Dynamic adaptive transaction method for interference data

Abstract

The invention belongs to the field of data transaction, and provides an interference data-oriented dynamic adaptive transaction method, which is applied to a corresponding system and is configured as follows: acquiring a data block for transaction and negotiation constraint content in the transaction of the data block so as to store transaction funds related to the transaction into a preset intelligent contract under the constraint of the negotiation constraint content and lock the transaction funds; based on a preset learning model, sensitive feature abstracts in the data blocks are extracted, feature vectors containing sensitive coefficients are generated according to the sensitive feature abstracts, and noise values are dynamically generated according to the feature vectors; performing noise addition processing on the data block according to the noise value, and encrypting the noise-added data block according to a preset encryption algorithm; and verifying the transaction information of the purchaser and the owner and the encrypted data block, so as to execute the distribution of transaction funds and the decryption of the data block according to a verification result. Direct exposure of the original data is avoided, and fairness of data transaction is guaranteed.

Description

Technical Field

[0001] This application relates to the technical field of data trading, and in particular to a dynamic adaptive trading method for data subject to interference. Background Technology

[0002] In the current era of rapid development of the digital economy, data, as a key factor of production, is experiencing a continuous increase in demand for circulation and trading. However, traditional data trading models contain irreconcilable contradictions that severely restrict the efficient utilization of data elements.

[0003] In traditional approaches, data transactions are often centered on the direct exchange of raw data. While this model allows buyers to obtain complete, unprocessed information, it has significant drawbacks. On the one hand, raw data contains a large amount of sensitive information, and direct transactions can easily lead to privacy leaks. On the other hand, data owners are cautious about transactions for fear of leaks or misuse of raw data, making it difficult for a large amount of valuable data to enter the market.

[0004] Meanwhile, traditional models lack effective data quality verification and transaction security mechanisms. Existing data transaction solutions mostly rely on smart contracts or zero-knowledge proofs to verify data validity, but this cannot be applied to all data. The validity of some data cannot be encoded into oracle functions within smart contracts, and similarly, ensuring the correctness of data encoded in zero-knowledge proof systems is quite challenging, sometimes even impossible. Furthermore, if substandard data quality is discovered after a transaction, there is a lack of fair and efficient dispute resolution mechanisms, and data owners face the risk of being refused payment after data delivery.

[0005] Therefore, there is an urgent need for a data trading solution that can both avoid the direct exposure of raw data and ensure the fairness of transactions. Summary of the Invention

[0006] This application aims to at least solve the technical problems existing in the prior art and provide a dynamic adaptive trading method for interference data.

[0007] This application provides a dynamic adaptive trading method for interference data, applied to a dynamic adaptive trading system for interference data. The method is characterized by the following steps: acquiring a data block for trading, and the negotiation constraints within the data block transaction, such that the buyer and owner of the data block deposit and lock transaction funds related to the transaction into a preset smart contract under the constraints of the negotiation constraints; extracting a sensitive feature summary from the data block based on a preset learning model, and generating a feature vector containing sensitivity coefficients based on the sensitive feature summary, to dynamically generate noise values ​​based on the feature vectors; adding noise to the data block based on the noise values, and encrypting the noise-added data block according to a preset encryption algorithm; verifying the transaction information of the buyer and owner, as well as the encrypted data block, to allocate the transaction funds and decrypt the data block based on the verification results.

[0008] In one embodiment of this application, based on the above technical solution, the step of adding noise to the data block according to the noise value and encrypting the noise-added data block according to a preset encryption algorithm includes: determining the data similarity between the noise-added data block and the original data block according to the distribution of the data block, and calculating the privacy strength of the noise value according to the noise type of the noise value; when both the data similarity and the privacy strength meet the preset evaluation criteria, the corresponding noise value is retained, and the noise-added data block is generated.

[0009] In one embodiment of this application, based on the above technical solution, when both the data approximation and the privacy strength meet a preset evaluation standard, retaining the corresponding noise value and generating a noisy data block includes: obtaining the retained noise value, and generating a first noise vector according to the distribution type and association parameters corresponding to the noise value; calculating a second noise vector according to the first noise vector and a preset entanglement probability formula; expanding the noise pair composed of the first noise vector and the second noise vector based on tensor product to obtain a noise matrix of the dimension corresponding to the data block; and encrypting the noise matrix based on a preset elliptic curve encryption algorithm to generate a noisy data block.

[0010] In one embodiment of this application, based on the above technical solution, the step of verifying the transaction information of the buyer and the owner, as well as the encrypted data block, to perform the allocation of transaction funds and the decryption of the data block according to the verification result includes: when the verification result is that the verification fails and / or the buyer has an objection, entering and executing a preset arbitration process to allocate the transaction funds according to the arbitration result output by the arbitration process; when the verification result is that the verification passes and the buyer has no objection, transferring the transaction funds to the allocation unit to allocate the transaction funds according to the negotiated constraints.

[0011] In one embodiment of this application, based on the above technical solution, when the verification result is that the verification fails and / or the buyer has objections, a preset arbitration process is entered and executed to allocate the transaction funds according to the arbitration result output by the arbitration process. This includes: within the arbitration process, verifying the availability of the data block through a preset expert pool and making a voting decision; based on the result of the voting decision, determining the party at fault between the owner and the buyer, and executing a reward and punishment process according to preset reward and punishment rules.

[0012] In one embodiment of this application, based on the above technical solution, the step of verifying the availability of the data block and making a voting decision through a preset expert pool in the arbitration process includes: each party to the transaction independently selects a corresponding number of self-selected experts from a preset arbitration panel according to a preset first ratio; based on the smart contract, a corresponding number of random experts are randomly selected from the preset arbitration panel according to a preset second ratio; and the self-selected experts and the random experts are combined to obtain a preset expert pool.

