A blockchain-based multi-source heterogeneous sensing data right authorization and privacy sharing method

Abstract

This invention discloses a blockchain-based method for the confirmation of ownership and privacy sharing of multi-source heterogeneous sensing data. The method includes: de-identifying and encapsulating multi-source heterogeneous data into standardized data packets, uploading them to a decentralized storage network, and simultaneously minting dynamic digital assets with updatable dynamic attributes on the blockchain; when the asset participates in a sharing task, off-chain nodes evaluate its marginal contribution based on game theory, generate zero-knowledge proofs, and submit them to the blockchain; a value assessment smart contract calculates the comprehensive data value by combining dynamic attributes and contribution; and an incentive allocation contract automatically calculates and settles the revenue share based on this value, while updating the asset's historical contribution and the provider's reputation. This invention achieves dynamic and accurate pricing of data under privacy protection and trustless distribution based on contribution, effectively stimulating the willingness to share high-quality data.

Description

Technical Field

[0001] This invention relates to the field of data privacy and security computing technology, and in particular to a method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain. Background Technology

[0002] With the development of the Internet of Things, edge computing, and industrial digitalization, massive amounts of multi-source heterogeneous sensing data are growing exponentially, and data has become a core production factor driving the new digital economy. Against this backdrop, to break down data silos and promote the secure flow of data elements, blockchain-based data ownership confirmation and sharing technologies have emerged. Blockchain, with its decentralized, immutable, and traceable characteristics, provides a solid foundation of trust for defining ownership and tracing the flow of sensing data assets. Simultaneously, combined with privacy-preserving computing paradigms based on cryptographic systems (such as zero-knowledge proofs and homomorphic encryption), heterogeneous data can be shared across domains without revealing the original plaintext information, achieving usability without visibility. This greatly promotes the value release and collaborative computing of multi-source sensing data in various complex intelligent application scenarios.

[0003] However, existing blockchain-based data ownership confirmation and sharing solutions still face several technical bottlenecks in practical applications. Specifically, traditional ownership confirmation mechanisms often employ static notarization, merely hashing data onto the blockchain. This lack of continuous tracking of the dynamic attributes of data assets throughout their lifecycle (e.g., quality evolution, access popularity, provider reputation) leads to a rigid characterization of data value, failing to accurately reflect its timeliness. Furthermore, in multi-party data sharing tasks, due to the modal heterogeneity of multi-source sensing data, existing technologies often employ a crude, uniform pricing or average allocation strategy. This fails to accurately quantify the actual incremental utility of a single data source to the overall task (i.e., marginal contribution), resulting in distorted value assessment and unfair incentive distribution, severely inhibiting the willingness of high-quality data providers to share. Moreover, when introducing complex value assessment and privacy screening, existing solutions struggle to verify the honesty of the computation process while ensuring data privacy. They lack efficient off-chain verification of complex computations and lightweight on-chain consensus mechanisms, making it difficult to support high-concurrency, large-scale data circulation networks. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a blockchain-based method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data, to address the problems mentioned in the background section.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain, comprising: Multi-source heterogeneous sensing data is encapsulated into standardized data packets, and the standardized data packets are uploaded to a decentralized storage network to obtain a storage location identifier. A smart contract is invoked to forge a dynamic digital asset for the standardized data packets on the blockchain. The dynamic digital asset records the storage location identifier and a set of dynamic attributes that can be updated by the smart contract. When the dynamic digital asset participates in a data sharing task, the off-chain computing node evaluates the contribution of the dynamic digital asset in the data sharing task, and submits the contribution, along with the dynamic attributes read from the dynamic digital asset, to the value assessment smart contract, which then calculates a comprehensive data value. The incentive distribution contract automatically calculates and settles the incentive share due to the dynamic digital asset owner from the total revenue of the data sharing task based on the calculated comprehensive data value.

[0007] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention creates dynamic digital assets with dynamic attributes on the blockchain and introduces a state write-back mechanism based on a time decay factor, so that the intrinsic attributes of data assets can dynamically evolve over time and with the participation of tasks, truly reflecting the timeliness of data value.

