12results about How to "Avoid data breaches" patented technology

The application discloses a kind of concealed communication detection methods based on federal learning, including the following steps: the sample data transmitted by several time slots antennas is stored in the form of matrix;Determine global model, center server initializes global model;Center server distributes initialized global model to user;Participating node downloads global model from center server, and uses local data to train and update local model with downloaded global model as local model;Participating node uploads local model to center server, and center server receives all local models and carries out model aggregation update, forms new global model;Center server distributes new global model to all training nodes, and carries out iterative training;User detects local data with the final updated global model, judges whether concealed communication data is contained in physical layer signal, and completes classification detection task.The application saves communication resources and protects data security.

This invention discloses a method, system, and computer equipment for electronic data forensics in police vehicles. The method includes: confirming whether the vehicle's onboard data storage device has read permissions; if read permissions exist, the investigating officer performs bidirectional authentication between the vehicle-mounted forensics equipment and the identity authentication module in the public security platform, activating and authorizing the secure, legal, and operable nature of the vehicle-mounted forensics equipment; if read permissions do not exist, data is directly read from the vehicle-mounted data storage device; and different data integrity checks and extractions are performed based on the two communication protocol types of the vehicle-mounted data storage device. This invention, in police enforcement scenarios, overcomes permission barriers and equipment damage constraints, enabling rapid acquisition and analysis of onboard data.

The application discloses a kind of multi-party secure computation methods for data asset audit into table, it is related to multi-party secure computation and data verification technical field, the application includes: step 1.log block and hash generation, step 2.confidential verification and cross window deduplication, step 3.abnormal identification and audit determination, the method carries out standardization format regulation and time sequence window division to e-commerce user behavior log, realizes the credible retention of each record by hash operation, guarantee original data complete traceable, complete data quality evaluation in combination with window effective matching rate, according to the evaluation result, execute dynamic window adjustment and extension recalculation, generate the user total number ciphertext after cross window deduplication, abnormal traffic window is screened by multidimensional behavior characteristics comparison, the whole process of whole set of procedures uses the mode of cryptogram operation, fully protect various data information, effectively reduce audit misjudgment probability, improve overall detection precision, adapt to the audit work of e-commerce field data asset compliance into table.

The utility model relates to the field of protection of real person verification equipment, in particular to a protection structure of intelligent construction site real person verification equipment, which comprises a verification equipment body. A protective shell is arranged on the outer surface of the verification equipment body, a movable door is movably connected to the rear end of the protective shell, a plurality of interfaces are formed in the rear end of the verification equipment body, the partition plate is arranged at the rear position in the verification equipment body, and a plurality of grooves are formed in the bottom of the partition plate; after the coded lock is unlocked, the movable cover is opened, the movable door is opened at the same time, then the first rotating block is rotated to drive the partition plate to move upwards, a needed connecting wire is inserted into the interface, then the partition plate moves downwards to cover the interface, and meanwhile the thin part of the connecting wire is inserted into the groove through the silica gel block. The problem that the information security and the operation stability of the whole construction site are threatened due to data leakage or unauthorized access of a system which is possibly caused by illegal connection due to the fact that an interface is easy to become a security hole is solved.

The application provides a query statement processing method, an electronic device and a computer readable storage medium. The method comprises the following steps: auditing a received query statement based on a preset rule in a trusted container to obtain auditing information of the query statement; and in response to the auditing information meeting a preset condition, sending query related data to a privacy computing platform, wherein the query related data comprises the query statement and remote proof material of the trusted container. By using the embodiment of the application, the security of executing the query statement can be improved, and data leakage can be avoided.

This invention relates to the field of multi-tenant cloud service technology and discloses a multi-tenant resource dynamic access method, cloud hosting system, and apparatus based on the MCP protocol, including: a client system, an intelligent agent system, an MCP routing service system, and an MCP service group system; the client system is used to initiate service requests to the intelligent agent system, receive the endpoint address and short token returned by the MCP routing service system, and establish and maintain an SSE long connection with the MCP service; the intelligent agent system is used to determine whether the service request needs to call an MCP tool, initiate an MCP tool call request through the SSE long connection, and process the returned result; the MCP routing service system serves as a central scheduling component. This invention uses tenant group dynamic routing, directing each tenant's MCP tool call to a dedicated endpoint, avoiding the mixing of tool interfaces from different tenants. GPT-4o testing verified that the MCP tool identification accuracy is improved from 60% in the prior art to over 90%, an improvement of 30%.