[0013] In one embodiment of this application, based on the above technical solution, the transaction funds include the purchase amount and first deposit of the buyer, and the second deposit of the owner; the step of determining the party at fault between the owner and the buyer according to the result of the voting decision, and executing the reward and punishment process according to the preset reward and punishment rules, includes: when the party at fault is the buyer, allocating the first deposit to the expert pool, and allocating the second deposit and the purchase amount to the owner; when the party at fault is the owner, allocating the second deposit to the expert pool, and allocating the purchase amount and the first deposit to the buyer.

[0014] The technical solutions of this application embodiment can bring at least the following effects through the above-described inventive content: This application generates feature vectors by training a digest of sensitive features in a data block, and dynamically generates noise values ​​based on the feature vectors. The data block is then encrypted based on the noise values ​​and a preset encryption algorithm, which avoids the direct exposure of the original data. At the same time, by storing and locking transaction funds related to the transaction in a preset smart contract under the constraints of negotiated constraints, and verifying the encrypted data block and the obtained transaction information, the allocation of transaction funds and the decryption of the data block are executed based on the verification results, which can improve the fairness of data transactions. Therefore, the technical solution of this application avoids the direct exposure of the original data while ensuring the fairness of data transactions.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this application. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating an exemplary embodiment of the dynamic adaptive trading method for interference data in this application; Figure 2 This is a schematic diagram of a Merkle hash tree illustrated in an exemplary embodiment of this application; Figure 3 This is a schematic diagram illustrating the information flow of a data transaction system according to an exemplary embodiment of this application; Figure 4 This is a flowchart illustrating a data transaction as shown in an exemplary embodiment of this application; Figure 5 This is a flowchart illustrating an arbitration mechanism as shown in an exemplary embodiment of this application. Detailed Implementation

[0017] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0018] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0019] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0020] The execution entity of the dynamic adaptive transaction method for interference data in this application includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in the embodiments of this application: a server, a terminal, etc. In other words, the dynamic adaptive transaction method for interference data can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0021] In existing technical solutions, data transactions mostly involve raw data. Even with encrypted transmission, the raw data is still directly exposed to the data buyer after decryption, posing a risk of sensitive information being misused or disseminated secondary.

[0022] In addition, during the noise addition process, in order to ensure that the original data of the data owner is not leaked, the noise addition process can only be operated by the data owner alone. The data buyer cannot verify whether the noise intensity meets the agreed privacy budget, which may lead to excessive noise addition and data invalidation. At the same time, there is a lack of reliable verification methods to verify whether the data block has been tampered with by the data buyer during the noise verification process.

[0023] Meanwhile, existing technical solutions rely on automatic arbitration of codes or centralized third-party arbitration, which has limitations in applications where data cannot be coded in some areas, risks of arbitration bias and untrustworthy third parties, and lack of transparency in dispute resolution processes, which can easily lead to secondary disputes. In addition, payment depends on the credit of both parties, and there may be situations where data buyers refuse to pay or data owners do not deliver the data after receiving payment, resulting in extremely high trust costs.

[0024] Figure 1 This is a flowchart illustrating a dynamic adaptive trading method for interference data, as shown in an exemplary embodiment of this application. Figure 1As shown in some embodiments of this application, in order to reduce the probability of data exposure during data trading and improve trading fairness, this application provides a dynamic adaptive trading method for interference data, applied to a dynamic adaptive trading system for interference data. The dynamic adaptive trading system for interference data is configured to perform at least the following steps: S100. Obtain the data block used for the transaction, as well as the negotiation constraints in the data block transaction, so that the buyer and owner of the data block will deposit the transaction funds related to the transaction into the preset smart contract and lock them under the constraints of the negotiation constraints. S110. Extract sensitive feature summaries from data blocks based on a preset learning model, and generate feature vectors containing sensitivity coefficients based on the sensitive feature summaries, so as to dynamically generate noise values ​​based on the feature vectors; S120. Add noise to the data block according to the noise value, and encrypt the noise-added data block according to the preset encryption algorithm. S130. Verify the transaction information of the buyer and the owner, as well as the encrypted data block, respectively, and execute the allocation of transaction funds and decryption of the data block according to the verification results.

[0025] It should be noted that for the English descriptions that appear throughout the text, you can refer to the explanation of the first appearance for comprehension.

[0026] Specifically, based on the aforementioned dynamic adaptive trading method for interference-prone data, in the initial stage of data trading, the data owner (DO) and data purchaser (DP) need to agree on the specific price of the traded data. (Purchase price, the price required to purchase the data block), the amount of deposit required from both parties. (First deposit, the deposit that the buyer needs to pay) and (The second security deposit, the deposit that the owner needs to pay), usually meets the following requirements. ,in, Multiple rounds of interactive negotiation were conducted regarding the margin ratio, the sensitive field classification system, and basic privacy constraint indicators. For example, specific constraints other than the margin, such as the specific scope of usage permissions, and usage scenario limitations under different permissions, etc.

[0027] For example, the sensitive field classification system includes a set of highly sensitive fields such as ID card numbers and bank card numbers, a set of moderately sensitive fields such as mobile phone numbers and addresses, a set of weakly sensitive fields such as age and gender, and a set of non-sensitive fields. The basic privacy constraint indicators include the lower limit of privacy protection strength and the data utility threshold, which serve as boundary conditions for noise generation.

[0028] After the above-mentioned negotiated constraints are agreed upon, the DP must, in accordance with the negotiated constraints of the Smart Contract (SC), transfer the transaction amount. With deposit Switching to a smart contract, DO will also synchronize... The funds are deposited into the same SC. At this point, the SC enters a fund-locked state. For example, its internal state variable status is set to pending. Only when all subsequent verification processes are passed and the DP has no objection to the transaction data, does not raise arbitration, or meets the time constraints, will the state transition to completed and the funds be allocated. Alternatively, if verification fails, it will jump to the funded state. This mechanism ensures the traceability of fund flows through the immutability of the blockchain.

