3112 results about "Server-side" patented technology

The present disclosure generally relates to a two-stage system for detecting malicious nodes in Wireless Sensor Networks (WSNs), enhancing network security and resilience. The system employs a distributed approach, leveraging Cluster Heads (CHs) and a central server for efficient and accurate detection. Initially, sensor nodes are monitored for comprehensive node and network metrics, statistically ranked by significance in identifying malicious behavior. CHs perform a resource-aware first-stage detection based on their resource weight, filtering potential threats locally. Results are then aggregated at a server for a second-stage analysis using a hybrid Machine Learning (ML) and Deep Learning (DL) approach. This advanced analysis, combined with statistically relevant metrics, significantly improves detection accuracy. By integrating resource-conscious CH operation with powerful server-side ML / DL, this system offers a scalable, energy-efficient, and highly effective solution for securing WSNs against malicious node attacks, surpassing traditional detection methods in both speed and precision.

The invention relates to the technical field of server upgrading, and discloses a server firmware remote upgrading method and system, and the method comprises the steps: a management node obtains a device identifier list and target firmware version information of a to-be-upgraded server, and uses a distributed task scheduling algorithm to divide tasks; transmitting an upgrading instruction and a data packet through an encryption channel, and generating an incremental packet by using a differential upgrading technology; data transmission is guaranteed through heartbeat detection and breakpoint continuous transmission; after pre-verification of the server-side security sandbox, upgrading and isolating tasks in an idle period; verifying by adopting a gray release strategy, and automatically rolling back if the verification fails; and storing an upgrade state and a log by means of a block chain network. The system comprises corresponding function modules. According to the method, the upgrading efficiency, safety and reliability are improved, the influence on services is reduced, data cannot be tampered and can be traced, the problems existing in a traditional upgrading mode are effectively solved, and the method is suitable for various server firmware remote upgrading scenes.

The invention discloses a mass concrete intelligent temperature monitoring system and method, and the system comprises a high-precision temperature sensor, a data collection node which is connected with the high-precision temperature sensor and is responsible for carrying out the primary processing and integration of temperature data collected by the sensor, and a data collection terminal which serves as a data convergence center of the whole monitoring system and is used for collecting the temperature data. The intelligent temperature monitoring system comprises a plurality of data acquisition nodes which are connected in a wired or wireless communication mode, a data monitoring platform which operates at a server end and is a core control and management unit of the whole intelligent temperature monitoring system, and a circulating water cooling control system which is composed of a cooling water pump, a cooling tower, a cooling water pipe network, a temperature control main station and the like. The cooling water pipe network is embedded in the concrete in advance to form a closed water circulation loop; the temperature control master station receives a cooling control instruction sent by the data monitoring platform, and adjusts the rotating speed of a cooling water pump and the working state of a cooling tower according to the instruction, so that the flow and temperature of cooling water in a cooling water pipe network are accurately controlled; through flowing of circulating water in the concrete, redundant heat in the concrete is taken away, effective regulation and control over the temperature in the concrete are achieved, key indexes such as the highest temperature and the temperature gradient of the concrete are controlled within a reasonable range all the time, and concrete cracks caused by temperature stress are avoided.

The invention relates to the technical field of neuromuscular electrical stimulation, and provides a multi-modal feedback self-adaptive adjustment neuromuscular regulation and control treatment system, which comprises wearable working equipment for outputting pulse electric signals to adjust nerves and muscles to form movement or adjustment feeling, electromyographic signals, motion data and user feeling input are collected to conduct feedback adjustment on output parameters of the output pulse electric signals; the handheld control equipment is used for being connected with the wearable working equipment and controlling the output working mode and the working state of the wearable working equipment; and the server-side equipment is used for managing user information and preference of the handheld control equipment, setting function authority of the handheld control equipment, acquiring data acquired by the wearable working equipment, and realizing self-adaptive output of output parameters of the pulse electric signals of the wearable working equipment based on a self-adaptive reinforcement learning algorithm. Adaptive optimization of millisecond-level dynamic electrical stimulation parameters is realized through a multi-modal feedback data fusion technology in combination with a layered reinforcement learning architecture.

