes integrity verification based on the predetermined verification algorithm until all faulty cloud servers with incomplete data storage are located. A faulty cloud server location entry is then set through the location contract, and a transaction is provided to the blockchain. The located faulty cloud servers locate damaged coded data blocks. After the faulty cloud server listens to the blockchain and obtains the transaction it provides, it performs fragment verification on its local storage to determine the specific damaged coded data blocks.

[0011] The beneficial effects of this invention are:

[0012] The method in this application eliminates the reliance on centralized cloud agents or third-party auditors, constructing a decentralized multi-cloud public audit framework coordinated by smart contracts. When the audit result is false, a progressive three-layer precise positioning mechanism is triggered to accurately locate the damaged coded data block, thereby achieving high precision in fault diagnosis and data regeneration. To protect the identity privacy of the drone, a secure key exchange protocol is embedded to achieve unconditional anonymity. Even if an adversary breaks the indistinguishability of the ciphertext, they cannot know the true origin and identity of the drone. Conversely, once malicious behavior, such as system abuse, is detected, the integrated regulatory service platform can reveal its true identity through the embedded traceability mechanism, effectively balancing anonymity and accountability. Attached Figure Description

[0013] Figure 1 This describes the communication process between the drone, the integrated regulatory service platform, multiple cloud servers, and the blockchain as described in the embodiments of this application. Detailed Implementation

[0014] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] refer to Figure 1 This application describes a method for verifying the integrity of multi-cloud storage data and locating errors in a low-altitude unmanned aerial vehicle (UAV) network according to an embodiment of the present application.

[0016] First, refer to Figure 1The numbers in the diagram represent the execution order. The system shown includes: the Unmanned Aerial Vehicle Operation Management (UOM) integrated management platform for drones and unmanned aerial vehicles, and secure information exchange to generate user public and private key pairs and securely outsource data collection.

[0017] To better understand this, we will first explain a method for anonymizing the identity of low-altitude UAVs and outsourcing secure data, and then, based on this method, we will present a method for verifying the integrity of multi-cloud storage data and locating errors in low-altitude UAV networks.

[0018] The method for anonymizing the identity of low-altitude drones and outsourcing secure data is executed through the integrated regulatory service platform, which includes the following stages: system initialization, user anonymization and private key generation, and data preprocessing and outsourcing.

[0019] Specifically, during the system initialization phase, the integrated monitoring platform, based on given security parameters... Generate system parameters, including the master private key. In the user anonymization and private key generation phase, the drone generates an anonymous identity and submits registration information to the integrated regulatory service platform to obtain the user's private key corresponding to the anonymous identity; in the data preprocessing and outsourcing phase, the drone uploads the data files to be uploaded. The data is divided into multiple data blocks, each data block is further divided into multiple data word blocks, and each data word block is encrypted using a symmetric encryption algorithm to obtain ciphertext sub-blocks. The subscript ij represents the j-th sub-block of the i-th data block. Then, the UAV encodes the ciphertext data block using erasure coding to generate multiple encoding vectors. For each encoded data block in the encoding vector, the UAV calculates its corresponding verification tag. The UAV obtains the number of available cloud servers, calculates the amount of encoded data stored on each cloud server, and then constructs an adapted random distribution record table of encoded data to record the encoding vectors and their verification tags corresponding to each cloud server, so as to allocate and transmit the storage data that the UAV needs to store to one of the multiple cloud servers for storage.

[0020] Next, each stage will be explained in detail.

[0021] System initialization phase: Given a security parameter The integrated regulatory service platform runs an initialization algorithm to generate system parameters, including the system master private key and public parameters. The specific steps are as follows:

[0022] set up and All are large prime numbers of order. Multiplication cyclic group, yes The generator. The integrated regulatory service platform defines a bilinear pair: Select random value ,and Used as the system's master private key, and its corresponding master public key is calculated. , ,in It is a model Remainder ring.

[0023] The integrated regulatory service platform has selected six secure, collision-resistant hash functions: the first hash function The second hash function The third hash function The fourth hash function ,in The fifth hash function is used for the set of natural numbers. The sixth hash function ,in For safety parameters The determined bit length of the hash function output, and the seventh hash function. .