[0029] Of course, the specific settings of various state variables in the original control program used to implement the dynamic adaptive trading method for interference data in the above example are only an exemplary reference to illustrate some of the implementation methods of this application, and do not impose any specific limitations.

[0030] Furthermore, to illustrate further, after the smart contract locks the transaction funds, DO first divides the original dataset Data into blocks of fixed block sizes, obtaining... .

[0031] Then, for each data block DO needs to extract the corresponding sensitive feature summaries. Including dimensional features Sensitive distribution characteristics and attribute features ,Right now .

[0032] Among them, dimensional features Used to distinguish whether data is high-dimensional or low-dimensional, sensitive distribution characteristics Represents the sensitivity ratio vector, attribute features Indicates whether the data type is discrete or continuous.

[0033] The noise value generation process in this application's technical solution is illustrated by example. A reinforcement learning (RL) agent model, Model, is initialized through a pre-defined Trusted Collaborative Center (TCC). The agent's state space S is defined as a sensitive feature summary, i.e. Action space Adjusting noise distribution parameters, such as standard deviation. or scale parameter Increase or decrease, initial strategy The agent is then sent to the DO after random initialization or pre-trained model settings.

[0034] DO summarizes the previously extracted sensitive features ,use The agent model is locally trained to compute gradients, and these gradients are sent to the Trusted Collaboration Center (TCC). The TCC receives and aggregates these gradients to update the RL agent. Value function, ,in It's a reward.

[0035] Furthermore, TCC builds a simulation environment in which the RL agent undergoes multiple rounds of training, each round being called an episode.

[0036] TCC obtains the current state from the environment. , It includes dimensional features Sensitive distribution characteristics and attribute features The feature vector for the current state RL agent selects an action Specific adjustments are made to the noise distribution parameters, such as adjusting... The choice of actions follows strategy. Then, a noise vector is generated based on the adjusted parameters. For the original data block Simulated noise addition is performed to obtain the noise-added data. .

[0037] In some embodiments of the application, when adding noise to a data block based on a noise value, the following steps may be included at least: Based on the distribution of data blocks, determine the data similarity between the noise-added data blocks and the original data blocks, and calculate the privacy strength of the noise values ​​based on the noise type of the noise values; When both data approximation and privacy strength meet the preset evaluation criteria, the corresponding noise value is retained, and a noisy data block is generated.

[0038] Specifically, based on the noisy data blocks, TCC evaluates the noise-adding effect using two core metrics. The first metric is utility loss. The Kullback-Leibler divergence is used to measure the difference in distribution between the original data and the noisy data. The smaller the value, the closer the distribution of the data after noise addition is to the original data, and the better the utility is preserved.

[0039] The second metric is actual privacy budget. The actual differential privacy strength satisfied is calculated based on the noise type, such as for a Laplace distribution. The smaller the value, the stronger the privacy protection. Rewards are calculated based on the evaluation results. As a response to the current action Feedback ,in and For hyperparameters, This represents the minimum level of privacy.

[0040] Finally, based on the reward renew Value function, ,in It's the learning rate. It is to perform an action The new state that is entered later It is a discount factor.

[0041] The above steps will be repeated multiple times. After each round, the agent's... Value functions more accurately reflect the true value of a strategy. By gradually optimizing and iterating until the RL agent converges, the agent has learned a stable optimal strategy and can automatically select the optimal noise parameter adjustment scheme based on any data characteristics, thus achieving a balance between privacy and utility.

[0042] Through the above implementation methods, the technical solution of this application can be based on reinforcement learning-driven adaptive noise optimization. By extracting sensitive feature summaries of data blocks to construct the RL agent state space, and adjusting the noise distribution parameters as the action space, the trusted collaboration center conducts multiple rounds of reinforcement learning training to dynamically optimize the noise generation strategy, and finally achieves a precise balance between privacy protection strength and data utility.

[0043] It should be noted that in traditional differential privacy data transactions, noise generation often relies on fixed parameters, easily leading to a dilemma of insufficient privacy protection or data utility loss. Furthermore, noise generation often employs independent noise addition modes, failing to consider the correlation structure between original data fields, which can easily disrupt the inherent logical relationships between fields. To address these technical problems, further, in some embodiments of this application, when both data approximation and the privacy strength meet preset evaluation criteria, the corresponding noise value is retained, and a noisy data block is generated. This may further include at least the following execution steps: Obtain the retained noise value, and generate a first noise vector based on the distribution type and associated parameters corresponding to the noise value; The second noise vector is calculated based on the first noise vector and the preset entanglement probability formula; The noise pair consisting of the first noise vector and the second noise vector is expanded based on the tensor product to obtain the noise matrix of the corresponding dimension of the data block; The noise matrix is ​​encrypted using a preset elliptic curve cryptography algorithm to generate a noisy data block.

[0044] It should be noted that, for the first noise vector in the above steps, the following example uses... The representation is performed, and the second noise vector is passed through To express.

[0045] Specifically, to illustrate by example, based on the learned optimization strategy TCC selects the final distribution type and parameters, and generates the corresponding noise vector. .

[0046] To preserve the relational structure information between fields in the original data and reduce the utility loss caused by independent noise addition, especially for high-dimensional relational data, noise is added using noise pairs. The resulting noise vector... It is the principal component of noise and determines the overall noise intensity.

[0047] After generating the principal components, TCC will use the sensitive feature summary. Medium sensitivity distribution characteristics Field correlation, calculate the Pearson correlation coefficient between fields. And the entanglement probability is calculated through a mapping function. ,in It is the magnification factor.

[0048] Then TCC generates a random number. ,like Then let ,otherwise .

[0049] Then generate a second random number. ,like Then let ,otherwise .

[0050] Finally, the associated noise components are calculated. Its value depends on the principal components. Entangled bit pairs generated previously The calculation formula is: ,in The correlation coefficient is... , To what extent Bell's inequality is violated, , It is the rotated bit. It is independent of random noise, ensure Not entirely by The decision retains a degree of randomness.

