333 results about "Communication bandwidth" patented technology

The invention discloses a power distribution equipment remote diagnosis method based on edge computing, and particularly relates to the technical field of intelligent monitoring of power equipment, and the method comprises the steps: an edge computing node collects the operation state data of the power distribution equipment in real time; performing diagnosis analysis locally at the node to generate a preliminary diagnosis result and key data; uploading the structured data to a cloud according to a preset strategy, and checking the integrity; and the cloud platform performs association analysis on the multi-node data to identify common anomalies, dynamically optimizes a diagnosis algorithm, automatically triggers alarms and work order distribution in a grading manner, and realizes rapid closed-loop processing in combination with the positions and skills of operation and maintenance personnel. Through cooperation of the edge and the cloud, communication bandwidth occupation is reduced, diagnosis accuracy and real-time performance are improved, fault response time is shortened, and the method is suitable for line-level monitoring and operation and maintenance management of the power distribution network.

The invention discloses an electric power measurement data management system based on cloud edge collaboration and an electric quantity distributed acquisition method, and relates to the technical field of electric power information acquisition and processing. The system comprises an intelligent electric meter terminal layer, an edge computing node layer, a regional collaborative gateway layer and a cloud data center. The intelligent terminal can dynamically adjust the sampling frequency according to the load change rate and the historical volatility, and identifies abnormity through an LSTM-ATT model; the edge nodes realize prediction caching based on Kalman filtering and RMSE judgment; the regional gateway executes dynamic time warping and consistency arbitration of topological weighting; and the cloud end completes digital twinborn modeling and exception evidence storage. According to the method, the real-time performance and accuracy of electric power data acquisition are improved, the communication bandwidth occupation is reduced, the system abnormity sensing capability and the operation stability are enhanced, and the method is suitable for an intelligent power distribution network and a distributed energy metering scene.

The invention relates to the technical field of wireless communication, and relates to a multi-terminal data processing method, a multi-terminal data processing device, communication equipment and a computer readable storage medium. The method comprises the following steps: an on-satellite UPF receives to-be-processed data sent by a terminal through a wireless access network, the to-be-processed data carries a joint task identifier, and the terminal is any one of a plurality of terminals corresponding to the joint task identifier; the satellite UPF sends at least part of data in the to-be-processed data to a ground UPF for processing to obtain ground processing data, and stores other data of the to-be-processed data; and the on-satellite UPF performs cooperative processing on the other data corresponding to the joint task identifier in the storage to obtain on-satellite processing data, and sends the on-satellite processing data to the ground UPF, so that the ground UPF performs forwarding based on the ground processing data and the on-satellite processing data. The data transmission amount can be reduced, the communication bandwidth can be saved, the utilization of resources can be optimized, and the data processing efficiency and timeliness can be improved.

The invention relates to an internet of vehicles edge computing task unloading and resource scheduling optimization method, which comprises the following steps: constructing an internet of vehicles edge computing system model fusing a task jump mechanism and dual-priority scheduling, the system model comprising a road test unit deployment model, a communication model, a computing model and a task jump model; establishing a joint optimization function aiming at minimizing the total time delay and the total energy consumption of the system; decomposing a joint optimization problem into three sub-problems of unloading decision, computing resource allocation and communication strategy selection; a deep reinforcement learning algorithm is adopted to intelligently sense a road environment state, an optimal communication priority strategy combination is dynamically selected, and a task unloading decision and computing resource allocation are collaboratively optimized; and in each scheduling time slot, the edge server allocates communication bandwidth and computing resources to the vehicle tasks according to the selected strategy, and triggers a task preemption and cross-server jump mechanism, thereby realizing maximization of system throughput and minimization of task processing time delay.

A georegistration (a.k.a. georectification) of an image captured by a camera in an aerial vehicle, such as a satellite, is based on identifying multiple features using descriptor sets, and sending to a ground station only the descriptors of the identified features and the associated locations in the captured image, without sending of the captured image itself, thus requiring a low communication bandwidth. Using a database of geosynchronized reference images, the ground station uses the received descriptors sets and the associated image locations to localize the features on a selected geosynchronized reference image from the database, and forms a mapping function that map any locations in the captured image to geographical coordinates on Earth. The mapping may be used to geosynchronize an additional feature identified in the aerial vehicle, or to geo synchronize a region that may be cropped from the captured image and sent to the ground station.