[0024] The integrated regulatory service platform selects a pseudo-random function. and a pseudo-random generator .

[0025] The integrated regulatory service platform sets up and publishes system parameters. and securely store the master private key. .

[0026] User anonymity and private key generation phase: Assuming each drone has a unique and authentic identity. The drone first generates an anonymous identity and sends it to the integrated regulatory service platform for registration. The platform then calculates the user's private key. The specific steps are as follows:

[0027] Set an initial login password for the drone. And select a random value And calculate the first user's public key and user anonymity ,in Anonymous identity information The validity period. Finally, the drone submits user registration information to the integrated regulatory service platform via a secure channel. ,in The session key, generated by the drone, is used to ensure the confidentiality of communication during subsequent interactions. This is a timestamp used to indicate when the registration information was generated.

[0028] When the registration information sent by the drone is received Then, the integrated regulatory service platform first uses the master private key. To restore the drone's identity and reset their login password. .

[0029] The integrated regulatory service platform selects a random value. And calculate the second user's public key. and its users' private keys Finally, the integrated regulatory service platform transmits the binary data via a secure channel. Send to the drone.

[0030] When receiving binary data from the comprehensive regulatory service platform Then, the drone verifies the validity of the user's private key using the following equation:

[0031] If the above verification equation holds true, the drone will receive the binary tuple. Otherwise, the drone rejects the binary pair. It also requires the integrated regulatory service platform to regenerate user private keys. Second user public key The question mark (?) in the above formula indicates a judgment on whether the equation is true. The question mark (?) in the formula has the same meaning.

[0032] Data preprocessing and outsourcing stage: Given a data file The drone performs the following algorithm steps to achieve secure data outsourcing.

[0033] For each data file to be uploaded Its file name is The drone first cut it into One data block. Each data block The subscript i represents the i-th data block, which is further divided into Data sub-blocks The subscript ij represents the j-th sub-block of the i-th data block. For each data sub-block... The drone chooses a secure symmetric encryption algorithm. To encrypt it, in order to generate ciphertext sub-blocks ,in Represents a symmetric encryption algorithm A security key. Therefore, the data file Encrypted into ciphertext , It is by Individual Model The data space composed of elements.

[0034] The drone selects an encoding matrix ,in The total number of vectors generated after encoding is given, and the codeword matrix is ​​calculated as follows: , where each encoded vector For each coded data block The drone selects a random value. And calculate its corresponding verification label:

[0035]

[0036] For random values The hash mapping value, The tags generated for the drone based on coded data sub-blocks, random values, and private keys are used for subsequent integrity verification. Therefore, each coded vector... The verification tag is:

[0037] .

[0038] To determine the number of cloud servers available in the system Then, the drone first calculated the amount of coded data stored on each cloud server. The codeword matrix follows. Construct an adapted table of randomly distributed coded data records. It is used to record the encoding vector and its verification label corresponding to each cloud server.

[0039] In some examples, a record table can be randomly distributed based on the coded data. Leaf nodes Then, based on the Merkel hash tree structure, the drones proceed in logical order on each cloud server. Calculate the hash root And thus obtain hash sub-roots Based on Merkel hash tree structure and hash sub-roots The drone can calculate the root hash value. Drone settings auxiliary metadata and generate a blockchain transaction. , For file tags.

[0040] Meanwhile, the drone is set to store data. And send it to the cloud server. ,do Let be the verification label for the i-th encoded vector.

[0041] Upon receiving stored data After that, cloud server Recalculate the hash root And retrieve blockchain transactions from the blockchain. To obtain the hash subroot .

[0042] Subsequently, cloud server Check the equation:

[0043] Check if they are equal. If not, store the data. Storage was denied. Otherwise, the cloud server... Check its correctness using the following formula: .in If the equation for checking correctness holds true, the cloud server... Preserve stored data The cloud server will store the data; otherwise, it will refuse to store the data. .