[0051] Based on the above steps, a pair of noises can be obtained. and .

[0052] Depending on the data dimension, tensor products can be used to transform two-dimensional noise pairs. and Expanded into a noise matrix of the corresponding dimensions .

[0053] Based on the above embodiments, in generating the noise matrix Then, TCC uses the Elliptic Curve Cryptography (ECC) algorithm with DP's public key. ,in The base point of the elliptic curve = , For DP's private key, Encryption is performed to obtain ciphertext. ,in , for The order is determined, and finally the encrypted noise matrix is ​​sent to DO for data noise interference.

[0054] Through the above implementation methods, the technical solution of this application can be based on the noise pair generation mechanism of quantum entanglement. By mining the correlation structure features between data fields, calculating the correlation degree and entanglement probability of fields to construct quantum entanglement constraint relationship, generating principal component noise and correlation noise with correlation dependence, forming a multidimensional noise matrix. While satisfying the privacy budget, it effectively preserves the correlation information between the original data fields and avoids the loss of high-dimensional data utility caused by independent noise addition.

[0055] Regarding the explanation of the aforementioned trusted collaboration center, the following clarification is provided: the trusted collaboration center can be an exemplary virtual transaction center in the technical solution of this application, used to receive and process various data during the transaction process, and output corresponding results according to the specific processing procedure. Furthermore, the feature analysis model can be established based on one or more existing algorithm models. For example, in the embodiments of this application, the feature analysis model can be established based on federated learning.

[0056] Federated learning is a distributed machine learning framework that enables joint modeling by allowing participants to train models locally and exchange only encrypted parameters. It eliminates the need to share raw data, effectively protecting privacy while improving model performance.

[0057] This application extracts a summary of sensitive features from a data block using the above method, and trains the extraction process based on a federated learning model to generate a feature vector containing sensitivity coefficients. Different distribution types are selected based on the vectors, and noise of different intensities is dynamically generated to ensure that data availability is maintained to the maximum extent while meeting the privacy budget.

[0058] Since the extraction of sensitive features changes continuously with the model training results, this application can dynamically generate noise values ​​based on the feature vectors trained by the model. The data block is then encrypted based on the noise values ​​and a pre-defined encryption algorithm, making the encrypted content less susceptible to interference and improving the transaction security of the data block.

[0059] The specific details of encrypting data blocks based on noise values ​​and a preset encryption algorithm can be explained using the following example: Specifically, the noise value generation process is illustrated below. The TCC of this application is based on feature vectors. Generate noise vector Specifically, TCC first considers dimensional features. Choose different distribution types. For high-dimensional data, choose the Gaussian distribution. High-dimensional data has complex field relationships, and the isotropic nature of the Gaussian distribution ensures that noise is uniformly diffused in multidimensional space, and its standard deviation is relatively low. , The adjustment factor is determined by the basic privacy constraint index; for low-dimensional data, the Laplace distribution is chosen because low-dimensional data has simple field relationships, and the Laplace distribution is more computationally efficient. .

[0060] It should be noted that in traditional data transaction encryption schemes, key generation often adopts a fixed parameter mode, without considering the structural differences and sensitive complexity of different data blocks. To solve these technical problems, further, in some embodiments of this application, after generating noise values, the owner performs lattice-based encryption on the data block, which may include at least the following execution steps: The sensitive feature vectors of the data block are transformed into high-dimensional point clouds, and the complexity features of the data block structure are extracted based on the high-dimensional point clouds according to the preset topology analysis. The public key parameters of the encryption algorithm are adjusted according to the complexity characteristics to generate a key pair that matches the structure of the data block, and the data block is encrypted based on the key pair.

[0061] Specifically, this means that when DO receives the message sent by TCC... Afterwards, at least the following implementation steps need to be performed: First, each data block Sensitive feature vectors It is considered as a high-dimensional point cloud, that is, each feature dimension corresponds to a coordinate axis in space, and the feature value corresponds to the coordinate value.

[0062] Then, the number of connected components of the point cloud is calculated using the Kruskal algorithm. The steps are as follows: calculate the pairwise distances of all points in the point cloud, connect the points in ascending order of distance, skipping any points where a connection would form a loop. The number of points in the final unconnected independent subset is the number of connected components. As a topological feature, it is used for dynamic adjustment of subsequent encryption parameters. The larger the value, the more complex the data block structure, and the stronger the encryption needs to be.

[0063] Next, a standard lattice basis matrix M is generated based on NTRU, which varies with topological features. Dynamically adjust the public key of lattice-based encryption , ,in This is a scaling factor to prevent the public key from becoming too large.

[0064] Private key Generated using the standard lattice basis reduction algorithm, ensuring compatibility with dynamic public keys. The private key must be able to efficiently decrypt content encrypted with the public key.

[0065] After generating the public-private key pair for lattice-based encryption, the data blocks... Mapped to binary polynomial Through public key Calculate lattice ciphertext This process leverages the computational difficulty of the lattice basis problem to ensure that a quantum computer cannot compute the solution in polynomial time. China Resumption This achieves quantum resistance.

[0066] Next, use DP's public key. ciphertext Homomorphic encryption is performed to obtain the encrypted double-layer ciphertext. Based on the additive homomorphism of the Paillier encryption algorithm, the encrypted double-layer ciphertext is processed within the ciphertext domain. Perform a noise-adding operation to obtain double-encrypted interference data. .

[0067] As can be seen, the technical solution disclosed in the above embodiments can generate dynamic keys based on topological data analysis. By converting data-sensitive feature vectors into high-dimensional point clouds and extracting data block structural complexity features using topological analysis, the public key parameters of lattice-based encryption are dynamically adjusted to generate lattice-based encryption key pairs that are compatible with the data structure. This achieves quantum attack protection while ensuring that the encryption strength matches the data-sensitive structural complexity.

