Abstract: Disclosed is a federated learning method for a k-means clustering algorithm. Horizontal federated learning includes the following steps: 1) initializing K clusters, and distributing, a local sample to a cluster closest to the sample; 2) calculating a new cluster center of the cluster; and 3) if the cluster center changes, then returning to step 1). Vertical federated learning includes the following steps: 1) running, the k-means clustering algorithm locally to obtain T local clusters and intersecting to obtain TL new clusters, or running an AP clustering algorithm to obtain Ti clusters and intersecting to obtain ?i=1LTi new clusters; 2) taking TL (?i=1LTi) new cluster centers as input samples, and initializing the K clusters; 3) distributing each sample to the cluster closest to the sample; 4) calculating a new cluster center of the cluster; and 5) if the cluster center changes, then returning to step 3).

Abstract: A digital currency intra-chain and cross-chain off-chain swap method based on trusted hardware. An initiator sends a swap message containing a public-private key pair to be swapped by a initiator to a responder, and deletes the public-private key pair to be swapped by the initiator; after receiving the swap message, the responder saves and locks the public-private key pair to be swapped by the initiator, locks the public-private key pair to be swapped by the responder, and sends a response message containing the public-private key pair to the initiator; after receiving the response message, the initiator saves the public-private key pair to be swapped by the responder and sends a confirmation message to the responder; after receiving the confirmation message, the responder deletes the public-private key pair to be swapped by the responder and unlocks the public-private key pair to be swapped by the initiator to complete the swap.

Abstract: A method for digital currency intra-chain and cross-chain off-chain payments based on trusted hardware. By creating and transferring a public-private key pair saved and managed by a trusted hardware system to replace the on-chain payment, the trusted hardware ensures the security of the transaction, avoids complicated on-chain transaction steps, reduces the long on-chain transaction time and high on-chain transaction cost, and neglects the differences among blockchain systems to support cross-chain transactions, thus greatly improving the intra-chain and cross-chain off-chain payment efficiency in digital currency and greatly reducing the payment cost in digital currency. In addition, on the basis of the above payment method, the corresponding rescue object generation process of different digital currency types and the process of using the rescue object to retrieve digital currency are further proposed, which avoids the problem of asset loss caused by the damage or loss of trusted hardware.

Abstract: A safety management and control system for full lifecycle of industrial control data, comprising: an administrator terminal holding an administrator public-private key pair and an administrator public key root certificate; a server comprising an encrypted database kernel and an initializer, the encrypted database kernel is located in a trusted execution environment and contains industrial control data; the initializer is configured to initialize the encrypted database kernel in combination with the administrator public key root certificate sent by the administrator terminal when the server is deployed for the first time, and return a self-signed certificate generated by the server to the administrator terminal; and a client that accesses the system through the certificate issued by the administrator terminal, and interacts with the server for industrial control data within the authority configured by the administrator terminal.

Abstract: An identity recognition method, an apparatus, a computer device, and a storage medium are provided. The method includes: obtaining a signal to be identified which includes a millimeter wave signal and an audio signal; performing living feature detection based on the millimeter wave signal and the audio signal, and acquiring a living millimeter wave signal and a living audio signal; performing feature fusion of the living millimeter wave signal and the living audio signal, and acquiring a fusion response diagram of a living voice signal; and performing identity recognition based on the fusion response diagram of the living voice signal.

Abstract: A method and an apparatus for privacy protection biometric authentication and an electronic device. The method comprises the following steps: constructing a corresponding a biological data template according to a biological information data set input by a user when registering; generating a pair of public key and private key by asymmetric cryptography technology; generating encrypted biological data by a secret sharing solution and OKVS technology according to the biological data template and the private key; sending the public key and the encrypted biological data to a server; recovering the private key by OKVS technology according to the biological data input by the user during authentication and the encrypted biological data; constructing a signature according to the recovered private key and the corresponding public key; sending the signature to the server, so that the server verifies the user according to the public key and the signature.