[0044] As in the aforementioned embodiments, after the blockchain deployment is completed, the integrated regulatory service platform, drones, and server clusters join the blockchain network and obtain a unique blockchain address. , and .

[0045] Furthermore, the integrated regulatory service platform UOM in this application embodiment deploys the following smart contracts on the blockchain: challenge contract. Validate the contract Positioning Contract .

[0046] Therefore, based on the foregoing embodiments, the method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle (UAV) network according to the embodiments of this application, based on the multi-cloud storage data completed by the foregoing method, performs integrity verification on low-altitude UAV data stored on multiple cloud servers, specifically including the following:

[0047] This is achieved through a smart contract deployed on the blockchain. This smart contract is responsible for performing integrity verification on low-altitude drone data stored on multiple cloud servers without downloading the full data. The method includes three phases: an audit challenge generation phase, a response proof generation phase, and a response proof verification phase.

[0048] Specifically, if the verification result determined in the response proof verification phase is a failure, the following actions are taken:

[0049] For low-altitude drone data stored on multiple cloud servers in a low-altitude drone network, a positioning contract deployed on the blockchain is used to locate faulty cloud servers. The set of multiple cloud servers undergoes integrity verification based on a predetermined verification algorithm until all faulty cloud servers with incomplete data storage are located. A faulty cloud server positioning entry is then set through the positioning contract, and a transaction is provided to the blockchain. The located faulty cloud servers then locate damaged coded data blocks. After the faulty cloud server listens to the blockchain to obtain the provided transaction, it performs fragment verification on its local storage to determine the specific damaged coded data blocks. The low-altitude drone data refers to the data entity collected by the low-altitude drone and stored in a multi-cloud environment in the aforementioned embodiment. Specifically, the low-altitude drone data consists of multiple coded data blocks generated by the drone dividing and encrypting the original data file, and encoding it using erasure coding technology, along with a homomorphic verification tag calculated and bound to each coded data block. The coded data blocks and their tags of the low-altitude drone data are distributed and stored on different cloud servers, collectively constituting the target data object for integrity verification and fault location.

[0050] The method will now be explained in detail.

[0051] Audit challenge generation phase: This phase is initiated by the first smart contract deployed on the blockchain. The specific steps for execution are as follows:

[0052] First Smart Contract Based on the current state of the blockchain and the block height, obtain the header hash value of the latest block. Extract the previous ones in reverse order. The block header hash. The state of the preceding blockchain includes at least the hash of the latest block in which the smart contract was invoked, and the hashes of a specific number of historical blocks traced back in sequence. These hashes collectively serve as a publicly verifiable source of randomness to generate unpredictable and tamper-proof audit challenges. First smart contract. Calculate the audit challenge seed key 1:

[0053] And Audit Challenge Seed Key 2:

[0054] .

[0055] In the above two equations, , , All are historical header hash values. This is the fifth hash function;

[0056] First Smart Contract Output Audit Challenges:

[0057]

[0058] in and These respectively represent the audit objectives The number of challenged encoded vectors and data blocks. Furthermore, to generate a challenge index for the encoded vectors, the first smart contract... Call the pseudo-random generator ,index ,in This is a counter variable used in the challenge index generation process to distinguish different challenge indexes generated in the same audit round. .

[0059] This generates a challenge index set for the encoded vectors. for:

[0060] ,in Indicates cloud server The set of indices of the encoded vectors that need to be challenged. Finally, the first smart contract. Record table based on random distribution of coded data Set up audit sub-challenges and generate a blockchain transaction. As the first transaction.

[0061] Response proof generation phase: This phase is handled by each cloud server. Execute them separately, with the specific steps as follows:

[0062] After listening to blockchain transactions After that, cloud server Obtaining the audit sub-challenge And retrieve the corresponding stored data. After that, the cloud server Computation in the Coding Challenge Data Vector Index Set The index of encoded data blocks below , Similarly, a counter variable is used to represent the data block challenge index generation process, distinguishing different data block challenge indexes generated in the same audit round. .