[0068] Through the above implementation method, a two-layer encryption mechanism based on inner lattice-based encryption and outer homomorphic encryption is employed. First, the data owner uses lattice-based encryption to perform inner-layer encryption on the highly sensitive original data, and then homomorphic encryption is used for outer-layer processing to support noise addition to the ciphertext domain. This ensures the long-term privacy and security of highly sensitive data in the quantum era while preserving the efficiency and verifiability of data transactions, achieving a dual guarantee of quantum-resistant security and transaction practicality.

[0069] Meanwhile, through double-layer encryption and homomorphic characteristics, accurate verification of noise addition is achieved. The Trusted Collaboration Center verifies whether the encrypted original data block and the encrypted noise block are consistent with the encrypted interference data block in the ciphertext domain by using the encrypted original data block sent by the data owner and the encrypted interference data block sent by the data buyer, and by using the locally stored original interference noise. This mechanism directly verifies the compliance of the noise addition operation through cryptographic operations without leaking the original data.

[0070] in, Let be the modulus of the Paillier encryption algorithm, satisfying , and All are large prime numbers, and the final result is generated An encrypted jamming data block, i.e. Finally, DO will use DP's public key. Encrypted interference data blocks Send to DP.

[0071] Based on this, this application can block the exposure of the original data from the source through a noise-adding mechanism. The data owner only transmits encrypted and interfered data blocks to the data buyer. After decryption, the data buyer can only obtain the noise-adding data and cannot restore the original information, which solves the privacy leakage problem of direct trading of original data to a certain extent.

[0072] Meanwhile, in the technical solution of this application, lattice-based encryption is used for the inner layer of highly sensitive data blocks to achieve quantum resistance, and homomorphic encryption algorithm is used for the outer layer to ensure operation in ciphertext state. The synergy of the two encryption algorithms forms the advantage of long-term security and privacy.

[0073] To further illustrate, in the technical solution of this application, DP receives an encrypted interference data block sent by DO. Then, use your own private key. Performing the decryption operation yields the ciphertext. At this point, the smart contract will trigger a temporary authorization mechanism, allowing the DP to temporarily obtain... Usage permissions, calculation get , That is, the noisy, usable data required by DP.

[0074] In some embodiments of this application, based on the above embodiments, Figure 2 This is a schematic diagram of a Merkle hash tree illustrated in an exemplary embodiment of this application, as shown below. Figure 2 As shown, DO sends to DP At the same time, with each Construct a Merkle Hash Tree (MHT) for the leaf nodes.

[0075] Specifically, DO first computes each leaf node. hash value Then, the hash of the parent node is calculated recursively according to the hierarchy, that is... ,in This represents the index of the node in this level, starting from 0. This represents the level at which the node resides. For the leaf layer, This is the layer above the leaf layer, and so on. For example, the hash value of the first node in the leaf layer is represented as... The Merkel hash tree is constructed through the above calculations until the root hash is generated. ,in Represents the number of nodes.

[0076] After the build is complete, DO will With each The hash path corresponding to the node contains from arrive All intermediate node hashes are sent to the TCC for subsequent integrity verification.

[0077] Based on this, after receiving the MHT message sent by the DO, the TCC uses a cryptographically secure pseudo-random number generator to generate... Random index and will Send to DP and DO respectively.

[0078] As an example, in the specific technical solution of this application, DP needs to retrieve data from index S. Extract the corresponding encrypted block It is important to note that here... The DO (Domain) sends the data to the DP (Data Provider) at the beginning of the transaction, so the DP doesn't need to perform any encryption or other operations. It only needs to select the corresponding encrypted interference data block based on the index, which improves the overall transaction efficiency. Finally, the DP selects the... Simply forward it to the TCC for verification.

[0079] Based on the above embodiments, for DO, it is necessary to refer to the received index. The corresponding original data block First, use the GLONASS public key. Calculate lattice ciphertext Then through DP's public key Encryption obtained , , Then use your own public key Encryption is performed to obtain , Finally, the encrypted original data block is sent to the TCC.

[0080] After encrypting the data block, the technical solution of this application verifies the encrypted data block and the obtained transaction information, and transfers the transaction funds to the allocation unit to perform the allocation of transaction funds and decryption of the data block according to the verification results.

[0081] Specifically, after receiving the encrypted jamming data block sent by DP, TCC first checks the data sent by DP. Perform MHT verification.

[0082] TCC first uses the same hash algorithm as DO on the encrypted and scrambled data blocks that need to be verified. Calculate its hash value, denoted as Next, the path information of the corresponding cryptographic interference data block is extracted from the hash path provided by DO, and then from... Initially, the hash of the parent node is calculated layer by layer upwards along the hash path, eventually yielding a derived root hash. ,like Equal to the original Merkle root sent by DO If the result is positive, it means that the encrypted interference data block sent by DP has not been tampered with and is the original data block; otherwise, it means that the data block has been tampered with, triggering the transaction termination process.

[0083] If the verification passes, the program settings will be executed. Then, the verification of noise addition will be carried out.

[0084] First, the TCC sends the verified encrypted jamming data block to the DP. Using DO's public key Encrypt to obtain , .

[0085] Then TCC uses the noise from the corresponding index it initially generated. Perform the same encryption, but note that the noise used here is stored in the TCC, and the DO cannot forge it to pass verification.

[0086] It is evident that the noise in this application is dynamically generated by an independent, trusted third-party collaborative center based on data feature perception, and is encrypted using an encryption algorithm before being sent to the data owner for noise addition. This not only ensures the compliance of noise generation, but also ensures that the data owner can only perform noise addition operations on the data in the encrypted domain and cannot access the original noise value.

[0087] Then use DP's public key. Perform the first encryption to obtain , Then use DO's public key Perform secondary encryption to obtain , .