The invention discloses a method of an optical anti-turbulence device for polarization multiplexing spatial optical communication. The mode demultiplexer is connected with the optical fiber; the single-mode optical fiber array is respectively coupled with each spatial intrinsic mode decomposed by the mode demultiplexer; the polarization CBC chip is connected with the other end of the single-mode optical fiber array; the driving control module is connected with the polarization CBC chip; the polarization multiplexing optical communication digital coherent receiver (Rx) is connected with the polarization CBC chip through a single-mode optical fiber; through the arrangement of the polarization CBC chip, the compatibility of a polarization multiplexing technology and a mode multiplexing coherent beam combination technology is realized, the problem of polarization dimension information loss or strong crosstalk caused by the use of a polarization state rotator (PSR) and a polarization maintaining optical fiber in the prior art is solved, modulation information of two orthogonal polarization dimensions is completely reserved, the communication bandwidth is improved, and the communication efficiency is improved. And the communication efficiency is greatly improved.

A dielectric waveguide structure of a semiconductor photonics device is formed prior to formation of the semiconductor photonics components of the semiconductor photonics device. This enables high-temperature processing techniques to be used to form the dielectric waveguide structure without concern for potential damage that might otherwise be caused to the semiconductor photonics components if the dielectric waveguide were to be formed above the semiconductor photonics components. The use of the high-temperature processing techniques may enable low optical loss to be achieved for the dielectric waveguide structure in that the high-temperature processing techniques may be used to achieve a low hydrogen concentration in the dielectric waveguide structure. The low hydrogen concentration in the dielectric waveguide structure enables higher performance to be achieved for the dielectric waveguide structure, including greater operating efficiency and increased communication bandwidth.

The invention provides a method for testing the communication bandwidth of a 10-gigabit network between an FPGA (Field Programmable Gate Array) and a CPU (Central Processing Unit), which is applied to a radar system. The FPGA completes a function of regularly generating data and a function of downlink transmission of the data; the 10-gigabit network exchange board card completes data transmission and sends data to the CPU board card through a 10-gigabit network; and the CPU board card receives the data and calculates the communication bandwidth and the network quality. According to the invention, the limitation of traditional single-port point-to-point transmission or simple multi-port parallel transmission on synchronization, load balancing or protocol overhead is solved, and a solution is provided for how to effectively utilize multiple ports, enable the multiple ports to simultaneously and efficiently work and ensure parallel transmission of data without mutual interference.

The invention discloses a digital in-memory computing architecture compiling and simulating method and a tool chain, and belongs to the technical field of in-memory computing. The method comprises the following steps: acquiring a target deep neural network model and a target in-memory computing architecture configuration parameter adapted to the target deep neural network model, the target in-memory computing architecture configuration parameters comprise at least one of the following parameters: the number of cores, the on-chip communication bandwidth between the cores, the size, the number and the bit width of a storage unit of each core, and the scale of a computing unit of each core; configuring a digital in-memory computing accelerator according to the target in-memory computing architecture configuration parameters; compiling the target deep neural network model into a target optimization instruction sequence based on a digital in-memory calculation accelerator; simulating and executing the target optimization instruction sequence, and performing performance evaluation according to performance indexes in the process of executing the target optimization instruction sequence.

The invention discloses an unmanned aerial vehicle group elastic path planning method and system for resisting an intelligent jammer, which can ensure the accurate synchronization of an interference strategy while realizing the efficient compression and transmission of information and reducing the communication bandwidth demand by an interference party by means of an encoder-decoder communication mechanism, and cooperate with a preferential experience playback technology to realize the flexible path planning of the unmanned aerial vehicle group. Learning is rapidly carried out from past experiences, convergence of an interference strategy is accelerated, efficient cooperative interference is achieved, and a strong threat is formed for a data collection task of the unmanned aerial vehicle group; at the defense party, an attention reviewer module is constructed based on an advanced attention mechanism, information having key influence on unmanned aerial vehicle decision making is accurately focused, the attention weight between intelligent agents is calculated, the weight is reasonably distributed for a state-action joint vector, and a more targeted decision is made in a complex environment; a double-commentator module is introduced, a value function is independently estimated by using two commentator networks, and the reliability and the stability of the algorithm are improved through smooth function fusion output and stable value function estimation.