[0063] Indexing each coded data block cloud server Calculate the corresponding weights for it. , Subsequently, the cloud server Calculate combined encoded data blocks , ,in cloud server The stored challenged encoded data blocks are aggregated with corresponding verification tags, which are represented as tuples. ,in:

[0064] In the above formula It is a value with the hash value of the random value corresponding to the challenged coded data block as the base and the corresponding weight as the exponent.

[0065] ;in It is a value with the label corresponding to the challenged coded data block as the base and the corresponding weight as the exponent.

[0066] In the above two equations, It is a large prime number. The multiplicative cyclic group.

[0067] Finally, cloud servers Generate response proof And generate a transaction on the blockchain. As a second transaction, it is stored in the block.

[0068] Response Proof Verification Phase: This phase is handled by verification contracts deployed on the blockchain. The specific steps for execution are as follows:

[0069] Verify the contract within the specified time frame. All second transactions were detected. And obtain the response proof. Subsequently, the contract was verified. Recalculate the data vector index set in the coding challenge The index of encoded data blocks below and their weights:

[0070] ,in Validate the contract Examine outsourced storage data using the following batch verification equation. Completeness:

[0071] Where Q is the verification parameter, If the integrity calculation equation holds, then verify the contract. Output "True" if the audit result is true; otherwise, output "False".

[0072] Validate Contract Create an audit log entry and generate a blockchain transaction. ,in The timestamp representing the output of the audit results.

[0073] The applicant further considered that existing data integrity verification mechanisms suffer from the problem of vague location of damaged data. When an audit fails, existing solutions can typically only locate the problematic cloud server, lacking the ability to accurately pinpoint the specific damaged data blocks, greatly increasing the difficulty and cost of data repair. In other words, when coded data is corrupted, data recovery is required. However, in a distributed storage system, data recovery presupposes the accurate location of the faulty cloud server and the coded data blocks it stores.

[0074] Therefore, based on the foregoing embodiments, this application also provides a method for locating multi-cloud storage data errors in a low-altitude unmanned aerial vehicle network, which mainly includes two stages: a stage for accurately locating damaged data and a stage for restoring damaged data.

[0075] Accurate location of damaged data stage: This stage consists of locating the faulty cloud server and the damaged coded data block.

[0076] Error location on cloud server: When the audit result is "False", the location contract deployed on the blockchain... Triggered to initiate the location of the faulty cloud server. Location contract. Initialize a candidate index set and an empty set The empty set Cloud server used to collect detected errors Indexes, and use management stacks. To manage the subset of indexes to be inspected. For each from the management stack The collection that pops up ,if Then its index is directly used as the cloud server for error handling. Add to collection Middle, otherwise Divided into two approximately equal subsets and Each subset must pass through the aforementioned equation:

[0077]

[0078] Perform integrity verification. If verification fails, locate the contract. Push it into the management stack This process is pending further recursive subdivision and integrity verification. It is repeated iteratively until all erroneous cloud server indexes are included in the set. In the middle. Finally, the positioning contract. Set up an entry for locating the faulty cloud server:

[0079] and provide a transaction to the blockchain. .

[0080] Location of damaged encoded data blocks: When an off-chain cloud server malfunctions... Transaction detected Then, the location of the damaged coded data block is initiated. Error occurred on the cloud server. First, the audit target (i.e., the target object) is retrieved. Data stored in and define a finite set Then, the cloud server encountered an error. Initialize another empty collection This is to record the complete encoded vector of the data. Simultaneously, a counter is set up. To track sets For each set Error in cloud server The verification tags are aggregated and calculated to generate a combined coded data block. And two aggregate signatures, respectively represented as: , ,in The hash value of the random value corresponding to the erroneous encoded data block. The label corresponding to the erroneous encoded data block. The index of the encoded data vector in the faulty cloud server is used. Integrity verification is performed using the following batch verification equation:

[0081] ,in The coded data block in the erroneous cloud server, where H is the seventh hash function, specifically a secure, collision-resistant hash function. .

[0082] If the above verification equation holds, in the set All indexes were added In the middle, and the counter Add one; otherwise, set There are damaged encoding vectors, and they are marked and Unchanged. Until all sets After all verifications were completed, an error occurred on the cloud server. Based on the properties of uniform (K, N) sets on a finite set, the damaged encoding vector is determined and output. .

