A cross-chain data privacy supervision method based on random feature optimization selection

Abstract

The present invention discloses a cross-chain data privacy supervision method based on random feature optimization selection, which is characterized by comprising the following steps: S1: a cross-chain governance domain links a supervision chain and a local chain, and performs identity registration through a chain registration protocol based on a HE-ECDH key negotiation algorithm when the chain is initialized; a one-time session seq value is determined through an off-chain trusted environment before chain registration; S2: a design is used to record the transaction process of both parties through an audit chain, and time control is performed on each stage of a cross-chain data privacy protection protocol based on random feature optimization selection. If one party uploads false data or fails to upload data within a timeout period, the audit chain records its violation and terminates the protocol. The technical problem to be solved by the present invention is to provide a cross-chain data privacy supervision method based on random feature optimization selection, which ensures the privacy security of cross-chain supervision data content while protecting the selected privacy from being leaked.

Description

Technical Field

[0001] The present invention relates to the technical field of data privacy supervision, and in particular to a cross-chain data privacy supervision method based on random feature optimization selection. Background Art

[0002] With the popularization and application of blockchain technology, many enterprise asset management systems have built their own blockchain network platforms, which not only promoted technological innovation, but also brought new challenges - how to achieve interoperability between blockchain platforms with different structures and consensus mechanisms. Although different blockchains can interact with each other off-chain, there is also the problem of data leakage. It is difficult to access data between chains at the same time, resulting in the inability to communicate or exchange value between blockchains, forming a "data island". The application potential of cross-chain technology is huge. It can promote asset liquidity, improve transaction efficiency, reduce costs, and enhance the transparency and security of the financial market. Existing researchers have proposed to design two parallel chain modules based on the cross-chain transaction scenario of digital assets to achieve fast transactions and secure payments. Existing cross-chain technology solutions still have shortcomings in privacy protection, scope of application, efficiency, scalability and versatility.

[0003] The security and privacy protection issues of cross-chain operations are becoming increasingly prominent, especially in cross-chain access control. An effective cross-chain access control mechanism is essential to ensure the secure transmission of data between different blockchains, prevent unauthorized access, and maintain the stability of the entire blockchain ecosystem. In order to meet these challenges, researchers have conducted in-depth research on blockchain cross-chain technology, for example, by using distributed private key control mechanisms, hash locking technology, side chains / relay chains, etc. to enhance the security and reliability of cross-chain transactions. A cross-chain identity authentication solution that includes decentralized identity (DID), smart contracts, and identity public key (IPK) security algorithms is proposed. These solutions not only improve the security of cross-chain systems, but also provide new ideas and methods for achieving efficient access control.

[0004] In terms of cross-chain data privacy protection, although blockchain technology is known for its openness and immutability, with the development of cross-chain technology, the issue of user privacy protection has become particularly prominent. Cross-chain interaction involves data transmission and communication between different blockchain platforms, which may have different privacy protection measures and security standards.

[0005] Existing cross-chain technologies are often designed to achieve asset liquidity and transaction interoperability, but pay insufficient attention to the privacy protection of data during transmission. This neglect may lead to the interception or leakage of sensitive information during the transmission between different blockchain systems, especially when it comes to personal identity information, transaction history, and asset details. The risk of privacy leakage is particularly serious. In the data supervision interaction scenario, when the regulator conducts cross-chain data supervision in various places, it often causes data selection privacy leakage, causing the regulated party to infer the regulator's intentions based on the data selection. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a cross-chain data privacy supervision method based on random feature optimization selection, so as to realize the optimized selection of arbitrary data by the receiver, so that the sender cannot infer the receiver's intention or behavior through the selection mode, thereby ensuring the privacy security of cross-chain supervision data content while protecting the selection privacy from being leaked.

