12results about How to "Guaranteed privacy and security" patented technology

The invention discloses a privacy-enhanced access control method for medical data sharing, which is characterized in that an efficient and privacy-protected attribute revocation mechanism is innovatively added on the basis of an original decentralized multi-authority attribute verification and strategy hiding mechanism, and an attribute revocation list (ARL) based on a Merkle tree and corresponding ZKP constraints are introduced, so that the privacy-protected access control method for medical data sharing is realized. Fine-grained dynamic authorization is supported, the security problem that user permission changes along with time in a medical environment is effectively solved, and triple privacy security of user attributes, access strategies and attribute validity states is ensured. Through cooperative work of a system initialization stage, a strategy deployment stage, an access request submission stage, an access control decision stage and a data access stage, strong privacy protection, decentralized architecture, fine-grained control and dynamic authorization in medical data sharing are realized; the problems of single-point failure, privacy disclosure and strategy stiffness in a traditional scheme are effectively solved, and a safe and reliable technical basis is provided for cross-institution medical cooperation.

The application discloses a cloud-edge collaborative and federated learning-based electroencephalogram prediction model self-evolution system and method. The system comprises an edge device and a cloud server cluster: the edge device collects physiological signals of a user and runs a first machine learning model, and uploads desensitized data of a marked event asynchronously; the cloud server performs continuous wavelet transform on received one-dimensional electroencephalogram signals to generate two-dimensional time-frequency images, trains a second deep learning model by using the images, generates updated weights, and completes silent updating by issuing the updated weights to the edge device. The application solves the model concept drift problem through a cloud-edge collaborative architecture, realizes continuous evolution of a prediction model, improves feature extraction depth and generalization ability through time-frequency conversion and a multi-specialist sub-network architecture, and effectively protects user privacy through a data desensitization and asynchronous uploading mechanism.

This invention discloses a multi-agent air-to-ground network resource allocation method based on federated learning. In the air-to-ground network, the ground network constitutes high-data-rate V2I links; the air network constitutes V2U links for direct communication with ground vehicles; the V2U links share the spectrum resources of the V2I links and use hybrid spectrum access technology for transmission; a network resource allocation system model consisting of M pairs of V2I links and K pairs of V2U links is constructed; a multi-agent resource allocation method is adopted, and a deep reinforcement learning model is constructed with the goal of minimizing the total transmission delay of the V2I link channels; federated learning is used to optimize the deep reinforcement learning model; during the execution phase, the V2U links obtain their current state based on observations, and the optimal resource allocation strategy is obtained using the trained model. This invention exhibits good stability in highly dynamic air-to-ground networks.

The invention relates to the technical field of artificial intelligence, can be applied to business system platforms of financial science and technology, medical health and the like, and discloses a product service strategy analysis method, device, equipment and medium, and the method comprises the steps: recognizing a business stage corresponding to a target user according to an interaction record, and determining a service stage threshold value of the target user; identifying an interaction intention of the target user according to the multi-modal interaction features; identifying context data corresponding to the multi-modal interaction features, and analyzing an intention perception weight of the target user according to the interaction intention and the context data; performing parameter optimization on the local model, and adjusting service strategy parameters in the optimized local model according to the intention perception weight; and analyzing a product promotion index of the target user by using a local model corresponding to the adjusted service strategy parameter, generating a product decision instruction according to the service stage threshold and the product promotion index, and determining a service strategy of the target product according to the product decision instruction. And the accuracy of product service strategy analysis is improved.

This invention discloses an end-to-end verification method, system, and product for video streams based on Schnorr signatures. The sending end first encodes the audio and video streams into a continuous sequence. For each encoded transmission packet, a point in the group is calculated using random numbers as the random commitment point for the frame signature, where is the generator of the selected elliptic curve group, and the group order is a prime number. Then, a signature scalar is calculated according to the Schnorr signature algorithm. Next, the original video frame data, the random point, and the signature scalar are combined into an authentication data packet, which is pushed to the server and client in real time via a network protocol. Finally, the client performs instant authentication and cache management, while the server performs frame-by-frame verification and aggregated storage. This invention utilizes the linear aggregation characteristics of Schnorr signatures to achieve efficient authentication and traceability of real-time video streams and historical stored data while significantly reducing storage space requirements, ensuring the integrity and evidentiary value of the video data.

