732 results about "Homomorphic encryption" patented technology

The present invention discloses a system and method for secure artificial intelligence-based financial technology governance and risk management, designed to provide real-time, autonomous, and verifiable compliance assurance within digital financial ecosystems. The invention integrates a secure artificial intelligence processing unit, a governance control processor, a cryptographically anchored storage unit, a federated learning coordination processor, and a quantum-resistant communication interface enclosed within a tamper-proof hardware structure. The system performs encrypted machine learning computations on financial transaction data using homomorphic encryption and trusted execution environments to preserve confidentiality during analysis. It computes a governance risk index based on probabilistic inference and anomaly detection to identify regulatory deviations, applies adaptive compliance reasoning across multi-jurisdictional frameworks, and automatically enforces governance actions through secure decision logic.

The invention relates to the technical field of information security management, and discloses an information security management method and system based on digital medical treatment. The method comprises the following steps: verifying doctor professional qualification, patient authorization range, data sensitivity level and time permission validity in a multi-institution medical data access request, and generating a permission verification result; analyzing ontology semantic attributes and sensitivity characteristics of medical search keywords input by a user, and determining a corresponding hierarchical encryption strategy based on the permission verification result; performing homomorphic encryption transformation on the medical term synonym set according to the hierarchical encryption strategy, and constructing a medical ciphertext database; and receiving a medical query request, executing symptom disease association matching and ciphertext state logical reasoning operation in the medical ciphertext database, and outputting an encrypted state search result. According to the method, the privacy leakage risk caused by search behavior trajectory analysis is effectively prevented, and the security and availability of medical data are improved.

A latent transformer architecture with latent attention mechanisms and expert processing systems for federated deep learning is disclosed. The system operates entirely within latent space, eliminating traditional embedding and positional encoding layers while maintaining full attention capabilities. Input data is compressed into latent vectors via variational autoencoder encoding, then processed by a latent attention module that computes query, key, and value matrices directly from latent representations. The architecture incorporates expert processing systems including gated latent expert networks for sparse computation and latent mixture of experts for collaborative processing. In the gated approach, a routing network selectively activates specialized expert modules based on latent vector characteristics. The mixture approach enables all experts to contribute through weighted combination, facilitating distributed computation and enhanced model expressiveness.

A collaborative transformation matrix learning system extends adaptive compression and encryption architectures through federated, privacy-preserving optimization. Each node analyzes local data distributions to generate anonymized distribution profiles using differential-privacy mechanisms, securely exchanging profiles and validated transformation matrices across a collaborative network. A trust and validation engine verifies mathematical properties and evaluates claimed performance metrics. Validated matrices are integrated into local optimization when trust and performance thresholds are satisfied. The system employs secure multi-party computation, homomorphic encryption, and conflict-resolution logic to ensure integrity of shared insights while preventing exposure of sensitive information. By combining collective learning with local adaptation, the invention accelerates convergence to optimal matrix configurations, mitigates cold-start inefficiencies, and improves compression-encryption efficiency and cryptographic strength across distributed deployments.

The invention discloses a real-time cleaning and aligning method for multi-source heterogeneous data, and particularly relates to the technical field of source heterogeneous data, a self-adaptive interface adapter is compatible with multiple types of data, and a semantic index atlas is generated through four-dimensional classification labeling; constructing a cloud-edge collaborative streaming processing framework, preprocessing edge nodes, and performing accurate cloud alignment; a dynamic rule cleaning and semantic-dimension-entity three-layer progressive alignment mechanism is adopted, and a four-dimensional quality evaluation system is combined for real-time monitoring; and through closed-loop optimization, homomorphic encryption, a block chain and other security mechanisms, data security and traceability are ensured. According to the real-time cleaning and aligning method for the multi-source heterogeneous data, the real-time performance and accuracy of data processing are effectively improved, the method is adaptive to multiple service scenes, and high-quality data support is provided for data value mining.