The invention discloses a transmission type communication system and method based on low-frequency carrier waves and high-frequency modulation. The transmission type communication system and method are suitable for complex medium environments such as deep sea, strata and biological tissues. Aiming at the contradiction that the traditional low-frequency communication bandwidth is insufficient and the high-frequency communication penetrating power is poor, according to the scheme, a low-frequency carrier wave (30Hz-100kHz) and a high-frequency modulation signal (such as OFDM (Orthogonal Frequency Division Multiplexing) and 256QAM (Quadrature Amplitude Modulation)) are fused through a nonlinear signal superposition technology to generate a composite carrier wave signal, long-distance penetration is realized by utilizing the low attenuation characteristic of the low-frequency carrier wave, and meanwhile, the data rate is improved through high-frequency modulation. Application scenes comprise deep sea detection (300-meter high-definition video return), a brain-computer interface (implantable antenna transmission neural signal) and an intelligent power grid (underground pipe network monitoring). Through the physical layer-information layer collaborative design, the physical boundary limitation of traditional communication is broken through, and high penetrability, high bandwidth and strong anti-interference capability are achieved.

The invention provides a chip routing method and device, electronic equipment and a storage medium, and relates to the technical field of chip design and manufacture, and the method comprises the steps: carrying out the first-stage interleaving among a plurality of input and output modules based on a memory address of a current data request, and determining a target input and output module; in the target input and output module, performing second-level interleaving among the plurality of links based on the memory address, and determining a target link; and performing third-level interleaving among the plurality of point-to-point channels of the target link, and determining the point-to-point channel of the current data request. According to the method and the device provided by the invention, a three-level progressive interleaving architecture which performs dynamic calculation completely based on the memory address is constructed, the problem that hardware resource consumption is sharply increased due to chip interconnection scale enlargement is solved, the chip cost and power consumption are remarkably saved, and through a step-by-step refined load balancing mechanism, the load balancing efficiency is greatly improved. The routing efficiency in the chip is improved, and the overall communication bandwidth and system performance of chip interconnection are improved.

The invention relates to an unmanned aerial vehicle cluster area search planning method, device, equipment and medium, and the method comprises a task pre-planning stage and a dynamic re-planning stage: in the task pre-planning stage, a ground control center carries out the grid division and numbering of a target area according to the unit time search capability of each unmanned aerial vehicle, and distributes the target area to each unmanned aerial vehicle; maintaining the distributed task state information in each unmanned aerial vehicle; in the dynamic re-planning stage, each unmanned aerial vehicle carries out autonomous search according to the network and the number, and task state transmission and cluster distributed information maintenance are carried out by adopting a long intermittent communication mode in which period triggering and event triggering are combined. Through combination of ground pre-planning and aerial distributed dynamic adjustment, synchronization and efficient and robust collaborative search of cluster task states are realized, the problem that unmanned aerial vehicle cluster area search excessively depends on communication bandwidth and a center node is effectively solved, the requirements for communication bandwidth and frequency are greatly reduced, and the method is suitable for large-scale unmanned aerial vehicle cluster area search. Therefore, the device can work stably in a severe electromagnetic environment.

The present application relates to the technical field of optical communication. Disclosed are an optical chip and a preparation method therefor, and an electronic device. The preparation method comprises: acquiring a substrate, wherein the substrate comprises a first base and a first device layer that are stacked, and the first device layer comprises a detection portion for preparing a photoelectric detector; forming a first waveguide on the substrate, removing the first base, and bonding a second base, wherein the first device layer and the first waveguide are located on the same side of the second base, and the resistivity of the second base is greater than the resistivity of the first base; and forming a first electrode of an electro-optic modulator on the first waveguide, and forming a second electrode of the photoelectric detector on the detection portion. By means of the preparation method and an optical chip obtained thereby, the process compatibility of the optical chip can be enhanced, the loss of optical signals is reduced, the communication bandwidth of the optical chip is increased, and the communication performance of an electronic device is thus improved.

The invention relates to a multi-agent reinforcement learning training method, system and device driven by an unreal engine and a medium. The method comprises the following steps: acquiring core attribute data of multiple agents, including real-time communication bandwidth, initial communication topology configuration, training stage progress, single agent computing power and cluster overall computing load state; analyzing the real-time communication bandwidth and the initial communication topology configuration to form an optimized topology configuration scheme; in combination with an optimized topology configuration scheme and a training stage progress, generating a cooperative training progress state by quantifying progress differences among intelligent agents; extracting load balance data according to the single agent computing capability and the cluster load state; integrating load balance data, a collaborative training progress state and an optimized topology configuration scheme, and constructing a multi-layer collaborative training framework; and integrated optimization is carried out to generate a cooperative training scheme for guiding multi-agent communication, training and resource allocation, so that the training efficiency is effectively improved, the resource load is balanced, and the cooperative synchronism is enhanced.