[0007] The present invention adopts the following technical solutions to achieve the invention objectives:

[0008] A cross-chain data privacy supervision method based on random feature optimization selection, characterized by comprising the following steps:

[0009] S1: The cross-chain governance domain links the regulatory chain and the local chain, and performs identity registration through the chain registration protocol based on the HE-ECDH key agreement algorithm during chain initialization;

[0010] Before chain registration, the one-time session seq value is determined through the off-chain trusted environment;

[0011] S2: Design an audit chain to record the transaction process of both parties, and perform time control on each stage of the cross-chain data privacy protection protocol based on random feature optimization selection. If one party uploads false data or fails to upload within a time limit, the audit chain will record its violation and terminate the protocol.

[0012] S3: To optimize the data receiving process in the cross-chain random feature optimization selection scenario, an arbitrary target optimization selection based on the on-chain Bloom filter is adopted. By selecting m in small batches multiple times, the transmission efficiency and time overhead are balanced to ensure that the receiver can quickly determine the repeatability of the selected key, significantly improving the flexibility and efficiency of data reception.

[0013] S31: The receiver expects to receive m blocks of data and initialize the corresponding Bloom filter on the blockchain;

[0014] S32: When m≥10, the local chain selects TPKi in groups of 10 keys, and calculates and verifies whether it exists in the Bloom filter on the chain. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed. The expected number of data blocks of the receiver becomes m=m-10, and S32 is repeated; when m<10, calculate and verify whether the remaining m blocks exist on the Bloom filter. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed, and it is executed to S33;

[0015] S33: The receiver successfully receives m secret data and performs hash verification on the data.

[0016] As a further limitation of the technical solution, the specific steps of S1 are:

[0017] S11: Local chain generates ECC key pair<LPK,LSK> , hash the concatenated string {CID, CName, LPK, RCAddress, Type, seq, ...} to get H0, and then send the transaction data packet Tx1 = [H0, LPK, Sig_LSK (H0)] to the cross-chain governance domain;

[0018] S12: The cross-chain governance domain uses LPK to verify Sig(H0) and generate a one-time ECC key pair<GPK,GSK> , calculate H1=hash(GPK, seq+1), sign Sig_GSK(H1), and call back data packet Tx2=[H1, Sig_GSK(H1), GPK] to the local chain;

[0019] S13: The local chain uses GPK to verify the signature. After successful verification, GPK and seq+1 are linked to calculate the hash to obtain H1', ​​and verify whether it is consistent with H1 in Tx2. After successful verification, GPK and LSK are calculated according to the ECDH negotiation algorithm to obtain the negotiated key, and KFD is used to unidirectionally calculate the symmetric key K, calculate H2=H{CID, CName, LPK, RCAddress, Type, seq+2}, use the LSK private key to sign H2 to obtain Sig_LSK(H2), use K' to encrypt parameters and signatures, and send Tx3=[Sig_LSK(H2), E(H2), E(CID, CName, LPK, RCAddress, Type)] to forward to GD;

[0020] S14: The cross-chain governance domain also uses the ECDH negotiation algorithm to calculate LPK and GSK to obtain K to decrypt Tx3 and use LPK to verify H2. Then, data={CID, CName, LPK, RCAddress, Type} is linked to seq+2 to calculate the hash to obtain H2', and verify whether H2' matches H2. After successful verification, data is linked to seq and the hash is calculated to obtain H0', and verify whether H0' matches H0. After successful verification, the GD contract executes the cross-chain session registration session={UID, ChainID, ChainName, PK, H(K), RCAddress, Type, TimeStamp}.

[0021] As a further limitation of the technical solution, the specific steps of S2 are:

[0022] T0: system initialization;

[0023] Audit chain initialization: execute chain registration operation, and upload cross-chain authorization public key to the supervision chain And RCID, local chain upload cross-chain authorization public key And LCID to the audit chain smart contract to register the authorized access certificate;

[0024] Initialization of the custody chain: registering the cross-chain public key with the audit chain and chain ID, and generates m one-time 256-bit Symmetric Key ;

[0025] Local chain initialization: First upload n blocks of transmission data to IPFS to obtain the CID address access identifier ,use Calculate its hash value to get ; The local chain obtains the public key of the supervision chain in the audit chain through the cross-chain contract And encrypt get ; Then generate n 256-bit Symmetric Key right Encrypted And record it as ; Finally adopted Elliptic curve parameter generation n Asymmetric key pair ;

[0026] T1: Local chain cross-chain authorized account will , and And the number of blocks n using the private key implement After elliptic curve signing, we get , , , and then upload the signature and ciphertext data to the audit chain through the cross-chain smart contract;

[0027] T2: The custody chain accesses the audit chain through a cross-chain smart contract , select m from them encryption get , with its cross-chain account private key implement Elliptic curve signature , upload the signature and ciphertext data to the audit chain, and the audit chain uses Verify whether it is a fake message. If it fails the signature verification, it will not be uploaded.