[0088] Finally, the encrypted raw data block was sent using DO. , Perform the following calculations within the ciphertext field: Based on the above calculation formula, after the calculation is completed, TCC generates a set of random numbers. , Then, the multiplicative homomorphism of the Paillier encryption algorithm is used to verify the result. ,like That is, the zero value of encryption proves If the noise added by the DO is correct, the transaction can continue; otherwise, it means that the DO added noise that was more sensitive than that generated by the TCC in order to better protect the privacy of its original data. More noise occurs, at which point SC automatically performs a refund operation, adjusting the Price and... Return to DP. Return to DO, transaction terminated.

[0089] As can be seen, the Trusted Collaboration Center of this application randomly generates an index, extracts a portion of data blocks for verification, uses Merkel proof to calculate the hash value of the extracted data blocks, and then derives the root hash layer by layer according to the hash path. If it is consistent with the root hash value provided by the data owner, it means that the sampled verification data block sent by the data purchaser is correct and has not been tampered with, which greatly reduces the computational cost and time cost of verification.

[0090] Through the above implementation methods, the dynamic noise generation mechanism based on data feature perception analyzes the distribution characteristics of the original data through federated learning, and then generates noise that is adapted to the data features. This not only ensures that highly sensitive data receives stronger privacy protection, but also avoids low-sensitivity data from losing statistical utility due to excessive noise, significantly improving the practicality of differential privacy technology in data transaction scenarios.

[0091] In some embodiments of this application, the process of verifying the transaction information of the buyer and the owner, as well as the encrypted data block, to allocate transaction funds and decrypt the data block based on the verification results may include at least the following steps: When the verification result is that the verification fails and / or the buyer has objections, the pre-set arbitration process will be initiated and executed to allocate the transaction funds according to the arbitration result output by the arbitration process. When the verification result is successful and the buyer has no objection, the transaction funds will be transferred to the allocation unit for allocation in accordance with the negotiated terms.

[0092] Specifically, based on the content of the above embodiments, the following further explanation is made: In the above verification process, after adding noise values ​​to the data block in the technical solution of this application, it is necessary to determine whether the noise addition to the encrypted data block is correct. If the noise addition is deemed correct, the transaction process continues; otherwise, it is terminated.

[0093] Specifically, the Trusted Collaboration Center conducts sampling checks on the noise addition. The Trusted Collaboration Center performs interference calculations on the sampled original data blocks encrypted with the public keys of both parties and the encrypted noise, verifies whether the noise addition results meet expectations, enforces the compliance of noise addition, and prevents data owners from maliciously manipulating data quality.

[0094] As can be seen, the technical solution of this application is based on the core design of transaction interference data rather than the original data, which reduces the risk of privacy leakage from the source. At the same time, it uses technical means to make the entire transaction process verifiable and disputes fair, providing an effective and innovative solution for the secure flow of data.

[0095] To illustrate, if the interference noise added by the DO is correct and the DP has no objection to the availability of the data, then the DP must send a confirmation signal to the SC within the time-lock period, such as 24 hours. This confirmation signal includes a digital signature. Using a private key for signing can prevent someone from maliciously forging confirmation signals.

[0096] SC received After successful verification, the pre-deposited transaction amount will be automatically credited. Send it to DO's account address, along with the margins for both DO and DP. and The refund will be returned to both parties' account addresses.

[0097] If the SC does not receive a message from the DP within the time-lock period If the DP has no objection to data availability, then it is assumed that the DP has no objection to the data availability; otherwise, the DP can propose arbitration. Therefore, if no data is received within the time-lock period... SC will also automatically perform the same operation as above.

[0098] In some embodiments of this application, based on the above embodiments, when the verification result is a failure and / or the buyer has objections, a preset arbitration process is initiated and executed to allocate transaction funds according to the arbitration result output by the arbitration process. This may include at least the following execution steps: Within the arbitration process, the availability of data blocks is verified through a pre-defined pool of experts, and a vote is taken to make a decision. Based on the voting results, the party at fault between the owner and the buyer is determined, and the reward and punishment process is carried out according to the preset reward and punishment rules.

[0099] Specifically, if the DP objects to the availability of data and requests arbitration, the arbitration process is initiated. This application introduces a pre-defined expert pool (EP) for data in different domains. Experts in the pool can assess the quality of the data. The number of experts K in the arbitration panel (AP) is set according to the scale of the transaction data; the larger the scale of the transaction data, the larger the value of K. When determining the experts for the arbitration panel, the DP selects from the EP. Renowned experts, DO selection Famous experts, remaining The experts will be randomly selected by SC using a random selection algorithm to form the arbitration panel. .

[0100] Meanwhile, to ensure that different experts do not interfere with each other during the voting process and to ensure the independence and fairness of the voting, the arbitration stage uses ring signature technology to achieve anonymous voting.

[0101] Each expert They all have a key pair ( , The set of ring members is .

[0102] Each expert Independent voting was conducted based on data quality checks, and the voting results were as follows. Where 0 indicates data is unavailable and 1 indicates data is available, the final voting results are... Obtain by signing .

[0103] To illustrate, based on the above-mentioned expert pool setup, when a dispute arises, each party selects 1 / 3 of the experts in the field, and the remaining 1 / 3 are randomly selected to form an arbitration panel. They vote anonymously using ring signature technology, and the results are stored on the blockchain for evidence.

[0104] Through the above implementation methods, anonymous voting avoids interference during the voting process, random selection ensures neutrality, and smart contracts automatically execute the adjudication results, thus achieving fairness and transparency in dispute resolution.

[0105] Based on the voting results of the aforementioned expert pool, the party at fault among the owners and buyers will be identified, and the reward and punishment process will be carried out according to the preset reward and punishment rules.

[0106] Specifically, the transaction funds include the buyer's purchase price and first margin, and the owner's second margin. Of course, the first and second margins mentioned here are just for ease of description, and their specific contents can be understood in correspondence with the margin for the purchase price and the owner's margin mentioned above.