The invention discloses an electric vehicle transportation state real-time monitoring method based on multi-sensor fusion, and belongs to the field of electric vehicle transportation state real-time monitoring. A comprehensive state index is obtained through integration of the multi-source data fusion model; acquiring a data transmission priority in the comprehensive state index, and if the data transmission priority is higher than a communication bandwidth threshold, allocating an additional transmission time sequence to obtain a dynamic communication scheduling sequence; carrying out resource allocation on the vehicle-mounted communication system according to the dynamic communication scheduling sequence, and judging whether key safety data such as battery overheating early warning is in the first position of the sequence or not so as to determine a real-time transmission path; and monitoring a communication congestion index in the real-time transmission path, and if the communication congestion index exceeds a threshold value, activating a standby data compression mechanism to obtain a reliable transmission data packet, and sending the reliable transmission data packet to a monitoring center.

The invention provides a switch cabinet partial discharge on-line monitoring method based on cloud-side cooperation, and belongs to the technical field of switch cabinet partial discharge detection. Comprising the steps that an edge computing device performs partial discharge preliminary monitoring by using discrete signals collected by an ultrasonic sensor and a ground wave sensor on a switch cabinet; if the partial discharge event is preliminarily monitored, discrete signals collected by all ultrasonic sensors and ground wave sensors on the corresponding switch cabinet before and after partial discharge are sent to cloud computing equipment; and the cloud computing device carries out partial discharge type identification and partial discharge position positioning according to the received discrete signals. According to the method, a collaborative architecture of edge preliminary monitoring and cloud deep analysis is adopted, and lightweight real-time monitoring is executed on the edge side; according to the invention, an intelligent trigger uploading mechanism based on an event is adopted, and the original sensing data before and after the event occurrence moment are uploaded only after the partial discharge event is preliminarily confirmed at the edge side, so that the communication bandwidth demand and the cloud processing load are greatly reduced.

The embodiment of the invention discloses an automatic rock extraction method and system based on deep learning and Mars rover camera images, and aims to solve the problems that an existing Mars rock segmentation algorithm is insufficient in generalization ability under a complex earth surface background, high in model calculation load and difficult to deploy on satellite-borne edge equipment. The core of the method is that a lightweight encoder-decoder network is constructed, and the network integrates three key modules: a frequency-assisted enhancement Mama module, which accurately captures rock texture and contour by fusing the global sequence modeling capability of Mama and the frequency domain enhancement of wavelet transform; the multi-scale feature intensifier is used for adaptively fusing multi-level features by using a parallel double attention mechanism; and the boundary perception auxiliary branch improves the integrity of the segmented contour through an explicit edge supervision and feature decoupling mechanism. According to the method, the rock extraction precision is remarkably improved, meanwhile, the model complexity and the calculation overhead are greatly reduced, the method is suitable for outer space exploration scenes with limited communication bandwidth and calculation resources such as Mars rovers, and effective balance of high precision and light weight is achieved.

The invention relates to the technical field of optical communication, and discloses an optical chip, a preparation method thereof and electronic equipment. The preparation method comprises the steps that a substrate is obtained, the substrate comprises a first base and a first device layer which are stacked, and the first device layer comprises a detection part used for preparing the photoelectric detector; a first waveguide is formed on the substrate, the first substrate is removed, a second substrate is bonded, the first device layer and the first waveguide are located on the same side of the second substrate, and the resistivity of the second substrate is larger than that of the first substrate; a first electrode of an electro-optical modulator is formed on the first waveguide, and a second electrode of a photodetector is formed on the detection section. According to the preparation method and the obtained optical chip, the process compatibility of the optical chip can be improved, the loss of optical signals is reduced, the communication bandwidth of the optical chip is improved, and the communication performance of electronic equipment is further improved.

The invention discloses a distributed communication method and device, electronic equipment and a storage medium. The method is applied to a distributed computing system comprising a plurality of computing units, the plurality of computing units are in communication connection, and the method comprises the following steps: dividing communication among the plurality of computing units into a plurality of levels based on the size of communication bandwidth among the computing units; for each hierarchy in the plurality of hierarchies: constructing a communication topology of the hierarchy based on hardware interconnection of the computing units within the hierarchy; executing a hierarchical communication process based on the constructed communication topology; wherein the plurality of levels of communication processes are executed in parallel.