Abstract: The present invention discloses a method for designing an interference noise of speech based on the human speech structure, including the following steps: (1): obtaining a large amount of speech data containing different speakers and different speech contents, extracting voiceprint information, and then building an initial speech data set; (2): for each user, obtaining a small amount of speech data of the user, extracting voiceprint information, and then matching the most similar speech data in the initial speech data set; (3): performing data augmentation on the matched speech data; (4): segmenting the augmented speech data with a phoneme segmentation algorithm to form a vowel data set and a consonant data set; (5): constructing three noise sequences based on the vowel data set and the consonant data set, and performing superimposition to obtain an interference noise; and (6): continuously generating and playing randomly generated interference noise, and continuously injecting the interference noise into rec

Abstract: A batch private information retrieval method and apparatus, including: agreeing on a hash function, calculating a serial number index in an original database, retaining a result of each serial number index for each hash function, and obtaining a first hash bucket; filling an index of a batch query into a second hash bucket of a same length as the first hash bucket by the client, and encoding a filling result by using a constant weight code encoding manner; encrypting an obtained encoding result as a batch query request by the client; comparing the batch query request with the first hash bucket by the server to obtain a corresponding intermediate result, calculating correspondingly the intermediate result and the original database to obtain a corresponding query result ciphertext; and decrypting the query result ciphertext by the client, and thus simultaneously obtaining a query result of the batch query request.

Abstract: A batch private information retrieval method and apparatus, including: agreeing on a hash function, calculating a serial number index in an original database, retaining a result of each serial number index for each hash function, and obtaining a first hash bucket; filling an index of a batch query into a second hash bucket of a same length as the first hash bucket by the client, and encoding a filling result by using a constant weight code encoding manner; encrypting an obtained encoding result as a batch query request by the client; comparing the batch query request with the first hash bucket by the server to obtain a corresponding intermediate result, calculating correspondingly the intermediate result and the original database to obtain a corresponding query result ciphertext; and decrypting the query result ciphertext by the client, and thus simultaneously obtaining a query result of the batch query request.

Abstract: An information retrieval method and apparatus, a computer device, and a storage medium are provided. The method includes receiving a request ciphertext sent by a client, the request ciphertext including an encrypted first code, the first code being generated based on a request entry of the client and a preset encoding algorithm; performing distributed matching on a plurality of candidate codes based on the request ciphertext to determine a second code matching the request ciphertext, the candidate codes being generated based on index entries of a database and the preset encoding algorithm; and sending a request information ciphertext associated with the index entry corresponding to the second code to the client.

Abstract: The disclosure provides a multimodal speech recognition method and system, and a computer-readable storage medium. The method includes calculating a first logarithmic mel-frequency spectral coefficient and a second logarithmic mel-frequency spectral coefficient when a target millimeter-wave signal and a target audio signal both contain speech information corresponding to a target user; inputting the first and the second logarithmic mel-frequency spectral coefficient into a fusion network to determine a target fusion feature, where the fusion network includes at least a calibration module and a mapping module, the calibration module is configured to perform mutual feature calibration on the target audio/millimeter-wave signals, and the mapping module is configured to fuse a calibrated millimeter-wave feature and a calibrated audio feature; and inputting the target fusion feature into a semantic feature network to determine a speech recognition result corresponding to the target user.

Abstract: A multi-party privacy computing method and device based on semi-trusted hardware, wherein the method applied to semi-trusted hardware comprises the following steps: acquiring random number masks and random seeds of all user terminals; generating a garbled circuit seed according to the random seeds; generating a garbled circuit according to a predetermined circuit description and the garbled circuit seed, wherein the garbled circuit comprises garbled tables, wire labels and decoding information; sending the wire labels corresponding to the inputs of all user terminals to a user terminal corresponding to the semi-trusted hardware by using an oblivious transfer protocol; and sending the garbled table and the decoding information to the user terminal corresponding to the semi-trusted hardware, so that the user terminal can compute an output value according to the garbled tables, the decoding information and the wire label corresponding to the inputs of all user terminals.

Abstract: A network device determines a priority sequence of a plurality of features, where the priority sequence of the plurality of features is used for random access resource selection. The network device sends first indication information indicating the priority sequence of the plurality of features to a terminal device.