[0083] Damaged data recovery phase: This phase is handled by the cloud server. Error-prone cloud server The joint execution will follow these steps:

[0084] Error cloud server Send a data recovery request to the help cloud server ,in This indicates the established help agreement. Upon receiving a data recovery request... Afterwards, it helps the cloud server retrieve stored data and sends recovery metadata. Error cloud server .

[0085] When the number of received recovery metadata reaches After the above, the cloud server encountered an error. From the encoding matrix Extract the corresponding submatrix The encrypted data file is recovered through the following calculations. : Ultimately, the cloud server malfunctioned. calculate To replace the erroneous encoding vector .

[0086] Based on the foregoing embodiments, for further explanation, the derivation process of some of the aforementioned formulas is provided.

[0087] The equation for checking correctness in the above embodiments is:

[0088]

[0089] The derivation process for using cloud server CS to verify the correctness of outsourced data sent by drones is as follows:

[0090]

[0091] The equation for calculating integrity in the foregoing embodiments is as follows:

[0092]

[0093] Used for the second smart contract The derivation process for verifying the integrity of outsourced data in cloud server CS storage is as follows:

[0094] in .

[0095] Therefore, this application implements an identity-based homomorphic hash signature algorithm to aggregate verification tags and data returned by different servers in a multi-cloud environment under a unified structure, making the verification equation linearly composable. This structure avoids the high communication and computational overhead caused by block-by-block signature verification in traditional solutions, and realizes a lightweight auditing process of off-chain aggregation and on-chain consistency verification. This solution significantly reduces computational complexity in multi-node, multi-block auditing scenarios, making it more suitable for the continuous outsourced storage and rapid verification needs of TB-level data from low-altitude UAVs.

[0096] In summary, this application implements a three-layer progressive error location mechanism to accurately locate faulty servers and corrupted data blocks. To address the difficulty of accurately locating damaged blocks in existing multi-cloud systems, a three-layer location mechanism of "server detection—subset determination—block-level verification" is constructed. The verification contract can automatically recursively divide the index space when auditing fails and perform batch verification on the partitions, effectively isolating the faulty server. The faulty node gradually shrinks the search space based on a uniform (K,N) set and combined signatures to accurately locate the corrupted encoded data block. This structured location method significantly reduces recovery costs and substantially improves system diagnostic efficiency.

[0097] This invention employs Reed-Solomon erasure coding to encode raw UAV data, ensuring all shards are evenly distributed across multiple cloud servers, significantly reducing the overhead of traditional replica-based redundant storage. In the event of cloud server failure or data corruption, the method helps the cloud server recover lost raw data blocks and their homomorphic tags based on the encoding matrix, achieving data regeneration without relying on centralized components. This method offers advantages such as high fault tolerance, high stability, and low-bandwidth recovery, meeting the long-term storage and robustness requirements of low-altitude UAV data.

[0098] To avoid the risk of information leakage due to the exposure of drone identities, this invention employs an identity-based key exchange and symmetric encryption mechanism to achieve conditional anonymity for users. In normal business operations, cloud servers and auditors cannot obtain the true identity of the drone, ensuring mission privacy and security. In the event of abuse or malicious behavior, the integrated monitoring service platform can trace the identity of the corresponding user, achieving a balance between anonymity and accountability.

[0099] Compared to traditional multi-cloud auditing, which relies on cloud agents or third-party auditors and is prone to trust bottlenecks and forgery risks, this application utilizes blockchain smart contracts to solidify the entire audit process on the chain, making the audit process open, transparent, automated, and tamper-proof. All results are traceable and require no local computing resources from the user, demonstrating outstanding practical feasibility and credibility in resource-constrained scenarios such as drones.