[0028] T3: The local chain obtains through the cross-chain smart contract , and use the n generated during initialization Decrypt each Get the m*n key matrix ; Then the local chain uses the m*n key matrix to encrypt the n symmetric keys generated during its initialization , get the m*n encryption matrix Right now , where the chain uses After signing, the encrypted matrix will be uploaded to the audit chain;

[0029] T4: Chain of custody obtained through cross-chain smart contracts , using the m symmetric keys generated during initialization Decrypt the correct ciphertext data of each row in the matrix and get m And Decrypted Then the custody chain uses the cross-chain account private key RSK to decrypt again and obtain m Address, by Hash operation verification compares the cross-chain audit chain uploaded by the local chain , as the basis for determining whether the data transmission process has been tampered with.

[0030] Compared with the prior art, the advantages and positive effects of the present invention are:

[0031] The present invention proposes a cross-chain key agreement protocol based on HE-ECDH. An extensible cross-chain governance domain is designed to realize secure and unified identity registration between heterogeneous chains, solving the problems of replay attack and man-in-the-middle attack in the traditional key agreement process. At the same time, the extensible cross-chain governance domain enables the new blockchain to be seamlessly integrated into the governance framework, with good scalability and flexibility. The present invention proposes a cross-chain data privacy protection protocol based on random feature optimization selection. The m out of n cross-chain random feature optimization selection protocol (C-OT) is designed to realize arbitrary target data optimization selection, that is, cross-chain data privacy supervision, so that the data sender cannot infer the intention or behavior of the receiver through the selection mode, thereby ensuring the privacy of cross-chain content transmission. At the same time, an audit chain is designed and a set of audit smart contracts are developed. The contract automatically executes various operations specified in the agreement to ensure the compliance of regulatory activities and the non-repudiation of data exchange, and records the party that fails to complete the corresponding operation within the specified time. The arbitrary data optimization selection of the receiver is realized, so that the sender cannot infer the intention or behavior of the receiver through the selection mode, ensuring the privacy of cross-chain regulatory data content while protecting the selection privacy from being leaked. Finally, when the regulator needs to obtain a large number of random feature optimization selection data blocks, it can use Bloom filter (BF) to quickly calculate and verify whether the selected receiving key is selected, thereby reducing the problem of exponential growth of the random feature optimization selection protocol time caused by a single random feature optimization selection m being too large. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a system framework diagram of the present invention.

[0033] Figure 2 Schematic diagram of the heterogeneous chain registration protocol of the present invention.

[0034] Figure 3 Schematic diagram of the cross-chain random feature optimization selection protocol of the present invention.

[0035] Figure 4 A schematic diagram of the total time consumption for random feature optimization selection of the present invention.

[0036] Figure 5 It is a schematic diagram of the Bloom filter verification process of the present invention. DETAILED DESCRIPTION

[0037] A specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings, but it should be understood that the protection scope of the present invention is not limited by the specific implementation.

[0038] The present invention comprises the following steps:

[0039] S1: The cross-chain governance domain links the regulatory chain and the local chain, and performs identity registration through the chain registration protocol based on the HE-ECDH key agreement algorithm during chain initialization.

[0040] Before chain registration, the one-time session seq value (positive integer) is determined through the off-chain trusted environment; the protocol flow is as follows Figure 2 As shown, the chain registration message field description is shown in Table 1.