[0107] Furthermore, based on the voting results, the party at fault between the owner and the purchaser is determined, and a reward and punishment process is executed according to preset reward and punishment rules, which may include at least the following steps: When the party at fault is the purchaser, the first deposit is allocated to the expert pool, and the second deposit and the purchase price are allocated to the owner. When the party at fault is the owner, the second margin is allocated to the expert pool, and the purchase money and the first margin are allocated to the purchaser.

[0108] Based on the above embodiments, the following illustrative description is provided, and all signatures... After submission to SC, SC will first verify each The validity of the signature is confirmed by checking whether the consistency condition of the ring signature is met, and the votes are counted.

[0109] like If this proves the data is unavailable, the DO is deemed dishonest, and the SC will automatically deduct the DO's pre-deposited margin. ,Will The proceeds from the arbitration are distributed to the experts in the AP as a reward, and the pre-deposited transaction amount in the DP is also used. and deposit The refund will be returned to DP's account address.

[0110] Conversely, if If the data is available, it means the DO (Data Provider) wants to obtain the data without paying, which constitutes dishonest behavior. Therefore, the SC (Security Controller) will automatically deduct the DP's (Data Provider's) margin. The proceeds will be distributed to the experts in the AP as a reward for arbitration, and the transaction amount due to the DO will also be distributed. and its deposit Send it to DO's account address.

[0111] As can be seen, the technical solution of this application introduces a margin mechanism, requiring both parties to deposit a margin in advance according to a certain percentage of the transaction price, and lock the transaction amount with the margin of both parties through a smart contract. With the help of the automatic execution and immutability of blockchain smart contracts, the flow of funds is ensured to be safe and traceable, and the existence of the margin enhances the honest intention of both parties to the transaction.

[0112] It should be noted that arbitration is not mandatory for this application, but optional. Under the arbitration mechanism of this scheme, dishonesty by either party will result in significant losses, so the probability of entering the arbitration stage is relatively small. The arbitration stage is only a preferred option to improve transaction efficiency.

[0113] Through the above implementation method, both parties deposit the margin and transaction amount into the contract before the transaction. Only after the data verification is passed and there is no dispute or the dispute is resolved, will the smart contract automatically complete the transfer or refund. This uses technical means to replace credit guarantee and reduces human intervention.

[0114] Figure 3 This is a schematic diagram illustrating the information flow of a data transaction system according to an exemplary embodiment of this application, such as... Figure 3 As shown, in order to more clearly disclose the technical solution of this application, the following further explanation is made in conjunction with the specific content of the above embodiments: The technical solution of this application breaks through the limitations of traditional raw data transactions. It innovatively uses interfering data as the core of the transaction and combines technologies such as federated learning, smart contracts, differential privacy, homomorphic encryption, lattice-based encryption, Merkle proof, and ring signature to construct a transaction mechanism that emphasizes both privacy protection and fairness throughout the entire process.

[0115] Specifically, Figure 4 This is a flowchart illustrating a data transaction as shown in an exemplary embodiment of this application, such as... Figure 4 As shown, the data transaction process in the technical solution of this application may include at least the following execution steps: 400. At the beginning of the transaction, the data owner and the data buyer negotiate and determine the data price, margin, sensitive field classification system and basic privacy constraint indicators. Both parties deposit the margin and transaction funds into a smart contract to lock the funds to ensure transaction security.

[0116] 410. After the funds in the smart contract are locked, the data owner divides the original data into blocks and extracts sensitive feature summaries. The trusted collaboration center initializes the reinforcement learning agent model and optimizes the noise generation strategy through multiple rounds of training.

[0117] 420. Generate a correlated noise matrix by combining the properties of quantum entanglement and encrypt it before sending it to the data owner.

[0118] 430. Data owners dynamically adjust lattice-based encryption parameters through topological data analysis, using lattice-based encryption to protect data from quantum errors, followed by homomorphic encryption. Noise is then added to the encrypted state to generate encrypted interference data blocks, which are sent to the data buyer. This dual approach of noise addition and encryption avoids privacy leaks, solving the core privacy pain point of traditional raw data transactions.

[0119] 440. To ensure data integrity and the correctness of noise addition, this application also introduces a Merkle hash tree and a sampling verification mechanism. The data owner generates a Merkle hash tree with encrypted interference data blocks as leaf nodes, submits the root value and hash path to the trusted collaboration center, and the trusted collaboration center randomly selects index values ​​for sampling inspection and sends them to both parties to the transaction. The data buyer submits the encrypted interference data block corresponding to the index, and the data owner submits the encrypted original data block corresponding to the index. After the trusted collaboration center verifies that the data has not been tampered with through Merkle proof, it further verifies whether the data owner has honestly added noise.

[0120] 450. If the noise is added correctly and the data buyer has no objection to the availability of the data, a confirmation signal is sent to the smart contract. The smart contract will transfer the transaction funds to the owner and refund the deposits of both parties within the agreed time.

[0121] Figure 5 This is a flowchart illustrating an arbitration mechanism as shown in an exemplary embodiment of this application, such as... Figure 5 As shown, based on the above transaction process, this application also proposes an arbitration mechanism. If the data buyer has no objection to the availability of the data, the smart contract will transfer the transaction funds to the owner and refund the deposits of both parties within the agreed time.

[0122] 500. If there is any objection, arbitration shall be initiated. Each party shall select one-third of the arbitrators from the pre-set pool of experts in the field, and the remaining one-third shall be randomly selected to form an arbitration panel.

[0123] 510. The arbitration panel will verify the availability of the data and vote on it. To prevent interference, ring signature technology will be used for anonymous voting. Finally, the voting results will be submitted to the smart contract.

[0124] 520. Based on the voting results, if the data owner's data quality is low, the smart contract will deduct the data owner's deposit and give it to the arbitration panel as a reward, while refunding the data buyer's transaction amount and deposit.