The invention provides an underwater robot intelligent thermal management energy-saving method, which comprises the following steps: acquiring temperature data and an operation load peak value of a processor in real time through a heat-sensitive sensor arranged in an underwater robot cluster, generating a thermal state vector group, and storing the thermal state vector group in a distributed data node to obtain an initial thermal state record set; analyzing task queue congestion conditions of other underwater robots based on the task capability limited list through a task transfer mechanism under the support of a cluster communication bandwidth, redistributing affected tasks, generating a task takeover distribution table, and determining a target takeover underwater robot combination; and according to the task takeover distribution table, the navigation trajectory of the target takeover robot is recalculated in combination with the water flow resistance change of the underwater environment and the terrain shielding coefficient, path overlapping conflicts are avoided, and an adjusted navigation trajectory group is obtained.

The invention relates to a backflow cable state online monitoring method and system of a distributed master-slave architecture, and belongs to the technical field of backflow cable monitoring. The monitoring method comprises the following steps: receiving a pre-configured equipment topological relation table; controlling the slave equipment to generate an initial characteristic signal; scanning a current channel state parameter of the backflow cable, and dynamically selecting a communication bandwidth and a modulation mode; modulating the initial characteristic signal, generating a carrier characteristic signal and injecting the carrier characteristic signal into the backflow cable; controlling host equipment to receive the carrier characteristic signal on the backflow cable, and extracting an equipment ID and a current timestamp in the carrier characteristic signal; a first abnormal flag bit and a second abnormal flag bit are generated through comparison, and a comprehensive abnormal decision code is generated through fusion; and querying a logic association relationship in the equipment topological relationship table, determining a fault section identifier, generating a state report and uploading the state report to a remote monitoring platform. According to the invention, intelligent and non-contact real-time monitoring of the connection state of the backflow cable in the high-voltage cable line is realized.

The invention discloses a frequency converter abnormal state detection method based on time series data analysis, which belongs to the technical field of electric power, and comprises the steps of: finally generating a structured abnormal event through acquisition, preprocessing, feature extraction, anomaly detection, anomaly level judgment and trend analysis of data of a multi-source sensor; accurate prediction and early warning of the operation state of a complex system are achieved, the technical problems of high real-time performance, low data redundancy and multi-source feature fusion of abnormal monitoring of the frequency converter are solved, redundant data are reduced, the data volume is reduced, the communication bandwidth pressure is reduced, it is ensured that high-frequency, low-frequency and event data are aligned on the same time axis, and the accuracy of abnormal monitoring of the frequency converter is improved. Abnormal data are transmitted preferentially, abnormal information is guaranteed to be complete and reliable, suspicious anomalies are screened rapidly, local and remote alarms are generated according to the abnormal levels, an abnormal database and a periodic summary report are stored, and operation and maintenance and remote monitoring are facilitated.

The invention discloses an insulating soft sling tension monitoring method and system based on the Internet of Things, relates to the technical field of Internet of Things and power equipment monitoring, and is used for solving the problems of high communication bandwidth occupation and poor real-time performance. Data redundancy is reduced, the transmission efficiency is improved, a cloud server integrates data from different sensors by adopting a multi-source data fusion algorithm to generate a more accurate tension distribution model, a dynamic early warning rule is constructed based on a sliding window statistical analysis method, a threshold interval is optimized through a self-adaptive adjustment strategy, and the accuracy of the tension distribution model is improved. According to the technical scheme, the sensitivity and accuracy of early warning rules are ensured, a visual operation interface is provided, a user can conveniently check data and adjust parameters, the usability and flexibility of the system are further improved, and the requirements of the industrial field for a high-reliability tension monitoring system are met by integrating the technical means.

The invention discloses a control instruction rate matching method of a distributed controller-driver, and relates to the technical field of distributed control, and the method comprises the steps: carrying out instruction frequency detection through employing a clock source of the driver, and obtaining a frequency error value of a controller instruction; performing interpolation caching processing on the controller instruction based on the frequency error value of the controller instruction to obtain frequency error information of instruction caching; and performing dynamic period adjustment according to the frequency error value of the controller instruction and the frequency error information of the instruction cache, and obtaining a control instruction rate matching result. Frequency detection and dynamic period adjustment are independently completed through the driver, a controller does not need to participate in clock calibration, the communication bandwidth is prevented from being occupied by a calibration instruction, and the system communication efficiency is remarkably improved; and in combination with interpolation cache processing, clock skew can be compensated in real time, the instruction frequency error is controlled within an extremely low range, the system fluctuation risk is effectively reduced, a communication protocol does not need to be modified, and the method is completely compatible with an existing hardware architecture.