The application discloses a principal component analysis multi-bias interaction-based longitudinal federal learning optimization method, an electronic device and a storage medium, and belongs to the technical field of privacy calculation. In order to improve the efficiency of a longitudinal federal neural network model in the case that the capacity of a data set is small, the application trains participating parties including a training initiator and a training assistant, trains data of the training initiator and data of the training assistant by using a forward propagation method, obtains a training result of the forward propagation method, trains the training result of the forward propagation method by using a backward propagation method, and the training assistant and the training initiator respectively update model parameters, so that one round of training is completed. The training is repeated until the training result reaches a precision requirement or a stop condition, and neural network training in a longitudinal federal scene based on multi-bias interaction is completed. The principal component analysis data dimension reduction method is used, so that the filtering function of features is guided by multi-party data information, and the result is more persuasive.

This invention discloses a method, apparatus, device, and medium for updating a fundus image recognition model, relating to the field of ophthalmology. The method includes: determining a set of target features for the fundus image based on the acquired knowledge of fundus image recognition model samples corresponding to the user terminal to be updated; acquiring the activation values ​​of neurons in the recording network corresponding to each user terminal to be updated; constructing a local fundus image recognition model subgraph based on the fundus image recognition model sample knowledge, activation values, and a set of common features of the fundus image; transmitting all local fundus image recognition model subgraphs to a server; performing subgraph fusion on all the local fundus image recognition model subgraphs to obtain a fused subgraph; and updating the local fundus image recognition model using the fused subgraph. This invention transforms the fusion and updating of the user terminal model into an operation of subgraph fusion to generate a full image, ensuring privacy and security throughout the model fusion cycle based on multiple privacy computing technologies, and achieving secure model transmission and incremental fusion.

This invention discloses a heterogeneous federated learning method based on reinforcement learning models and joint optimization algorithms. This method, used in scenarios with given heterogeneous data client pricing and computing power, calculates the contribution of client-side weights to the global model through reinforcement learning. Combined with a dynamic programming algorithm that considers both total time and contribution, it achieves shorter model training time and better final model performance. The method includes the following steps: training a machine learning model using federated learning; the client-side sends model weights back to the server; the server calculates the contribution of the client-side models to the global model using the reinforcement learning model, records the model's performance on the client-side test set and the training time of each model; then, based on a hierarchical dynamic programming algorithm, selects a client set that minimizes the objective function containing both total contribution and the longest training time.

The application relates to the technical field of mobile crowd sensing, and discloses a mobile crowd sensing control method, wherein when sensing data is transmitted between participants and a processing platform, the sensing data is encrypted and transmitted by using a public key generated by an RSA encryption algorithm, and the encrypted sensing data needs to be decrypted by using a private key which is not publicly disclosed, so that even if the sensing data is intercepted, the sensing data cannot be decrypted, the reliability of sensing data encryption can be improved, the privacy of a user can be prevented from being leaked in the sensing data transmission process, and the safety of the privacy of the user is ensured. The application further discloses a mobile crowd sensing control device and system and a storage medium.

The application discloses a kind of regional power grid cooperation methods based on knowledge graph driving, comprising the following steps: collecting regional power grid data, generating multi-source heterogeneous power grid data set;Construct regional power grid knowledge graph, initialize federation heterogeneous graph neural network and graph neural controlled differential equation joint framework;Perform federation heterogeneous graph neural network coding, generate local embedding, mode weight decoupling parameter and local coding parameter;Run graph neural controlled differential equation modeling, generate continuous time state trajectory and dynamics parameter set;Perform federation aggregation on client set results, generate global model parameters;Issue global model parameters and update local model;Based on knowledge graph data, control path input and updated model parameters, generate regional power grid cooperation result set.The application significantly improves the cross-regional power grid cooperation prediction accuracy and calculation efficiency, effectively guarantees the operation safety and dispatching reliability.

The application discloses a kind of location privacy protection method and system based on homomorphic encryption, belong to data processing technical field;The method in which client carries out homomorphic encryption for location data as digital matrix ciphertext;Digital matrix ciphertext is passed to server by byte stream;The ciphertext obtained by server is matched with matrix in database;The result matched by database is sent to client.The application combines data information with homomorphic encryption, adopts the mode of matrix, directly calculates and matches on encrypted data without decrypting data, can securely compare and does not disclose any information, fully protects the privacy information of client and server, achieves that private data is available and invisible;Server does not need to contact plaintext data, greatly improves the security of user location privacy data.