[0125] 530. If the data purchaser is dishonest and raises malicious objections, the data purchaser's deposit will be deducted and given to the arbitration panel as a reward. At the same time, the data owner's deposit will be refunded, and the transaction amount will be sent to the data owner.

[0126] The arbitration panel determines liability through anonymous voting using ring signatures. If the data quality is substandard, the data owner's deposit is deducted to reward the arbitrator, and the data purchaser's funds are refunded. If the data purchaser raises malicious objections, their deposit is deducted, and payment is made to the owner.

[0127] As can be seen, the technical solution of this application introduces a dynamic arbitration and ring signature voting mechanism. The arbitration panel consists of one-third experts selected by the data owner and one-third experts randomly selected by the smart contract. The experts vote anonymously through ring signatures to ensure the independence and fairness of the voting. The smart contract automatically executes rewards and penalties based on the voting results, reducing human intervention.

[0128] It should be noted that the dynamic adaptive trading system for interference data provided in this embodiment and the dynamic adaptive trading method for interference data provided in the above embodiments belong to the same concept. The specific methods of execution of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the dynamic adaptive trading system for interference data provided in this embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0129] In the description of this specification, the references to terms such as "an embodiment," "some embodiments," "example," "specific example," "a implementation," "a preferred implementation," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0130] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A dynamic adaptive trading method for interference data, applied to a dynamic adaptive trading system for interference data, characterized in that: The dynamic adaptive trading system for interference-oriented data is configured to perform the following steps: Obtain the data block used for the transaction, as well as the negotiation constraints in the data block transaction, so that the buyer and owner of the data block will deposit the transaction funds related to the transaction into a preset smart contract and lock them under the constraints of the negotiation constraints. Based on a preset learning model, a summary of sensitive features is extracted from the data block, and a feature vector containing sensitivity coefficients is generated based on the summary of sensitive features, so as to dynamically generate noise values ​​based on the feature vector; The data block is noise-added according to the noise value, and the noise-added data block is encrypted according to a preset encryption algorithm; The transaction information of the buyer and the owner, as well as the encrypted data block, are verified respectively, so as to perform the allocation of transaction funds and the decryption of the data block according to the verification results.

2. The dynamic adaptive trading method for interference data according to claim 1, characterized in that, The step of adding noise to the data block based on the noise value and encrypting the noise-added data block according to a preset encryption algorithm includes: Based on the distribution of the data blocks, determine the data similarity between the noise-added data blocks and the original data blocks, and calculate the privacy strength of the noise values ​​based on the noise type of the noise values; When both the data approximation and the privacy strength meet the preset evaluation criteria, the corresponding noise value is retained, and a noisy data block is generated.

3. The dynamic adaptive trading method for interference data according to claim 2, characterized in that, When both the data approximation and the privacy strength meet the preset evaluation criteria, the corresponding noise value is retained, and a noisy data block is generated, including: Obtain the retained noise value, and generate a first noise vector based on the distribution type and associated parameters corresponding to the noise value; The second noise vector is calculated based on the first noise vector and the preset entanglement probability formula; The noise pair consisting of the first noise vector and the second noise vector is expanded based on the tensor product to obtain the noise matrix of the corresponding dimension of the data block; The noise matrix is ​​encrypted using a preset elliptic curve cryptography algorithm to generate a noisy data block.

4. The dynamic adaptive trading method for interference data according to claim 1, characterized in that, The step of adding noise to the data block based on the noise value and encrypting the noise-added data block according to a preset encryption algorithm includes: The sensitive feature vector of the data block is converted into a high-dimensional point cloud, and the complexity features of the data block structure are extracted based on the high-dimensional point cloud according to a preset topology analysis. The public key parameters of the encryption algorithm are adjusted according to the complexity characteristics to generate a key pair that matches the structure corresponding to the data block, and the data block is encrypted based on the key.

5. The dynamic adaptive trading method for interference data according to claim 1, characterized in that, The step of verifying the transaction information of the buyer and the owner, as well as the encrypted data block, respectively, and then performing the allocation of transaction funds and the decryption of the data block based on the verification results, includes: When the verification result is that the verification fails and / or the buyer has objections, the preset arbitration process is initiated and executed to allocate the transaction funds according to the arbitration result output by the arbitration process. When the verification result is successful and the buyer has no objection, the transaction funds are transferred to the allocation unit to allocate the transaction funds in accordance with the negotiated constraints.

6. The dynamic adaptive trading method for interference data according to claim 5, characterized in that, When the verification result is a failure and / or the buyer objects, a preset arbitration process is initiated and executed to allocate the transaction funds according to the arbitration result output by the arbitration process, including: Within the arbitration process, the availability of the data block is verified through a pre-defined pool of experts, and a vote is taken to make a decision. Based on the voting results, the party at fault between the owner and the purchaser is determined, and a reward and punishment process is executed according to the preset reward and punishment rules.

7. The dynamic adaptive trading method for interference data according to claim 6, characterized in that, Within the arbitration process, the availability of the data block is verified through a pre-defined pool of experts, and a voting decision is made, including: Both parties to the transaction independently select a corresponding number of self-selected experts from the pre-set arbitration panel according to a pre-set first proportion. Based on the smart contract, a corresponding number of random experts are randomly selected from the preset arbitration panel according to a preset second ratio value. The self-selected experts are combined with the randomized experts to obtain a preset expert pool.

8. The dynamic adaptive trading method for interference data according to claim 6, characterized in that, The transaction funds include the purchase price and first deposit of the buyer, and the second deposit of the owner; the process of determining the party at fault between the owner and the buyer based on the voting decision, and executing the reward and punishment process according to the preset reward and punishment rules, includes: When the party at fault is the purchaser, the first deposit is allocated to the expert pool, and the second deposit and the purchase price are allocated to the owner; When the party at fault is the owner, the second security deposit is allocated to the expert pool, and the purchase money and the first security deposit are allocated to the purchaser.

Images