[0008] 2. This invention introduces a marginal contribution quantification model based on game theory Shapley value. By stripping away the actual utility increment of a single data source to the overall task and combining it with market popularity indicators for on-chain weighted aggregation, it abandons the traditional extensive model of charging based on data volume or number of records. Combined with a zero-tolerance negative value truncation protection mechanism and automatic liquidation by smart contracts, it fundamentally ensures the absolute fairness of profit distribution and stimulates the willingness of high-quality data owners to share.

[0009] 3. Furthermore, this invention eliminates the default risk associated with pre-payment or post-payment transactions in traditional data trading by using on-chain asset custody and smart contract atomic transfers for data requesters. Simultaneously, after revenue settlement, a hyperbolic tangent nonlinear smoothing function is triggered to automatically update the data provider's reputation, constructing a virtuous cycle of contribution assessment, utility-based incentives, asset and reputation appreciation, and attracting more high-value tasks. This effectively curbs malicious or fraudulent transactions. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the overall process of a blockchain-based method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data, as described in one embodiment of the present invention. Detailed Implementation

[0011] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0012] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0013] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0014] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0015] Example 1 Reference Figure 1 This is the first embodiment of the present invention, which provides a method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain, including: S1. Encapsulate multi-source heterogeneous sensing data into standardized data packets, upload the standardized data packets to a decentralized storage network to obtain storage location identifiers, call smart contracts, and mint a dynamic digital asset for the standardized data packets on the blockchain. The dynamic digital asset records the storage location identifier and a set of dynamic attributes that can be updated by the smart contract.

[0016] It's important to note that in multi-party IoT and edge computing scenarios, the data collected by sensing devices (such as high-definition cameras, temperature and humidity sensors, and radar) exhibits strong heterogeneity in terms of modality, sampling frequency, and data structure. Directly uploading or sharing the raw plaintext data to the blockchain would not only lead to extremely high storage costs but also pose serious risks of privacy breaches and difficulties in uniformly measuring value. Therefore, it is necessary to clean, de-identify, and format the raw sensing data.

[0017] Furthermore, for the original heterogeneous sensing data of multiple modalities, this embodiment uses an adaptive multimodal desensitization autoencoder deployed on the edge node side for processing. The network structure of the encoder includes: an input layer, independent encoder groups for different modal data (for example, a one-dimensional convolutional 1D-CNN combined with LSTM layers is used for temporal sensor data, and a ResNet residual network is used to extract spatial features for image / video frames), a latent space feature layer with differential privacy constraints, and a decoder layer.

[0018] Specifically, heterogeneous data from multiple sources are input into their respective encoders and mapped to high-dimensional feature vectors of a unified dimension. Then, in the latent space feature layer, a Laplacian mechanism is introduced for differential privacy-preserving noise addition, thereby removing sensitive individual information while retaining the overall statistical distribution characteristics (availability) of the data. The feature representation formula after noise addition is as follows: in, Indicates the encoder's network parameters The feature mapping function is as follows; This is raw, heterogeneous sensing data; The scale parameter is represented as The Laplace noise distribution; The global sensitivity of this type of perceived data (i.e., the maximum L1 norm change in the feature output caused by changing any single data record, representing the ease with which the data can be reverse-engineered). This is a privacy budget parameter; its value depends on the data source type. For highly sensitive data (such as surveillance videos containing faces), a very small budget is allocated. This injects strong noise; for low-sensitivity data such as ambient temperature, a larger amount is allocated. Preserve higher data accuracy.

[0019] Furthermore, the features after adding noise The decoder reconstructs the output, completing the data cleaning and desensitization, and finally transforming it into a unified JSON or Protobuf format sequence.

[0020] It should be noted that the aforementioned adaptive multimodal desensitization autoencoder needs to undergo a pre-training process before being deployed to edge nodes. This pre-training process is as follows: First, historical plaintext sensing data for the corresponding modality is collected as an unlabeled training sample set. This training sample set is then input into an autoencoder for training, where the joint loss function of the autoencoder is... It consists of reconstruction loss and privacy regularization term, and the formula is: in, The original input data, To reconstruct the output data for the decoder, It measures the fidelity (i.e., usability) of the data; The feature latent space distribution of the autoencoder output. The target prior distribution after introducing Laplace noise, The KL divergence is used to measure the difference between the two. Hyperparameters designed to balance usability and privacy.