The application provides a multi-AUV system formation coordination control method with non-convex control input constraints, converts the coordination control of the multi-AUV system formation without a leader into a formation consistency problem, and defines the consistency state of the multi-AUV system formation without a leader. In the case that the water acoustic communication bandwidth is limited, the application selects a double-layer communication channel mainly composed of position and speed information. Considering the existence of communication delay and non-convex control input constraints, a discrete-time leaderless formation consistency constraint controller algorithm with communication delay is designed by introducing a constraint operator. By using the properties of graph theory, random matrix and SIA matrix, and selecting appropriate controller parameters, the multi-AUV system formation can reach the defined consistency state and maintain the stability of the formation shape.

This invention discloses a distributed consistency control method and device for primary-distribution-microgrid collaborative energy storage, relating to the field of power system energy storage control technology. The method includes the following steps: acquiring the SOC deviation and voltage deviation of each energy storage node; converting the voltage deviation into a dynamic feedback gain through nonlinear mapping; performing asymmetric adaptive correction of the SOC deviation using the difference in dynamic feedback gains between adjacent nodes; generating a local active power regulation quantity based on the corrected SOC deviation and upper-level scheduling instructions; controlling the local energy storage converter to perform power regulation through the active power regulation quantity, updating the SOC and voltage of the local node, and using the updated SOC and voltage as the initial values ​​for the next control cycle. This invention achieves spontaneous transfer of physical stress in a microgrid group and self-organized survival of the system without increasing the communication bandwidth of the centralized controller.

This application relates to a multimodal Transformer parallel inference method and system based on dynamic operator decomposition. For each multimodal request, the method dynamically decomposes the computational tasks of the multimodal Transformer model at the operator level based on the network architecture configuration information of the multimodal Transformer model, the memory and communication bandwidth status of each computing device node in the target computing cluster, and the load characteristics of the multimodal request. For each operator, multiple candidate device allocation strategies are determined. The total cost corresponding to the operator under each candidate device allocation strategy is determined. Based on the total cost, the target device allocation strategy corresponding to the operator is determined from the multiple candidate device allocation strategies. Each operator and the data it processes are scheduled to the computing device nodes indicated by the target device allocation strategy for parallel inference computation. This method can reduce the end-to-end inference response latency of the model.

The invention discloses an ionosphere real-time monitoring method, device and system, and a storage medium, and the method comprises the steps: configuring a multi-frequency GNSS receiver on a low-orbit satellite, obtaining a navigation satellite signal in real time, rapidly calculating the total electron content of an oblique ionosphere through the dual-frequency carrier phase difference, and calculating the observation confidence based on a Kalman filtering model; further, the satellite end carries out anomaly monitoring through a dynamic threshold value, and respectively generates and downloads a compressed observation flow containing a quality index and original carrier phase observation data fragments in a time window before and after the anomaly is triggered; and the ground station performs weighted fusion on the compressed observation flow according to the confidence coefficient to generate high-precision TEC data. According to the technical scheme, high-precision TEC data products for ionosphere research and abnormal original data products for deep diagnosis are finally produced through on-satellite intelligent processing and selective downloading, communication bandwidth occupation can be remarkably reduced, and observation real-time performance and data reliability are improved.

The application discloses an outsourcing encrypted database system and method based on a TEE server supporting a secure aggregation operation and a secure query operation. The system comprises a client and a cloud server configured with a rich execution environment (REE) and a trusted execution environment (TEE). The client generates a key pair, and adopts a Paillier threshold encryption strategy to split the private key into two shards, which are respectively sent to the REE and the TEE, and the data encrypted by the public key is stored in the REE. When responding to a query or aggregation instruction, the REE and the TEE cooperatively execute a cryptography operation protocol by using the private key shards held by each other through a hardware isolation mechanism. In order to further improve the processing efficiency, the application also introduces amortization optimization technology, which uses the property of homomorphic encryption to package multiple independent comparison tasks into a ciphertext for one-time joint decryption, thereby significantly reducing the calculation overhead and communication bandwidth.