[0100] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle (UAV) network, characterized in that, In low-altitude drone networks, after registering through a comprehensive monitoring service platform to generate user private and public keys, drones send stored data to multiple cloud servers for multi-cloud storage. This includes the drone sending the data files to be uploaded. The data is divided into multiple data blocks, each data block is further divided into multiple data word blocks, and each data word block is encrypted using a symmetric encryption algorithm to obtain ciphertext sub-blocks. The subscript ij represents the j-th sub-block of the i-th data block; then, the UAV encodes the ciphertext data block using erasure coding to generate multiple encoding vectors. For each encoded data block in the encoding vector, the UAV calculates its corresponding verification tag; the UAV obtains the number of available cloud servers, calculates the amount of encoded data stored on each cloud server, and then constructs an adapted random distribution record table of encoded data to record the encoding vectors and their verification tags corresponding to each cloud server, so as to allocate and transmit the storage data that the UAV needs to store to one of the multiple cloud servers for storage; Then, integrity verification is performed on the low-altitude drone data stored on multiple cloud servers, including: a first smart contract deployed on the blockchain executes an audit challenge generation phase, wherein the first smart contract generates audit challenges for multiple cloud servers and publishes the audit challenges as the first transaction to the blockchain; the first smart contract generating audit challenges for multiple cloud servers includes: the first smart contract obtaining the header hash value of the latest block based on the current state of the blockchain and the block height. And extract the previous historical header hash values ​​in reverse order, the total number is Based on the latest block header hash and historical header hash, calculate the audit challenge seed; based on the audit challenge seed, use a pseudo-random function to generate the audit challenge containing the challenge index; Multiple cloud servers storing target data encoded data blocks execute the response proof generation phase respectively. Each cloud server listens to the blockchain, and after obtaining the audit challenge in the first transaction, it generates a response proof based on its own stored encoded data blocks and publishes its respective response proof as a second transaction to the blockchain. The verification contract deployed on the blockchain performs a response proof verification phase. The verification contract listens for and obtains multiple response proofs from the second transaction published by all cloud servers within a preset time. According to a predetermined verification algorithm, the aggregated multiple response proofs are batch verified to determine the integrity of the target data. The audit log containing the verification results is then published to the blockchain as a third transaction.

2. The method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle network according to claim 1, characterized in that, The calculation of the audit challenge seed based on the latest block header hash and historical header hash includes: First Smart Contract Calculate the audit challenge seed key 1: And Audit Challenge Seed Key 2: ； In the above two equations, , , All are historical header hash values, in the index. This represents the total number of previous historical header hashes. This is the fifth hash function; First Smart Contract Output Auditing Challenge ,in and These respectively represent the audit objectives The number of challenged encoding vectors and data blocks in the middle. To remain anonymous The filename for the data file uploaded by the drone; The process of generating the audit challenge containing the challenge index using a pseudo-random function based on the audit challenge seed includes: The first smart contract calls the pseudo-random number generator. ,index ,in To challenge the counting variables in the index generation process; , For auditing the pseudo-random generator corresponding to seed key 1; Generate challenge index set for: ,in Indicates cloud server The set of indices of the encoded vectors that need to be challenged. For the number of cloud servers, For the first One cloud server.

3. The method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle network according to claim 2, characterized in that, The step of generating a response proof based on its own stored encoded data blocks includes: Based on the challenge index in the aforementioned audit challenge, determine the index of the encoded data block in this server that needs to be challenged. , , To challenge the counting variables in the index generation process, and And calculate its corresponding weight. ; in, It is a pseudo-random generator; Calculate the challenged coded data block : in, The challenged encoded data blocks stored on the cloud server. It is a model Remaining rings, For index set; The aggregated and challenged encoded data blocks correspond to the verification tags, and the verification tags are... ,in: ； ； In the above formula, The value is based on the hash value of the random value corresponding to the challenged coded data block, and exponentiated by the corresponding weight. The value is calculated with the label corresponding to the challenged coded data block as the base and the corresponding weight as the exponent. To verify the bilinear group to which the label belongs; The response proof is constructed by aggregating the encoded data blocks and the aggregated verification tags.