[0021] Then, the Adam optimizer is used to iteratively update the network parameter weights in the autoencoder and decoder through the backpropagation algorithm until the joint loss function converges.

[0022] Finally, the trained network parameters are solidified and distributed to each edge computing node to perform local de-identified inference.

[0023] It should be noted that this pre-training operation enables the present invention to meet privacy compliance requirements while ensuring that data availability features are not lost.

[0024] Furthermore, it is encapsulated according to a unified data metadata standard to generate standardized data packets.

[0025] Specifically, after anonymization and formatting, in order for blockchain smart contracts to uniformly identify and schedule this heterogeneous data, the data needs to be encapsulated according to a predefined unified data metadata standard. This standard strictly defines a three-layer structure for data packets: Static attribute structure: includes the data generation timestamp, unique identifier (DID) of the sensing device, geographical coordinate range of the device, and data modality label. Once the static attributes are identified, they cannot be tampered with and are the cornerstone for tracing the physical source of the data.

[0026] Initial quality metric (static quality assessment): To eliminate persistent invalid data generated by crashed sensors, the initial physical information richness of the current batch of data packets is calculated based on information entropy as a basic quality metric. : in, The discretized sensory data features state; This represents the probability of this state occurring in the current data packet. These are the prior weights for this type of modal data; The higher the value, the richer the dynamic sensing information contained in the data packet.

[0027] Dynamic attribute reserved slots: Reserve an expandable field structure for dynamic digital assets that will be mapped on the blockchain.

[0028] Furthermore, it is uploaded to a decentralized storage network to obtain a storage location identifier.

[0029] Specifically, the standardized data packet (which is still quite large due to the presence of a large amount of anonymized plaintext sensor data) is uploaded to a decentralized storage network such as IPFS (InterPlanetary File System). The IPFS network will calculate a unique hash value based on the content of the standardized data packet using the SHA-256 algorithm, and return a unique hash value as the storage location identifier (i.e., Content Addressing Identifier, CID).

[0030] It should be noted that this step is mainly used to decouple the on-chain and off-chain data content storage from the asset ownership confirmation logic. This avoids the state explosion of the blockchain and uses the hash collision resistance feature to ensure the absolute immutability of off-chain data packets.

[0031] Furthermore, smart contracts are invoked to mint dynamic digital assets on the blockchain and initialize dynamic attributes.

[0032] Specifically, the obtained storage location identifier, along with encapsulated static attributes and initial quality metrics, is sent to the blockchain node as the transaction payload. A smart contract (e.g., an asset ownership contract conforming to and extending the ERC-1155 / ERC-721 standard) is triggered and executed, forging a dynamic digital asset for this standardized data packet within the blockchain state machine. Unlike traditional blockchain credentials that only serve as static proof, this smart contract not only records the storage location identifier but also initializes a set of dynamic attributes in the contract state storage area that can be invoked and updated across contracts by subsequent smart contracts (e.g., value assessment contracts, identity contracts). Based on the actual data circulation lifecycle, these dynamic attributes are specifically instantiated as the following state variables: Data quality score: Starting from the initial basic quality indicators, the score is dynamically adjusted based on the cross-validation results of other requesting nodes after participating in multiple federated computing or shared tasks.

[0033] Data access count: Initialized to 0. Each time the access control smart contract successfully authorizes the reading of the data corresponding to the storage location identifier, this variable is incremented by 1, directly reflecting the real demand for the data in the market.

[0034] Historical contribution value: This records the cumulative marginal utility value created by the asset in all historical data sharing and collaborative tasks since its creation, reflecting the asset's normalized value.

[0035] The data provider's reputation score is mapped to an identity contract and bound to the edge node (DID) that submits the data. If the provider consistently provides high-quality and valid data, the reputation score accumulates positively; if the provider is detected to have provided false or spliced ​​data, a penalty mechanism is triggered to deduct the reputation score.