The invention provides a medical data-oriented efficient privacy computing fusion engine method and system, and the method comprises the steps: constructing a medical data encryption and decryption system as a basis, and carrying out the standardized packaging of federated learning, homomorphic encryption, secure multi-party computing and other technologies, a unified technical framework comprising a data encryption interface, a ciphertext calculation interface and a result decryption interface is defined; an intelligent task analysis mechanism is adopted, high-level medical analysis requirements can be automatically decomposed into privacy calculation subtask sequences, and multi-technology collaboration is achieved through an execution engine; meanwhile, the functions of resource monitoring and performance analysis are achieved, technology switching and parameter adjustment and optimization during operation are supported, and a self-optimization execution environment is formed. And finally, displaying through a friendly application interface and visualization. According to the method, the problems of single technical route, low calculation efficiency, poor resource utilization rate, serious memory limitation and the like in medical data privacy protection are solved.

The invention discloses a homomorphic encryption-based cross-industry health insurance data sharing and privacy protection method, and relates to the technical field of health insurance data security, and the method specifically comprises the steps: standardizing medical and insurance industry core data, quantifying key indexes such as a disease clinical risk value and an occupational basic risk coefficient, achieving the format unification of cross-industry data, and achieving the privacy protection of the cross-industry data; further generating ownership, risk and business labels, carrying out encryption processing on data through homomorphic encryption on the risk labels based on three scenes of data sensitivity and business influence grading, business label binding pricing, underwriting and claim settlement prediction, adding an integrity check code to high-risk data, screening ciphertexts based on the labels, and then executing fusion calculation; the comprehensive risk adjustment coefficient ciphertext is generated, the data integrity is verified, the comprehensive risk adjustment coefficient is obtained after collaborative decryption, the comprehensive risk adjustment coefficient is applied to premium pricing, underwriting conclusion generation and claim risk pre-judgment, and the scientificity and accuracy of health insurance pricing and underwriting decision making are improved.

The invention discloses a method and system for privacy data protection in a big data analysis process, and relates to the technical field of civil aviation data security and privacy protection, and the method comprises the steps: obtaining civil aviation multi-source heterogeneous data, extracting passenger identity features, travel track features and consumption behavior features, and generating a structured feature matrix; extracting an association relationship between fields through a convolutional neural network, calculating a field sensitivity initial value, generating a field sensitivity weight in combination with a field uniqueness index and the sensitivity initial value, and generating a privacy level mapping table containing the field sensitivity weight according to the sensitivity weight; calculating query sensitivity and a privacy budget parameter, generating a differential privacy noise parameter and a homomorphic encryption key, and generating a hierarchical protection strategy; performing noise injection and ciphertext conversion on the original data to generate a desensitized data set; passenger flow statistics is executed through secret state calculation, and a public analysis result is generated. Over protection of low-sensitivity data is avoided, and sufficient safety of high-sensitivity data is ensured.

The invention relates to the technical field of target behavior prediction systems based on causal inference and multi-task learning, and particularly discloses a target behavior prediction system based on causal inference and multi-task learning. The system comprises a central coordination server and a plurality of participant clients, constructs a global causal graph through a federated causal discovery algorithm in a collaborative manner, determines a causal feature subset of each prediction task, and carries out multi-task model training through a federated average algorithm under the constraint. In the process, differential privacy and homomorphic encryption technologies are comprehensively applied to protect data privacy. According to the method, more accurate causal discovery and more reliable prediction model training can be realized on the premise of protecting data privacy of all parties.

The present invention relates to a privacy-preserving EV charging authorization and billing system, and a method for the same. The proposed system is configured to integrate permissioned blockchain with fully homomorphic encryption. The present invention aims to eliminate plaintext exposure mitigates single point of failure, by performing all authorization and billing computations on encrypted data and recoding transactions immutably, wherein an EV user securely generates encrypted authorization and billing requests using FHE-based public keys. The charging station routes these encrypted requests to the blockchain network, which records immutable encrypted transactions and verifies them via consensus. The FHE computation layer performs secure operations on the encrypted data for authorization and billing, while smart contracts execute automated verification and billing computations, ensuring transparency and auditability.