To provide a method, an apparatus, a system, an architecture, and an interface for a wireless transmit / receive unit (WTRU) to discover available edge network (EN) management (ENM) servers associated with either an edge network (EN) service and an EN application.SOLUTION: A method may include sending a first message to a core network entity including information indicating a request for discovery of an ENM server, receiving a second message from the core network entity responding to the request for discovery of an ENM server, the second message includes ENM server information indicating (1) any number of available ENM servers and (2) respective available ENM server endpoint information, and performs communication with one or more of the available ENM servers.SELECTED DRAWING: Figure 2

The invention relates to the technical field of Web front-end development, and discloses a front-end intelligent menu caching and dynamic authority control method. The method comprises the following steps: firstly, checking whether menu cache data in a local storage of a browser exists and is valid, if so, directly using the cache data, otherwise, obtaining latest menu data from a server and caching according to a preset strategy; the menu data change state of the server side is monitored in a timed polling or WebSocket mode, and the local cache is updated in time when the data change; acquiring role permission information of the current user, matching the role permission information with the menu data, and dynamically generating an accessible menu tree; when routing jump is carried out, the type of a target address is intelligently judged, permission verification is carried out on an internal routing address, and it is ensured that a user can only access pages within the authorization range of the user. According to the application, the server request frequency is reduced, the response speed of the front-end application is improved, and meanwhile, the real-time performance of the menu data and the access security are ensured.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses a federal collaborative optimization method, device and equipment and a medium. Each participant terminal executes instruction tuning training and value alignment training based on the initial global model, generates model parameters and uploads the model parameters, and the server side aggregates the received model parameters, generates an iterative global model and repeats the training process until a preset performance index is reached. According to the method, instruction tuning and value alignment training are carried out in parallel at the terminal side of the participant, and the optimization result is fused at the server side, so that the adaptability and training efficiency of the model to multi-source data are improved, the problems that in the prior art, the optimization target is single, and aggregation strategies are lack of collaboration are solved, and the practicability and generalization ability of the model are enhanced.

A method includes receiving, at a server from a user device, a user prompt segment to a large language model (LLM), obtaining an additional prompt segment from a prompt data source, identifying a electronic address in the prompt segment, replacing the electronic address with a placeholder to generate a updated prompt segment, generating a LLM prompt comprising the updated prompt segment and the user prompt segment, and sending the LLM prompt to the LLM. The method further includes receiving a response to the LLM prompt from the LLM, the response comprising the placeholder, replacing the placeholder with the electronic address to generate an updated response, and sending the updated response to the user device.

The invention discloses a user-noninductive cold and hot sensing remote memory exchange system, which belongs to the field of computer storage, and is characterized in that cluster nodes are divided into extensible remote memory service nodes and computing client nodes according to roles; the remote memory node comprises a remote memory exchange server and a memory resource registration module; the computing client node comprises a FrontSwap-based remote memory exchange client, and a remote memory exchange server side is registered as a memory resource which is non-inductive to a user; furthermore, popularity statistics based on a popularity histogram is realized in a kernel mode; the memory exchange behavior is guided through the dynamic decision-making module based on online reinforcement learning, the memory management efficiency and the system performance are remarkably improved, and efficient and non-inductive memory expansion service is provided for user application.

Systems and methods are presented herein for requesting a version of media content from a server that includes haptic feedback rending criteria compatible with the haptics capabilities of a client device. At a server, a request is received for a media asset for interaction on a haptic enabled client device, wherein the media asset comprises haptic feedback rendering criteria. Based on the request, haptic feedback capabilities of the haptic enabled client device associated with the request is determined. The haptic feedback capabilities of the haptic enabled client device are compared to the haptic feedback rendering criteria of one or more versions of the media asset available via the server. In response to the comparing, a version of the media asset comprising haptic feedback rendering criteria compatible with the haptic feedback capabilities of the haptic enabled client device is transmitted from the server to the haptic enabled client device.

The invention discloses an NB-IOT technology-based firmware remote upgrading method and an electric energy meter. The upgrading method mainly comprises the following steps of S1, an upgrading preparation stage, wherein the upgrading preparation stage comprises server end preparation and electric energy meter end preparation; s2, a data transmission stage: firmware data transmission, a breakpoint resume mechanism and fragment verification are included; s3, a firmware updating stage which comprises data analysis and storage, preparation before updating and firmware writing and starting; s4, upgrading verification and feedback, including function verification and feedback information sending; and S5, security measures including identity authentication, data encryption and digital signature. An NB-IOT module is arranged in the electric energy meter, and data transmission between the electric energy meter and the server is achieved through the NB-IOT communication technology. According to the remote firmware upgrading method based on the NB-IOT technology and the electric energy meter provided by the invention, reliable remote upgrading of the firmware of the intelligent electric energy meter is realized by adopting an NB-IOT communication mode and combining a fragmented data transmission mechanism, a firmware verification mechanism, a breakpoint resume mechanism and an upgrading feedback mechanism.