[0036] S2. When a dynamic digital asset participates in a data sharing task, the off-chain computing node evaluates the contribution of the dynamic digital asset in the data sharing task. The contribution, along with the dynamic attributes read from the dynamic digital asset, is submitted to the value assessment smart contract, which calculates a comprehensive data value.

[0037] It's important to note that in multi-source, heterogeneous sensing data circulation scenarios, the true value of data is not static but highly dependent on the specific application context (i.e., the data sharing task). For example, the same segment of high-definition camera data plays drastically different roles in traffic flow statistics and nighttime security alarm tasks. Furthermore, directly processing complex data utility assessments on the blockchain leads to high gas fees and poses privacy risks. Gas fees, in this context, refer to the computational resource costs paid by users in a blockchain network for executing transactions or smart contract operations, similar to fuel costs for a car. These fees compensate validators for their transaction processing work and prevent network abuse. Therefore, this embodiment employs a collaborative architecture combining off-chain privacy computing and on-chain value aggregation.

[0038] Furthermore, before the data sharing task begins, a privacy access screening based on zero-knowledge proofs is performed.

[0039] Specifically, to prevent malicious requesters from using data evaluation as a pretext for data theft, this invention introduces an assertion screening mechanism. Specifically, this mechanism first involves the data requester defining a data screening assertion (e.g., requiring sensing data with a temperature greater than 30°C and located within a target industrial area) and publishing its hash value as a public state. Subsequently, the data provider performs calculations locally on an edge device, inputting the data into a conditional judgment circuit built based on the zk-SNARKs protocol without exposing the original plaintext data. In this circuit structure: the local sensing data plaintext and its private encryption key are used as private inputs; the assertion condition hash set by the data requester and the on-chain recorded identifier of the dynamic digital asset's ownership storage location are used as public inputs. The circuit performs assertion logic locally, verifying whether the private data meets the public assertion conditions and whether the data indeed belongs to the corresponding source of the storage location identifier. If the local data satisfies the assertion, a first zero-knowledge proof containing only a Boolean result (True) is generated. Finally, this first zero-knowledge proof is submitted to an access control smart contract on the blockchain for verification. Only after successful verification will the smart contract allow the dynamic digital asset to be officially added to the current data sharing task pool.

[0040] It should be noted that the introduction of this assertion screening mechanism mainly realizes the supply and demand matching of data under the condition of availability but not visibility, filtering out irrelevant or inferior data from the source, thereby reducing the waste of complex evaluation computing power.

[0041] Furthermore, the contribution of dynamic digital assets is evaluated by off-chain computing nodes using a game theory-based marginal contribution quantification method.

[0042] Specifically, for each dynamic digital asset in a data sharing task (e.g., multi-source sensing data fusion computation or federated learning model training), due to the heterogeneity of the data in terms of modality and information content, traditional average allocation mechanisms cannot reflect the true utility of individual data. This embodiment introduces the Shapley value from game theory as an evaluation model, and determines the final contribution by iteratively calculating the marginal utility increment under different data combinations.

[0043] Specifically, let the total set of dynamic digital assets participating in the current data sharing task be... Any one of the data asset nodes is denoted as ( Define the utility function. Representing a subset of data The overall utility gain generated when jointly participating in a task (e.g., the information entropy gain in fusion computing, or the improvement in model accuracy in federated learning). Then, data assets. marginal contribution The calculation formula is: in, The total number of data items participating in the shared task; This refers to all assets that do not include data assets. Subset combination; For subset The number of data points in the data; This indicates that data assets Add to the original subset After that, the marginal utility increment of the overall utility of the task occurs.

[0044] It should be noted that this formula not only considers the value of the data asset in its individual form, but also exhaustively lists the synergistic effects that can be brought about when the data is combined with other multi-source heterogeneous data in the task. By calculating the expected value of the marginal increment under all possible permutations and combinations, the true technical contribution of the data asset in this specific task can be extracted in an absolutely fair manner.

[0045] Furthermore, off-chain computing nodes generate a second zero-knowledge proof for the evaluation process and submit it to the blockchain.