4. The method for verifying the integrity of multi-cloud storage data in a low-altitude unmanned aerial vehicle network according to claim 3, characterized in that, The predetermined verification algorithm is as follows: Where Q is the verification parameter, , The master public key, It is the user's public key. It is a bilinear pair. It is a generator of the multiplicative cyclic group. This is the fourth hash function, where U is the second user's public key, and the ? in the formula indicates whether the equation is true; If the equation is true, the verification contract outputs a true result; otherwise, it outputs a false result.

5. A method for locating errors in multi-cloud storage data in a low-altitude unmanned aerial vehicle (UAV) network, implemented based on the data integrity verification method for multi-cloud storage data in a low-altitude UAV network as described in any one of claims 1-4, characterized in that... If the verification result determined in the response verification phase is a failure, the following steps are executed: For low-altitude drone data stored on multiple cloud servers in a low-altitude drone network, the faulty cloud servers are located through a positioning contract deployed on the blockchain. The set of multiple cloud servers is subjected to integrity verification based on the predetermined verification algorithm until all faulty cloud servers with incomplete data storage are located. A faulty cloud server positioning entry is set through the positioning contract, and a transaction is provided to the blockchain. The faulty cloud server that has been located locates the damaged coded data block. After listening to the blockchain to obtain the transactions it provides, the faulty cloud server performs fragment verification on its local storage to determine the specific damaged coded data block.

6. The method for locating multi-cloud storage data errors in a low-altitude unmanned aerial vehicle network according to claim 5, characterized in that, The collection of multiple cloud servers undergoes integrity verification based on the predetermined verification algorithm until all faulty cloud servers with incomplete stored data are identified, including: Initialize a candidate index set containing all cloud server indexes and an empty set for collecting indexes of erroneous servers; Push the candidate index set onto the management stack; Pop a subset of indexes to be checked from the management stack. If the size of the subset is 1, add its index to the set of servers that have failed. If the size is greater than 1, divide it into two subsets. For each subset after partitioning, the predetermined verification algorithm performs integrity verification; The subset that failed verification is pushed back into the management stack, and the above popping, partitioning and verification process is repeated until the stack is empty, thus obtaining the final set of erroneous cloud servers.

7. The method for locating multi-cloud storage data errors in a low-altitude unmanned aerial vehicle network according to claim 5, characterized in that, The step of performing fragmented verification on the local storage of the faulty cloud server to determine the specific damaged coded data blocks includes: The cloud server that encountered the error first retrieved the audit target. Data stored in ,in To remain anonymous Define the filenames for the data files uploaded by the drone, and define a finite set. ;in, For encoding vectors, To verify the label; Next, the cloud server initializes an empty set. This is to record data and store complete encoded vectors; and a counter is set up to track the set. ; The cloud server that encountered the error performs aggregate calculations on its verification tags to generate a combined coded data block: And two aggregate signatures, respectively represented as: , ; in, The index of the encoded data vector in the cloud server where the error occurred; W is the first user's public key. This is the fourth hash function, where i represents the i-th data block, j represents the j-th sub-block of the i-th data block, and y is the total number of sub-blocks. The hash value of the random value corresponding to the erroneous encoded data block. The labels corresponding to the erroneous encoded data blocks are used, and integrity verification is performed using the following batch verification equation: In the formula, The 'e' in the encoded data block of the cloud server is a bilinear pair. It is a generator of the multiplicative cyclic group. This is the third hash function, U is the second user's public key, and H is the seventh hash function. It is the master public key; If the above verification equation holds, in the set All indexes are added to the empty set. If the condition is met, the counter is incremented; otherwise, the set is closed. There are damaged encoded vectors, which are marked and the counter remains unchanged; until each set After verification, the damaged encoding vector is determined and output. .

8. The method for locating multi-cloud storage data errors in a low-altitude unmanned aerial vehicle network according to claim 7, characterized in that, It also includes a damaged data recovery phase, which is performed after the damaged coded data block location phase, and includes: The faulty cloud server initiates a data recovery request to other normal help cloud servers; After receiving recovery metadata returned by the help cloud server that is no less than the minimum threshold required for data recovery, the faulty cloud server reconstructs the original data or the correct encoded data block based on the recovery metadata and the pre-stored encoding matrix to replace the local damaged encoded data block.

Images