[0041] Design a cross-chain data privacy supervision system framework such as Figure 1 , through the scalable cross-chain governance domain to link various heterogeneous chains, complete cross-chain identity registration and verification, realize cross-chain random feature optimization selection and reporting, that is, the supervision chain optimizes the selection of arbitrary target data of the local chain, and designs an audit chain to ensure the compliance and traceability of the entire process.

[0042] Local chain: an enterprise blockchain with autonomous data, acting as the data reporter in the cross-chain regulatory system;

[0043] Chain of Supervision: As the data supervisor in the cross-chain system, it supervises the local chain data and performs corresponding operations according to the business;

[0044] Cross-chain governance domain: As a relay chain, it provides chain registration and identity authentication services for the entire cross-chain regulatory system;

[0045] Audit Chain: responsible for auditing the cross-chain random feature optimization selection protocol process to ensure the compliance, security and transparency of data exchange;

[0046] IPFS: InterPlanetary File System is a peer-to-peer distributed file system for storing and sharing data, which allows users to store and access files globally without relying on a specific centralized server. Blockchains can share and transfer files through IPFS, securely transferring large amounts of data without storing the entire file on the chain.

[0047] Table 1 Chain registration message fields

[0048]

[0049] The specific steps of S1 are:

[0050] S11: Local chain (LC) generates ECC key pair<LPK,LSK> , hash the connection string {CID, CName, LPK, RCAddress, Type, seq, ...} to get H0, and then send the transaction data packet Tx1 = [H0, LPK, Sig_LSK (H0)] to the cross-chain governance domain;

[0051] S12: The cross-chain governance domain (GD) uses LPK to verify Sig(H0) and generate a one-time ECC key pair<GPK,GSK> , calculate H1=hash(GPK, seq+1), sign Sig_GSK(H1), (cross-link routing) callback data packet Tx2=[H1, Sig_GSK(H1), GPK] to the local link;

[0052] S13: The local chain uses GPK to verify the signature. After successful verification, GPK and seq+1 are linked to calculate the hash to obtain H1', ​​and verify whether it is consistent with H1 in Tx2. After successful verification, GPK and LSK are calculated according to the ECDH negotiation algorithm to obtain the negotiated key, and KFD is used to unidirectionally calculate the symmetric key K, calculate H2=H{CID, CName, LPK, RCAddress, Type, seq+2}, use the LSK private key to sign H2 to obtain Sig_LSK(H2), use K' to encrypt parameters and signatures, and send Tx3=[Sig_LSK(H2), E(H2), E(CID, CName, LPK, RCAddress, Type)] to forward to GD;

[0053] S14: The cross-chain governance domain also uses the ECDH negotiation algorithm to calculate LPK and GSK to obtain K to decrypt Tx3 and use LPK to verify H2. Then, data={CID, CName, LPK, RCAddress, Type} is linked to seq+2 to calculate the hash to obtain H2', and verify whether H2' matches H2. After successful verification, data is linked to seq and the hash is calculated to obtain H0', and verify whether H0' matches H0. After successful verification, the GD contract executes the cross-chain session registration session={UID, ChainID, ChainName, PK, H(K), RCAddress, Type, TimeStamp}.

[0054] S2: Design an audit chain to record the transaction process of both parties, and perform time control on each stage of the cross-chain data privacy protection protocol based on random feature optimization selection. If one party uploads false data or fails to upload within a time limit, the audit chain will record its violation and terminate the protocol.

[0055] The specific steps of S2 are:

[0056] T0: system initialization;

[0057] Audit chain initialization: execute chain registration operation, and upload cross-chain authorization public key to the supervision chain And RCID, local chain upload cross-chain authorization public key And LCID to the audit chain smart contract to register the authorized access certificate;

[0058] Initialization of the custody chain: registering the cross-chain public key with the audit chain and chain ID, and generates m one-time 256-bit Symmetric Key ;

[0059] Local chain initialization: First upload n blocks of transmission data to IPFS to obtain the CID address access identifier ,use Calculate its hash value to get ; The local chain obtains the public key of the supervision chain in the audit chain through the cross-chain contract And encrypt get ; Then generate n 256-bit Symmetric Key right Encrypted And record it as ; Finally adopted Elliptic curve parameter generation n Asymmetric key pair ;