[0046] Specifically, to ensure the honesty of the off-chain evaluation process and prevent off-chain nodes from forging or tampering with contribution results, after the off-chain computing nodes complete the contribution evaluation, the calculation trajectory of the Shapley value needs to be transformed into an arithmetic polynomial, and a second zero-knowledge proof reflecting the correctness of the calculation process needs to be generated using technologies such as ZK-Rollup. Then, the off-chain nodes encapsulate the calculated contribution and the second zero-knowledge proof together into a transaction payload and submit it to the value evaluation smart contract on the blockchain.

[0047] Furthermore, the value assessment smart contract is executed on-chain for verification, and the overall data value is calculated.

[0048] Specifically, after receiving the submitted data, the value assessment smart contract first performs on-chain cryptographic verification of the second zero-knowledge proof. It is important to emphasize that this verification process consumes minimal computational resources. The computational complexity is constant. If verification fails, the submission is rejected and a malicious penalty is triggered; if verification passes, it indicates that the contribution of the off-chain submission is credible. Once verification is successful, the smart contract will automatically read the dynamic attributes of the dynamic digital asset (including the number of times the data has been accessed, the reputation score of the data provider, etc.) from the blockchain state storage area, and combine them with the basic quality indicators in S1 to perform a weighted aggregation of the comprehensive data value.

[0049] Furthermore, the weighted set process is as follows: First, market popularity indicators are calculated based on dynamic attributes. The formula is as follows: in, To determine the number of times data read from dynamic attributes is accessed, a logarithmic function is used. This is to smooth out the explosive effect of metrics caused by excessive access to top data, thus conforming to the economic law of diminishing marginal utility. The reputation score of the data provider reflects the long-term historical credit of the data source; and This is the adjustment coefficient.

[0050] It should be explained that the market popularity index is calculated to characterize the popularity of data from sociological and market supply and demand dimensions.

[0051] Subsequently, the value assessment smart contract weights and aggregates the basic quality indicators (intrinsic physical attributes), market popularity indicators (external market attributes), and the contribution of this task (task context attributes) to obtain the final comprehensive data value. : in, , and These are the weighting coefficients for each of the above indicators.

[0052] It should be noted that, in order to ensure the overall democracy and adaptability of the system, these weight coefficients are not hard-coded in the contract, but are dynamically determined by the data sharing alliance members through voting in the on-chain governance contract (DAO) and recorded in the value assessment smart contract.

[0053] S3. The incentive distribution contract automatically calculates and settles the incentive share due to the dynamic digital asset owner in the total revenue of the data sharing task based on the calculated comprehensive data value.

[0054] It should be noted that in traditional centralized data trading markets, default and trust crises often arise due to the practice of payment before payment or payment before receipt, and the crude, uniform pricing model cannot effectively incentivize the continuous production of high-quality data. Therefore, this embodiment, based on the "code is law" characteristic of blockchain smart contracts, constructs an automatic, trustless mechanism for distribution based on work and value feedback.

[0055] Furthermore, before the data sharing task begins, the on-chain assets of the total revenue from the data sharing task are held in custody.

[0056] Specifically, when a data requester publishes a multi-source heterogeneous sensing data sharing task (e.g., initiating a federal traffic model training for a specific geographical area), the total revenue of the task (in the form of blockchain native tokens or stablecoins) must be pre-locked in the segregated fund pool of the smart contract by calling the payment interface of the incentive distribution contract.

[0057] It should be noted that this escrow operation aims to eliminate the risk that data requesters will refuse to pay after obtaining data utility, thereby providing absolute liquidity guarantee for automated clearing.

[0058] Furthermore, after receiving the aggregated data value of all participating tasks, the incentive allocation contract performs automated calculation of the incentive share.