The invention discloses a digital fingerprint generation method for terminal equipment, and relates to the technical field of digital fingerprint generation, comprising the following steps: S1, collecting multi-dimensional hierarchical features; s2, privacy enhancement; s3, fingerprint anti-regeneration generation is carried out; and S4, fingerprint life cycle management. According to the terminal equipment digital fingerprint generation method, multi-dimensional fine-grained feature acquisition is realized from a non-sensitive layer, a semi-sensitive layer and a high-sensitive layer, a feature weight is optimized through a federated learning framework, fingerprint drift caused by hardware replacement and software updating is effectively resisted, and the user experience is improved. Dynamic Gaussian noise is injected into numerical semi-sensitive features to balance privacy protection and feature effectiveness, 2048-bit key Paillier homomorphic encryption is adopted for high-sensitive features, meanwhile, cross-scene tracking risks are blocked through scenarized sub-fingerprint generation and an original feature immediate destruction mechanism, and the privacy protection and feature effectiveness are improved. And an anti-duplication architecture fusing a Merkle tree and a bloom filter is introduced to greatly reduce the fingerprint coincidence collision rate.

The invention discloses an engineering vehicle driver behavior analysis system based on federal learning. The system comprises a multi-source sensing layer which forms vehicle-road-person collaborative all-weather scenarized data input; the edge calculation layer is used for capturing a complex space relationship of a driving scene, embedding a lightweight cross-modal attention module, calculating an attention weight, and processing long-sequence driving operation data by adopting a Mamba state space model to capture a long-term dependency relationship of a driver state; the federated learning layer adopts a pFedMe algorithm to realize a driver customized model, during local training, the model learns local data distribution characteristics through a Moreau Envelopes regularization item, the cloud aggregates distribution of adaptive local data of each driver, models a federated learning process as a meta-learning problem, and performs training updating; a privacy and security layer; and differential privacy dynamic adjustment, Paillier improved homomorphic encryption and a Hyperledger Fabric block chain are adopted to prevent a log from being tampered.

The invention discloses a medical record large model privacy enhancement fine tuning method based on a federated learning framework. The method comprises the following steps: S01, constructing a three-level medical exclusive federated learning framework of a client, an edge node and a federated server; s02, the client performs privacy enhancement preprocessing on the medical record, generates a medical exclusive feature vector and adds differential privacy noise; s03, a three-level privacy enhancement mechanism of a data layer, a transmission layer and a training layer is designed, dynamic differential privacy noise injection is adopted in the data layer, homomorphic encryption and secure aggregation are adopted in the transmission layer, and gradient mask and federated distillation technologies are adopted in the training layer; s04, adopting a federal fine tuning strategy adaptive to medical characteristics; and S05, constructing privacy security verification, clinical compliance auditing and a model iterative optimization closed loop. Therefore, the privacy protection intensity is obviously improved, the medical data adaptability is optimized, the heterogeneous deployment flexibility is enhanced, the privacy and performance are dynamically balanced, the clinical compliance is guaranteed, and the cooperation efficiency is improved.

The invention discloses a file digital circulation privacy protection method and system based on a zero-trust architecture, and relates to the technical field of data security, and the method comprises the steps: building a mapping table of semantics and encryption fragments based on a semantic context tag set and a dynamic authorization intention in combination with the operation granularity in a homomorphic encryption domain, generating a homomorphic operation plan and a re-encryption credential set; according to the homomorphic operation plan and the re-encryption credential set, executing logic fragmentation and homomorphic encryption on the to-be-transferred file to form a cryptographic container set, obtaining an access commitment instance, and constructing an access commitment chain initial fragment; homomorphic cooperation operation is executed in the secret state container set, the access commitment chain initial fragment is checked in real time, node revocation and authorization reversal are executed according to the commitment chain state, and an updated access commitment chain and an instant trust reconstruction record are generated. According to the invention, the security, controllability and traceability of the file collaboration process are improved.

The invention discloses a data security storage system and method suitable for a resource-constrained network node, and relates to the technical field of data security storage, and the method comprises the steps: carrying out the homomorphic encryption of to-be-transmitted data through employing an elliptic curve public key, packaging a homomorphic ciphertext pair through employing a session key, obtaining a transmission message, and transmitting the transmission message to a network node; transmitting the transmission message to a relay node; determining a target storage node by calculating the distance between the storage node and the aggregated ciphertext pair, performing secondary packaging on the aggregated ciphertext pair by using a storage key, performing disk falling on the target storage node, generating a Verkle proof, and establishing a data storage index; and the secondary ciphertext load is unpacked by using the storage private key, and decryption is performed by using the elliptic curve private key to obtain plaintext data. According to the method, distributed mapping is carried out, secondary packaging is carried out through a storage key derived by homology, lightweight verifiable disk falling and multi-copy redundancy are completed in combination with Verkle certification, and single-node indexing and certification expenses are reduced.