[0060] T1: Local chain cross-chain authorized account will , and And the number of blocks n using the private key implement After elliptic curve signing, we get , , , and then upload the signature and ciphertext data to the audit chain through the cross-chain smart contract; the audit chain smart contract uses the cross-chain authorized account public key uploaded when the chain is registered The data signature is verified. If the verification fails, the data will not be uploaded any further. This step needs to be executed within the specified T1 time after the system is initialized. If the audit chain does not receive the legal data and signature uploaded by the local chain within the specified time, the system will determine that the local chain has violated the rules, record the violation in the audit chain, and terminate the protocol.

[0061] T2: The custody chain accesses the audit chain through a cross-chain smart contract , select m from them encryption get , with its cross-chain account private key implement Elliptic curve signature , upload the signature and ciphertext data to the audit chain, and the audit chain uses Verify whether it is a forged message. If the signature verification fails, it will not be uploaded. This step needs to be completed within the T2 time period after the system is initialized. If the audit chain does not receive the legal data and signature that the supervision chain should upload within the specified time, the system will determine that the supervision chain is in violation, record the violation in the audit chain, and terminate the protocol.

[0062] T3: The local chain obtains through the cross-chain smart contract , and use the n generated during initialization Decrypt each Get the m*n key matrix (Only one decryption key in each row of the matrix is ​​correct, and each correct decryption result corresponds to the m symmetric keys generated by the previous stage of the custody chain , since the regulator has selected m To encrypt its own symmetric key, the supervisor can know which key in each row of the matrix is ​​correctly decrypted, but the local chain cannot know it); then the local chain uses the m*n key matrix to encrypt the n symmetric keys generated during its initialization , get the m*n encryption matrix Right now (Similarly, there is only one key in each row. is correct encrypted, i.e. the chain of custody knows which of each row is the correct encrypted result), where the chain uses After signing, the encrypted matrix will be uploaded to the audit chain; this step needs to be completed within the T3 time period after the system is initialized. If the audit chain does not receive the legal data and signature that the local chain should upload within the specified time, the system will determine that the supervision chain is in violation, record the violation in the audit chain, and terminate the protocol.

[0063] T4: Chain of custody obtained through cross-chain smart contracts , using the m symmetric keys generated during initialization Decrypt the correct ciphertext data of each row in the matrix and get m And Decrypted Then the custody chain uses the cross-chain account private key RSK to decrypt again and obtain m Address, available through Hash operation verification compares the cross-chain audit chain uploaded by the local chain , as the basis for determining whether the data transmission process has been tampered with. If there is an objection, the supervision chain needs to submit an objection arbitration to the audit chain within the T4 time period after the system is initialized. If there is no objection, the agreement is executed.

[0064] Table 2 Regulatory data packet format

[0065]

[0066] S3: To optimize the data receiving process in the cross-chain random feature optimization selection scenario, an arbitrary target optimization selection based on the on-chain Bloom filter is adopted. By selecting m in small batches multiple times, the transmission efficiency and time overhead are balanced to ensure that the receiver can quickly determine the repeatability of the selected key, significantly improving the flexibility and efficiency of data reception.

[0067] Introduction to Bloom Filter (BF)

[0068] Bloom Filter is a probabilistic data structure with high space efficiency, which is mainly used to determine whether an element is in a set. It is known for its small storage space and extremely fast query speed, but it has a certain false positive rate, that is, it may mistakenly judge an element that is not in the set as being in the set.

[0069] Bloom filter uses multiple hash functions and a bit array to implement its function. The specific steps are as follows:

[0070] Initialization: Create a bit array of length m with all bits initialized to 0. Select k independent hash functions, each of which maps the input element to a value in the range An integer.

[0071] Adding an element: When adding an element to a Bloom filter, the element is hashed using k hash functions to obtain the index of a k-bit array. Then, the bits corresponding to these indexes are set to 1.