[0059] Specifically, once the value assessment smart contract in S2 completes the comprehensive data value assessment of all dynamic digital assets participating in the current task, the incentive distribution contract will be automatically triggered. To prevent the negative marginal contribution caused by poor-quality data from causing the smart contract's revenue sharing logic to collapse, this invention introduces a value truncation protection mechanism. The incentive distribution contract first performs a non-negative mapping on the comprehensive data value of all inputs. The incentive share is then calculated based on the proportion of effective value to total effective value. The calculation formula is as follows: in, For the first The incentive share that each dynamic digital asset (owner) is entitled to in this task; The total collection of dynamic digital assets that ultimately successfully participate in and complete this data sharing task; The effective comprehensive value after nonnegative mapping; The sum of the effective combined value of all participating tasks; The total revenue of tasks pre-managed by the demander in the contract. It should be noted that in the above formula, if a data asset's effective comprehensive value after non-negative mapping is 0 due to providing poor-quality data, its incentive share for this instance will be 0, and its misconduct will be severely punished in subsequent reputation status write-backs. This not only ensures the robustness of mathematical operations on the blockchain but also embodies the fair principle that no one benefits without effective contribution.

[0060] Furthermore, it executes automatic transfers and settlements to the addresses of each dynamic digital asset owner.

[0061] Specifically, the incentive allocation contract calculates the array of all shares. Then, using the native transfer instructions of the blockchain virtual machine, a batch transfer is initiated at once to the owner wallet addresses of each dynamic digital asset bound to the blockchain. It is important to note that this transfer process is atomic, meaning that either all data providers successfully receive their corresponding incentive shares, or the entire state is rolled back. There are no intermediate states of partial success or partial failure, thus ensuring absolute fairness in multi-party settlement and consistency of the ledger.

[0062] Furthermore, the smart contract automatically triggers the write-back and update of the dynamic attributes of the dynamic digital assets and the reputation of the data provider.

[0063] It is important to emphasize that the confirmation and sharing of rights based on blockchain is not a one-time, static transaction, but a closed-loop ecosystem where data value continuously evolves. To accurately reflect the timeliness and credit accumulation of data assets throughout their lifecycle, after the aforementioned fund settlement, the incentive distribution contract will trigger the identity contract and asset confirmation contract through a cross-contract call mechanism, executing the following update operations: First, the historical contribution value attribute of the dynamic digital asset is updated. To emphasize the timeliness of the data (i.e., recently high-performing data should have higher weights, while older contributions will gradually decay), this invention uses an exponential moving average (EMA) algorithm with a time decay factor for dynamic write-back: in, To contribute value to the updated history; It contributes value to the old history prior to this mission; This is a preset time decay factor (e.g., a value range of [0.1, 0.3], representing the rate of forgetting historical records). This refers to the actual incentive share obtained in this instance; Adjustment weights to balance commercial value and pure technical contribution.

[0064] It should be noted that this update operation can continuously increase the intrinsic value of high-quality data assets, while automatically depreciating them.

[0065] Secondly, the reputation score of the data provider recorded in the identity contract is updated. To prevent the reputation score from inflating or fluctuating drastically due to a single abnormally high gain, this embodiment uses a nonlinear smooth growth model based on the hyperbolic tangent function to update the reputation: in, and These are the reputation scores of the data provider before and after the update; This is the maximum increment of reputation points that can be awarded for a single task, used to curb fraudulent activities such as order manipulation. This is the dynamic sensitivity scaling factor, used to scale the actual contribution to the appropriate working range of the tanh function; It is a hyperbolic tangent activation function, and its output range is between (-1, 1).

[0066] It should be further noted that, due to the significant differences in the physical dimensions and absolute values ​​of marginal contributions across different types of data sharing tasks, the dynamic sensitivity scaling factor is set to the reciprocal of the average of all baseline data contributions in the current task (i.e., ), used to standardize and map the absolute contribution generated under heterogeneous tasks to the unsaturated linear working interval of the tanh function (usually [-2,2]). Furthermore, as data providers consistently offer data with positive contributions, their reputation will steadily rise. In new data-sharing tasks, due to their higher reputation scores, they will more easily obtain a higher initial system weight through the market popularity assessment in S2; however, if the system detects that they have uploaded inferior or even malicious data ( This smoothing function will quickly deduct its reputation score.

[0067] It should be noted that, through the above processing, this invention establishes a virtuous cycle of positive incentives at the system architecture level, which consists of contribution evaluation → utility-based incentives → asset / reputation appreciation → attracting more tasks, thus solving the problem that owners of high-quality data are unwilling or afraid to share.