The invention provides a controllable-precision lattice-based homomorphic encryption inner product data similarity retrieval method and system, and relates to the technical field of privacy calculation and security retrieval. The retrieval terminal performs polynomial coding on the retrieval vector to obtain a retrieval plaintext polynomial; performing lattice-based homomorphic encryption on the retrieval plaintext polynomial to obtain a retrieval ciphertext; the server performs homomorphic multiplication on the retrieval ciphertext and the data ciphertext to obtain a discrimination ciphertext; the retrieval terminal decrypts the discrimination ciphertext; determining whether retrieval matching succeeds or not based on the constant term coefficient of the decryption polynomial; the data ciphertext is obtained based on lattice-based homomorphic encryption of a data plaintext polynomial; the data plaintext polynomial is obtained by encoding a polynomial of the data feature vector based on a similarity threshold; the constant term coefficient of the product of the data plaintext polynomial and the retrieval plaintext polynomial is equal to the difference between the inner product of the data feature vector and the retrieval vector and the similarity threshold. The problem that traditional searchable encryption only supports accurate matching and easily leaks query intentions is solved.

The invention relates to the field of artificial intelligence and federated learning, and discloses a supply chain data security sharing method based on federated learning. According to the method, a decentralized multi-participant collaborative architecture is constructed, local differential privacy disturbance is applied to a local gradient in each round of iteration, and a threshold secret sharing and Paillier homomorphic encryption dual mechanism is adopted to segment, encrypt and transmit a gradient share; and each participant reconstructs a global gradient through linear superposition, and locally completes model updating. The method does not need a trusted third party, effectively prevents leakage of original data and gradient information, gives consideration to model effectiveness and strong privacy protection, and is suitable for a cross-enterprise supply chain intelligent collaboration scene.

A data processing method, comprising: receiving a first fragment among two fragments obtained by performing secret fragmentation on input data (S602); dividing an original calculation function for processing the input data into a non-linear function in a symmetric interval and a linear function in an asymmetric interval (S604); obtaining a first representation vector of the first fragment on the basis of a Fourier series expression of the non-linear function, sending to a second computing service provider a first ciphertext obtained by performing homomorphic encryption on the first representation vector, receiving a blinded ciphertext returned by the second computing service provider on the basis of the first ciphertext and a second fragment among the two fragments, and decrypting the blinded ciphertext to obtain a result fragment of the non-linear function corresponding to the first fragment (S606); and, on the basis of the result fragment of the non-linear function corresponding to the first fragment and a result fragment of the linear function corresponding to the first fragment, obtaining a first result fragment, which is used for, together with a second result fragment obtained by the second computing service provider on the basis of the second fragment, generating a computing result corresponding to the input data (S608).

Identity nodes within bidirectional dual node hashchain (BDNH) bundles generate hash keys by processing partial metadata and linking them through identity graphs. Translation nodes perform identity resolution, data anonymization, and re-keying by transforming original identifiers into anonymous IDs and re-key IDs. Laplace noise enabler capsules introduce Laplace noise into translation layer logs for differential privacy. BDNHs share metadata between adjacent bundles for secure and efficient data transmission. A cognitive analytics layer shared among all bundles verifies hashed encrypted data packets using a gossip protocol for real-time validation. An analytics workspace layer activates re-key IDs and creates digital tags for datasets using BERT transformers, linking data through knowledge graphs. The system manages parallel processing and includes machine learning modules, homomorphic encryption, federated learning, zero-knowledge proofs, attribute-based encryption, and anomaly detection. Bundles' AI engines work with the Laplace noise enablers to analyze data, optimize noise, and enhance the data clean room functionality.