Abstract: A method and an apparatus for privacy protection biometric authentication and an electronic device. The method comprises the following steps: constructing a corresponding a biological data template according to a biological information data set input by a user when registering; generating a pair of public key and private key by asymmetric cryptography technology; generating encrypted biological data by a secret sharing solution and OKVS technology according to the biological data template and the private key; sending the public key and the encrypted biological data to a server; recovering the private key by OKVS technology according to the biological data input by the user during authentication and the encrypted biological data; constructing a signature according to the recovered private key and the corresponding public key; sending the signature to the server, so that the server verifies the user according to the public key and the signature.

Abstract: A multi-party privacy computing method and device based on semi-trusted hardware, wherein the method applied to semi-trusted hardware comprises the following steps: acquiring random number masks and random seeds of all user terminals; generating a garbled circuit seed according to the random seeds; generating a garbled circuit according to a predetermined circuit description and the garbled circuit seed, wherein the garbled circuit comprises garbled tables, wire labels and decoding information; sending the wire labels corresponding to the inputs of all user terminals to a user terminal corresponding to the semi-trusted hardware by using an oblivious transfer protocol; and sending the garbled table and the decoding information to the user terminal corresponding to the semi-trusted hardware, so that the user terminal can compute an output value according to the garbled tables, the decoding information and the wire label corresponding to the inputs of all user terminals.

Abstract: A method and a device for remote acquisition of correlated pseudo-random numbers based on semi-trusted hardware. When applied to semi-trusted hardware, the method comprises: acquiring a random seed of a sender and a selected number of a receiver; generating a plurality of first correlated pseudo-random numbers and first commitment seeds according to the random seed and a predetermined category of the correlated pseudo-random number; generating a first commitment value and a first open value by a commitment mechanism according to the first correlated pseudo-random numbers and the first commitment seeds; generating a Merkle proof according to the first commitment value and the selected number; sending the first correlated pseudo-random numbers, the first commitment value, the first open value and the Merkle proof corresponding to the selected number to the receiver.

Abstract: A method and a device for remote acquisition of correlated pseudo-random numbers based on semi-trusted hardware. When applied to semi-trusted hardware, the method comprises: acquiring a random seed of a sender and a selected number of a receiver; generating a plurality of first correlated pseudo-random numbers and first commitment seeds according to the random seed and a predetermined category of the correlated pseudo-random number; generating a first commitment value and a first open value by a commitment mechanism according to the first correlated pseudo-random numbers and the first commitment seeds; generating a Merkle proof according to the first commitment value and the selected number; sending the first correlated pseudo-random numbers, the first commitment value, the first open value and the Merkle proof corresponding to the selected number to the receiver.

Abstract: This invention discloses a method for discovering vulnerabilities of operating system access control based on model checking. In this method, security attribute and security specifications of operating system access control module are analyzed to construct the access control model. To discover vulnerabilities in the model, security analysis is performed for access control functionality with theorem proving techniques, and consistency of abstract machine specification and correctness and completeness of the components are verified with model checking tools. This method provides theoretical and technical support for studies in the field of operating system security.

Abstract: An efficient, secure and low-communication vertical federated learning method, includes: all participants select part of features of a held data feature set and a small number of samples of the selected features; the participants add noise satisfying differential privacy to part of samples of the selected features, and then send them to other participants together with data indexes of the selected samples; all participants take the received feature data as a label, take each missing feature as a learning task, and train each model with the feature data originally held in the same data index, respectively; all participants predict the data of the other samples with the trained model to complete the missing feature; the participants jointly train a model through horizontal federated learning. The present disclosure can protect data privacy and provide quantitative support for data privacy protection while efficiently training the model with horizontal federated learning.

Abstract: Disclosed is a database encryption method supporting composable SQL query, which mainly comprises the following steps: (1) a user encrypting and preprocessing data based on the encryption scheme provided by the present disclosure and uploading an encryption result and preprocessed data to a service provider; (2) setting and uploading a SQL query instructions: the user uploads the query instruction to the service provider according to actual needs, and uploads auxiliary parameters for the query instruction at the same time; (3) data query: the service provider performs SQL query according to the query instruction and auxiliary parameters received from the user, saves a calculation result, updates the data and returns a query result to the user.