[0068] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for multi-source heterogeneous sensing data rights and privacy sharing based on blockchain, characterized in that, include: Multi-source heterogeneous sensing data is encapsulated into standardized data packets, and the standardized data packets are uploaded to a decentralized storage network to obtain a storage location identifier. A smart contract is invoked to forge a dynamic digital asset for the standardized data packets on the blockchain. The dynamic digital asset records the storage location identifier and a set of dynamic attributes that can be updated by the smart contract. When the dynamic digital asset participates in a data sharing task, the off-chain computing node evaluates the contribution of the dynamic digital asset in the data sharing task, and submits the contribution, along with the dynamic attributes read from the dynamic digital asset, to the value assessment smart contract, which then calculates a comprehensive data value. The incentive distribution contract automatically calculates and settles the incentive share due to the dynamic digital asset owner from the total revenue of the data sharing task based on the calculated comprehensive data value.

2. The blockchain-based multi-source heterogeneous perception data ownership certification and privacy sharing method according to claim 1, characterized in that, The dynamic attributes include: Data quality score, number of times data was accessed, historical contribution value, and reputation score of the data provider. 3.The method of claim 1, wherein, Calculating the value of the comprehensive data includes: The comprehensive data value is obtained by weighting and aggregating the preset quality indicators, market popularity indicators, and contribution of the dynamic digital assets. The market popularity index is calculated based on the number of times the data is accessed in the dynamic attributes and the reputation score of the data provider. 4.The method of claim 1 or 3, wherein, The assessment of the contribution includes: A game theory-based marginal contribution quantification method is adopted to determine the contribution degree for each dynamic digital asset in the data sharing task by iteratively calculating its marginal utility increment under different data combinations.

5. The blockchain-based multi-source heterogeneous perception data ownership certification and privacy sharing method of claim 4, wherein, After completing the contribution assessment, the off-chain computing node generates a zero-knowledge proof for the assessment calculation process and submits the zero-knowledge proof and the contribution together to the value assessment smart contract. After receiving the contribution score, the value assessment smart contract performs on-chain verification of the zero-knowledge proof. Once the verification is successful, it calculates the comprehensive data value.

6. The method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain as described in claim 1, characterized in that, Before the data sharing task begins, the following is also included: The data requester defines a data filtering assertion; The data provider verifies the assertion locally without exposing the original data, and generates a first zero-knowledge proof for the verification process; The first zero-knowledge proof is submitted to the access control smart contract for verification, thereby enabling privacy access screening of the data corresponding to the dynamic digital asset.

7. The method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain as described in claim 1, characterized in that, The incentive distribution contract automatically calculates and settles the incentive share due to the dynamic digital asset owner from the total revenue of the data sharing task based on the calculated comprehensive data value, including: The data requester pre-deposits the total revenue from the data sharing task into the incentive distribution contract in the form of digital assets; After receiving the total data value of all dynamic digital assets participating in the task, the incentive distribution contract calculates the incentive share of each dynamic digital asset according to the proportion of the value of each dynamic digital asset to the total value, and executes automatic transfer and settlement to the address of the owner of each dynamic digital asset.

8. The method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain as described in claim 1 or 2, characterized in that, After the method is executed, it also includes: Automatically triggered by the smart contract, the contribution or incentive share calculated in this data sharing task is used to update the historical contribution value in the dynamic attributes of the dynamic digital asset, and to update the reputation score recorded by the data provider in the identity contract.

9. The method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain as described in claim 1, characterized in that, The process of encapsulating multi-source heterogeneous sensing data into standardized data packets includes: The raw sensory data is cleaned, desensitized, and formatted, and then encapsulated according to a unified data metadata standard, which predefines the structure of the static attributes, quality indicators, and dynamic attributes of the data.

10. The method for confirming ownership and sharing privacy of multi-source heterogeneous sensing data based on blockchain as described in claim 3, characterized in that, The weighting coefficients of the various indicators used in the weighted aggregation are determined by the data sharing alliance members through voting in the on-chain governance contract and recorded in the value assessment smart contract.

Images