The invention relates to a Boolean function calculation method and device based on fully homomorphic encryption and electronic equipment, and the method comprises the steps: carrying out the format conversion of each received first type of ciphertext based on a target conversion format, and obtaining at least one second type of ciphertext; constructing a multi-layer cascaded CMUX circuit, and based on a preset logic rule, controlling the multi-layer cascaded CMUX circuit to dynamically select pre-encrypted truth table data by taking the second type ciphertext as a control signal to obtain at least one third type ciphertext; and extracting the third type ciphertext coefficient to obtain at least one fourth type ciphertext, and performing blind rotation operation reconstruction on the fourth type ciphertext to obtain at least one encrypted data. Therefore, the problems of low function calculation efficiency, poor expandability and limited calculation precision in the prior art are solved, the resource utilization rate and the parallel processing capability of fully homomorphic encryption in ciphertext calculation are remarkably improved, and meanwhile, the calculation precision and the safety of encrypted data are ensured.

Mechanisms are provided for optimizing a deep learning (DL) computer model for homomorphic encryption (HE) workload processing. The mechanisms receive an original DL computer model architecture that is to be optimized for HE workload processing, and modifying the original DL computer model architecture by replacing a self-attention layer of the original DL computer model with an HE friendly self-attention layer that comprises a Power SoftMax function that does not have exponent terms, to thereby generate a modified DL computer model architecture. The mechanisms execute a machine learning training of the modified DL computer model architecture, approximate one or more elements of the Power SoftMax function with polynomials to generate a trained HE optimized DL computer model, and output the trained HE optimized DL computer model for execution on HE workloads.

The invention belongs to the technical field of data security, and particularly relates to a community data sharing and exchanging security control method based on a block chain, which realizes fine-grained and dynamic access control by performing refined permission division on user characteristics and data attributes and constructing a verifiable attribute permission mapping relationship in an intelligent contract. The slave logic layer prevents unauthorized access vulnerabilities; a mode of combining a fully homomorphic encryption algorithm and a lightweight homomorphic encryption algorithm is adopted, direct operation of data in an encryption state is realized, a permission verification mechanism without data content leakage is constructed in combination with zero-knowledge proof, and dual protection of'calculation without decryption and verification without leakage 'is formed. According to the method, the data leakage risk caused by smart contract code vulnerabilities is fundamentally immunized, an extensive permission allocation mode is broken through by an attribute-based permission management method, refined permission division can be performed according to actual attributes of users and data, permission abuse and unauthorized access are effectively avoided, and the normalization and security of data access are greatly improved.

The invention relates to the technical field of medical information sharing, and discloses a medical information sharing application method and system, and the system comprises a medical information collection module, a data preprocessing module, a security encryption module, a sharing strategy management module, a path optimization module, a monitoring auditing module, a data synchronization module, and a feedback optimization module. When medical information sharing is carried out, standardized medical data are encrypted and anonymized based on privacy protection requirements, and safe medical data are generated in real time by adopting homomorphic encryption and differential privacy technologies, so that whether the privacy disclosure problem occurs in the transmission process of the shared data can be dynamically verified; according to the method, the data is protected in time when being accessed and tampered in an unauthorized mode, the privacy leakage risk of a patient is reduced, a sharing strategy is dynamically matched according to user roles and permission setting information, permission control is automatically executed in combination with an attribute-based encryption technology and an intelligent contract, and the accuracy and controllability of data sharing are guaranteed.

The invention relates to the technical field of block chains, and discloses a block chain data encryption query method, device, equipment and medium, the method comprises the following steps: extracting data content information of to-be-chained data, using a homomorphic encryption algorithm to encrypt the data content information to obtain ciphertext content information; taking the ciphertext content information and the corresponding ciphertext hash value as on-chain evidence storage data to be stored in a block chain in a chaining manner; performing association mapping on the on-chain evidence storage data and the ciphertext access rule through the smart contract to generate an authority association rule corresponding to the on-chain evidence storage data; performing permission verification on the access permission in the data query request based on a permission association rule; after the permission verification is passed, querying the on-chain evidence storage data in a homomorphic encryption state according to the data query request to obtain an encrypted query result; and decrypting the encrypted query result to obtain target plaintext query data. The reliability of data encryption query can be improved.