[0072] Query verification element: When checking whether an element is in the Bloom filter, the element is hashed through k hash functions to obtain k indexes. Then, check whether the bits corresponding to these indexes are all 1. If all are 1, the element may be in the set; if any one is 0, the element is definitely not in the set.

[0073] False Positive Rate: The probability that a Bloom filter incorrectly determines that an element is in a set when it is not actually in the set.

[0074] The misjudgment rate of the Bloom filter can be calculated by the following formula:

[0075] Where: P is the false positive rate;

[0076] k is the number of hash functions;

[0077] n is the number of elements inserted;

[0078] m is the size of the bit array;

[0079] Bloom filter (BF) application

[0080] From the statistical experiment chart above, we know that when random feature optimization selection increases with the increase of random feature optimization selection m, the time consumed by the protocol increases exponentially. In order to optimize the data reception process in the cross-chain random feature optimization selection scenario, we proposed an innovative mechanism based on the on-chain Bloom filter. This mechanism balances transmission efficiency and time overhead by selecting m in small batches multiple times, ensuring that the receiver can quickly determine the repeatability of the selected key, and significantly improving the flexibility and efficiency of data reception.

[0081] Therefore, the present application designs a Bloom filter for each batch of random feature optimization selection data in the custody chain, records the calculation of the key TPK selected by the random feature optimization selection data and writes it into the BF.

[0082] The specific calculation is to calculate TPK through k hash functions to obtain k index positions, check whether the corresponding positions are all 1, if all are 1, it is determined that the data has been accessed before, and the current key selection is not valid. At the same time, due to the false alarm rate of the Bloom filter, it may falsely report that a non-existent element exists. However, the Bloom filter will not miss an actually existing element, and for this scenario, it will not miss any previously accessed data, thereby ensuring the timeliness of verifying the access data.

[0083] We design a batch of random feature optimization selection data with 100 blocks of data and control the false alarm rate at 0.01%. We need to calculate the bit array size L and the number of hash functions k respectively.

[0084] There is a bit array with the following calculation formula bits, number of hash functions The calculation result is rounded up to L=1914 bits and k=13.

[0085] S31: The receiver expects to receive m blocks of data and initialize the corresponding Bloom filter on the blockchain;

[0086] S32: When m≥10, the local chain selects TPKi in groups of 10 keys, and calculates and verifies whether it exists in the Bloom filter on the chain. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed. The expected number of data blocks of the receiver becomes m=m-10, and S32 is repeated; when m<10, calculate and verify whether the remaining m blocks exist on the Bloom filter. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed, and it is executed to S33;

[0087] S33: The receiver successfully receives m secret data and performs hash verification on the data.