This application provides a base station parameter calibration method and electronic device. The method includes: acquiring base station parameters of a base station; sending a request to a real-time dynamic carrier phase differential (RTK) proxy server to perform differential measurement of the base station parameters, and acquiring differential correction values ​​of the base station parameters from the RTK proxy server; and sending the differential correction values ​​to the base station to perform base station parameter calibration. Therefore, this application can solve the problem in related technologies where collecting, calculating, and processing base station parameters at the base station side increases the security risks of data leakage and tampering, thereby improving the security and accuracy of base station parameter calibration.

Embodiments of the present application provide an audio playing control method and device, equipment, system and storage medium, wherein the audio playing control method based on a sound box cloud server comprises: searching target audio metadata in a cloud sound box private music library in response to an on-demand instruction of a target audio, the cloud sound box private music library storing local audio metadata, the local audio metadata comprising an audio name, an audio ID and a local playing address of the audio; and performing local playing according to the searched target audio metadata, the local playing comprising sending an audio playing instruction to a sound box, the audio playing instruction comprising the local playing address of the target audio. The technical solution provided by the present application does not need to upload local audio of a local device to a cloud database, avoids occupying a large amount of cloud space, and avoids the problem of data leakage.

The invention discloses a novel size detector control method, which comprises a remote data management server module, a detection terminal gateway module, a size detection execution module and a user operation module, and is characterized in that a safety communication service platform is arranged between the remote data management server module and the detection terminal gateway module; the security communication service platform comprises an application layer, a platform layer, a transmission layer and a communication layer, the application layer provides a detection security plug-in SSDK and a detection task management client, the platform layer is provided with capacity interfaces of detection flow management, data synchronization, task scheduling and the like, the transmission layer is based on the P2P technology, the encryption transmission and routing optimization capacity of detection data is achieved, and the communication layer is used for communication with the detection security plug-in SSDK and the detection task management client. And the communication layer comprises a three-network direct connection data channel and an industrial-grade data private line communication service. According to the invention, through fusion design of a secure communication architecture, intelligent voice interaction, workpiece-face dual verification and a high-precision feature extraction technology, security, intelligence, precision and standardization of a detection process are realized.

This invention provides an application programming interface (API) interception method and system. It obtains a preset interception policy and synchronizes the policy to the user space of the API interceptor. If the interception policy is blocking, the policy in the user space is synchronized to the kernel space of the API interceptor via a BPF map, and blocking is performed in the kernel space. If the interception policy is tampering, the message content is tampered with and sent in the user space using an AF_XDP socket. The interception policy and content can be customized according to actual needs, flexibly addressing various API interception scenarios, improving the security and compliance of API calls, and enhancing accuracy and speed.

The utility model provides an isolated transmission terminal based on industrial network interconnection, including terminal shell, the transmission mechanism is equipped with in terminal shell inner bottom wall, and terminal shell inner bottom wall one side is equipped with moving mechanism, and terminal shell inner bottom wall other side is equipped with limit mechanism, and fixed mounting of fixed plate has in transmission terminal shell one end, the utility model discloses through fingerprint sensor carries out the right verification, only the fingerprint of authorized personnel is identified, motor will start, drives screw rod rotation, and then drives the mobile block and the moving plate to move down, makes the terminal machine to show in the user's face, and the user operation is convenient, has guaranteed that the terminal machine will not be contacted at random under the unauthorization condition, effectively prevents data disclosure and illegal operation, and the user can carry out data transmission operation fast, conveniently after through fingerprint verification, realizes efficient data exchange through data transmission mouth, has improved work efficiency greatly.

The invention relates to the technical field of talent evaluation, in particular to a multi-modal talent evaluation data intelligent analysis and safety operation and maintenance system and method based on a large model, and the system comprises a multi-modal data collection layer which is provided with a sensing device and an interaction terminal and is used for collecting multi-modal original data; the data preprocessing layer is used for performing standardized conversion, noise filtering and feature extraction processing on the multi-modal original data to obtain structured feature data; the large model analysis layer is deployed with a pre-training multi-modal large model and is used for executing evaluation dimension reasoning based on the structured feature data and outputting an evaluation analysis result; and the result output layer is used for outputting the evaluation analysis result in a visual form or a structured report form. The data preprocessing layer is used for effectively processing data, intelligent and accurate evaluation analysis is achieved by means of the large model analysis layer, a user can conveniently check an evaluation result through the result output layer, and meanwhile the safety of the system in the whole operation process is guaranteed through the safety operation and maintenance layer.