The invention relates to the cross technical field of computer information security and medical data processing, in particular to a medical data sharing privacy protection system based on a block chain and homomorphic encryption, which comprises a data representation and preprocessing layer, an access control and task execution layer and a result credibility guarantee layer. The preprocessing layer maps original data to a specific mathematical structure through implicit computable coding according to semantic features of medical data and a target computing task, and adapts to standard homomorphic operation; the access control layer defines a fine-grained access strategy and dynamically generates a task binding token in combination with attribute-based encryption and an intelligent contract; and the result guarantee layer adopts a fusion verification label and threshold secret sharing technology to realize lightweight calculation result verification. The system improves the data sharing suitability and the result credibility on the premise of guaranteeing the privacy security of the medical data, and is suitable for a medical data alliance collaborative sharing scene.

The invention discloses a privacy protection Transform reasoning method based on function secret sharing, which is characterized in that homomorphic encryption is adopted to encrypt an input token sequence, nonlinear operation is processed through a secure multi-party computing protocol, a distributed interval function key is generated by using function secret sharing, and secure comparison and dynamic token pruning are realized. The privacy protection reasoning specifically comprises the steps of an initialization stage, an input encryption stage, a cooperative calculation stage, a safety pruning stage, a result recovery stage and the like. Compared with the prior art, the method has the advantages that the online communication complexity of security comparison is reduced to a constant level from a quadratic level through a function secret sharing technology, the communication bottleneck problem of token pruning in the prior art is effectively solved, and the security comparison efficiency is improved on the premise of protecting client input data and server model parameter privacy. The reasoning efficiency of the Transform model in the privacy sensitive scene is remarkably improved, the method is particularly suitable for the scene where Transform model reasoning needs to be carried out on sensitive data and the privacy requirement is high, and the application prospect is good.

A system determines whether to execute a first operation of a distributed machine learning model (MLM) on at least one server or on at least one client device. In response to determining that the first operation should be executed on the at least one server, the system: encrypts data associated with the first operation using a specific encryption scheme; and transmits the encrypted data to the at least one server for execution of the first operation on the encrypted data. In response to determining that the first operation should be executed on the at least one client device, the system performs the first operation on the data using the at least one client device without encrypting using the specific encryption scheme.

The invention discloses a credit risk joint modeling system and method based on federal learning. The system comprises a data preprocessing module, a heterogeneous data adaptation module, a federal learning dynamic modeling module, a security communication module and a risk assessment decision module. The data preprocessing module adopts a hierarchical encryption strategy, sensitive fields are subjected to CKKS homomorphic encryption, super-sensitive fields are bound by adopting TFHE full homomorphic encryption and combining with biological characteristics, and non-sensitive fields are encrypted by adopting SHA-3 Hash. And the heterogeneous data adaptation module realizes cross-mechanism data semantic alignment through a three-level mapping network and adversarial training. And the federated learning dynamic modeling module is used for carrying out local training by using a modified MobileNet-V3 network, and dynamically adjusting an aggregation weight based on Bayesian optimization. And the secure communication module ensures data interaction security based on the block chain and zero-knowledge proof. According to the method, multi-mechanism collaborative modeling without local data is realized, the AUC value of the model is increased from 0.82 to 0.94, and the accuracy of credit risk assessment is effectively improved.

The invention discloses a hybrid multi-mode multi-key fully homomorphic encryption method based on a ring learning error problem. Comprising the following steps of public parameter initialization and data set preparation, multi-participant key generation and evaluation key joint generation, data encryption, ciphertext homomorphic operation, noise evaluation and control, participant partial decryption and server aggregation decryption. According to the method, the security advantage under a ring structure, the efficient homomorphic multiplication capability of the BFV model and the floating-point number calculation mechanism of the CKKS model are combined, and the functions of multi-party key independent generation, public evaluation key joint construction, multi-modal fusion, ciphertext homomorphic multiplication, relinearization, noise evaluation control and bootstrap refresh are supported; the problem that an existing homomorphic encryption system has performance bottlenecks and potential safety hazards in the scenes of multi-participant key isolation, ciphertext multiplication expandability, noise control and dynamic participation is solved, and the method has good safety, expandability and calculation efficiency.

A Zero-Knowledge Inference Sandbox provides a secure computational environment integrating cryptographic zero-knowledge proofs, secure multi­ party computation, and homomorphic encryption to execute i