[0088] The above disclosure is only a specific embodiment of the present invention, but the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A cross-chain data privacy supervision method based on random feature optimization selection, characterized in that: The following steps are involved: S1: The cross-chain governance domain links the regulatory chain and the local chain, and performs identity registration through the chain registration protocol based on the HE-ECDH key agreement algorithm during chain initialization; Before chain registration, the one-time session seq value is determined through the off-chain trusted environment; S2: Design an audit chain to record the transaction process of both parties, and perform time control on each stage of the cross-chain data privacy protection protocol based on random feature optimization selection. If one party uploads false data or fails to upload within a time limit, the audit chain will record its violation and terminate the protocol. The specific steps of S2 are: T0: system initialization; Audit chain initialization: execute chain registration operation, and upload cross-chain authorization public key to the supervision chain And RCID, local chain upload cross-chain authorization public key And LCID to the audit chain smart contract to register the authorized access certificate; Initialization of the custody chain: registering the cross-chain public key with the audit chain and chain ID, and generates m one-time 256-bit Symmetric Key ; Local chain initialization: First upload n blocks of transmission data to IPFS to obtain the CID address access identifier ,use Calculate its hash value to get ; The local chain obtains the public key of the supervision chain in the audit chain through the cross-chain contract And encrypt get ; Then generate n 256-bit Symmetric Key right Encrypted And record it as ; Finally adopted Elliptic curve parameter generation n Asymmetric key pair ; T1: Local chain cross-chain authorized account will , and And the number of blocks n using the private key implement After elliptic curve signing, we get , , , and then upload the signature and ciphertext data to the audit chain through the cross-chain smart contract; T2: The custody chain accesses the audit chain through a cross-chain smart contract , select m from them encryption get , with its cross-chain account private key implement Elliptic curve signature , upload the signature and ciphertext data to the audit chain, and the audit chain uses Verify whether it is a fake message. If it fails the signature verification, it will not be uploaded. T3: The local chain obtains through the cross-chain smart contract , and use the n generated during initialization Decrypt each Get the m*n key matrix ; Then the local chain uses the m*n key matrix to encrypt the n symmetric keys generated during its initialization , get the m*n encryption matrix Right now , where the chain uses After signing, the encrypted matrix will be uploaded to the audit chain; T4: Chain of custody obtained through cross-chain smart contracts , using the m symmetric keys generated during initialization Decrypt the correct ciphertext data of each row in the matrix and get m And Decrypted Then the custody chain uses the cross-chain account private key RSK to decrypt again and obtain m Address, by Hash operation verification compares the cross-chain audit chain uploaded by the local chain , as a basis for determining whether the data transmission process has been tampered with; S3: To optimize the data receiving process in the cross-chain random feature optimization selection scenario, an arbitrary target optimization selection based on the on-chain Bloom filter is adopted. By selecting m in small batches multiple times, the transmission efficiency and time overhead are balanced to ensure that the receiver can quickly determine the repeatability of the selected key, significantly improving the flexibility and efficiency of data reception. S31: The receiver expects to receive m blocks of data and initialize the corresponding Bloom filter on the blockchain; S32: When m≥10, the local chain selects TPKi in groups of 10 keys, and calculates and verifies whether it exists in the Bloom filter on the chain. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed. The expected number of data blocks of the receiver becomes m=m-10, and S32 is repeated; when m<10, calculate and verify whether the remaining m blocks exist on the Bloom filter. If it is verified, the key is reselected, and then the cross-chain random feature optimization selection protocol is executed, and it is executed to S33; S33: The receiver successfully receives m secret data and performs hash verification on the data.

2. The cross-chain data privacy supervision method based on random feature optimization selection according to claim 1 is characterized in that: The specific steps of S1 are: S11: Local chain generates ECC key pair<LPK,LSK> , hash the concatenated string {CID, CName, LPK, RCAddress, Type, seq, ...} to get H0, and then send the transaction data packet Tx1 = [H0, LPK, Sig_LSK (H0)] to the cross-chain governance domain; S12: The cross-chain governance domain uses LPK to verify Sig(H0) and generate a one-time ECC key pair<GPK,GSK> , calculate H1=hash(GPK, seq+1), sign Sig_GSK(H1), and call back data packet Tx2=[H1, Sig_GSK(H1), GPK] to the local chain; S13: The local chain uses GPK to verify the signature. After successful verification, GPK and seq+1 are linked to calculate the hash to obtain H1', ​​and verify whether it is consistent with H1 in Tx2. After successful verification, GPK and LSK are calculated according to the ECDH negotiation algorithm to obtain the negotiated key, and KFD is used to unidirectionally calculate the symmetric key K, calculate H2=H{CID, CName, LPK, RCAddress, Type, seq+2}, use the LSK private key to sign H2 to obtain Sig_LSK(H2), use K' to encrypt parameters and signatures, and send Tx3=[Sig_LSK(H2), E(H2), E(CID, CName, LPK, RCAddress, Type)] to forward to GD; S14: The cross-chain governance domain also uses the ECDH negotiation algorithm to calculate LPK and GSK to obtain K to decrypt Tx3 and use LPK to verify H2. Then, data={CID, CName, LPK, RCAddress, Type} is linked to seq+2 to calculate the hash to obtain H2', and verify whether H2' matches H2. After successful verification, data is linked to seq and the hash is calculated to obtain H0', and verify whether H0' matches H0. After successful verification, the GD contract executes the cross-chain session registration session={UID, ChainID, ChainName, PK, H(K), RCAddress, Type, TimeStamp}.

Images