The invention provides a five-axis series-parallel numerical control machine tool machining process virtual monitoring simulation method and system based on digital twinning, and the method comprises the steps: building an information interaction mechanism between a physical machine tool and a digital twinning model, enabling a server side to collect sensor data, and transmitting the real-time data to a client side through a communication protocol; the client processes the data and drives the digital twin model. And aiming at potential errors of twin system forecast, tool setting error compensation can be carried out in real time, so that the machining precision is improved. Meanwhile, the system supports an offline simulation function, combines an inverse kinematics algorithm and an NC code, fuses a material removal model, simulates a machining process, evaluates machining performance and errors, and provides an optimization basis for machining strategies of workpieces with different geometrical shapes and materials. According to the method, bidirectional interaction between the physical entity and the virtual model is achieved, the real-time monitoring and visualization capability of the machining process is enhanced, a reliable simulation evaluation and optimization means is provided for machining of complex parts, and the method has wide industrial application prospects.

The invention provides a personalized federal learning method and system for a shared model, and the method comprises the steps: enabling a server to store and initialize the shared model, a global model and a global class prototype of a client, and enabling the client to initialize a local learning weight vector; the server sends sharing models of other clients to one client, and the client sets a local sharing model as a global model and trains the global model, and uploads the sharing model and a local class prototype to the server; and the server calculates and obtains a global class prototype and a global model according to all the received shared models and local class prototypes. According to the method, the knowledge sharing problem in federal learning is solved, and the performance of a personalized model is improved; the problem of offset in local model training of the client is solved, and the obtained personalized prototype contains more abundant global information than the personalized prototype; while state dependence and communication overhead are reduced, effective fusion of global knowledge and local characteristics is realized so as to improve robustness and adaptability of the model in a complex scene.

The invention relates to a large model data protection method and device based on a trusted environment, equipment and a medium. In the scheme, after a server side obtains original large model data, the original large model data needs to be encrypted and signed, and then the original large model data and an encrypted information table are issued to a target application side; when the target application end calls the large model data, performing exception check according to a credible strategy; if the check is passed, performing signature verification on each encrypted data slice; if the signature verification succeeds, decrypting the encrypted slice data by using the encrypted information table to obtain original large model data; wherein the server side and the target application side are both in a trusted environment. It can be seen that the trusted environment and the key system are combined, and a dual protection mechanism for large model data is achieved; moreover, before the large model data is called, the operation environment of the target application end needs to be subjected to exception check, so that an attacker is prevented from acquiring or tampering the sensitive data, and all-around data protection is provided for the large model data.

The invention relates to the technical field of medical rehabilitation training, in particular to an intelligent hand training method and system, electronic equipment and a storage medium, and the method comprises the steps: receiving a training mode instruction selected by a user; real-time biomechanical data of the affected side hand are collected through a multi-source sensor array, the real-time biomechanical data comprise finger joint angles, electromyographic signal intensity and a touch pressure distribution matrix, and when the affected side hand is in a mirror image training mode, the six-degree-of-freedom motion trail of the uninjured side hand is synchronously collected; calling a corresponding algorithm model for processing based on the training mode, and generating a control instruction set containing pneumatic pressure parameters, electrical stimulation waveforms and virtual interaction instructions; the control instruction set is sent to a server side, a mechanical glove arranged on the hand of the affected side in a sleeving mode executes corresponding pneumatic-electric combined driving operation, and multi-sense feedback is synchronously presented. According to the application, the composite requirements of patients in different rehabilitation stages on training mode diversity, control accuracy and interactive immersion can be met.

Disclosed in the present invention are a large language model federated fine-tuning method and apparatus based on gradient compression. The method comprises the following steps: constructing, on the basis of a gradient tensor generated during fine tuning of a large language model, a raw data set having a time series relationship, performing inference by means of an autoencoder to obtain a reconstructed gradient data set, and constructing a reconstruction loss function to optimize the autoencoder; and initializing a base model of the large language model as a global model at a server end, the server end updating the global model to a client, using a pre-trained encoder to obtain a compressed gradient at the client, and at the server end, using a pre-trained decoder to decode and aggregate the compressed gradient, and then updating the global model. The present invention can improve the fine-tuning efficiency of the large language model and reduce computing resource requirements while ensuring data privacy protection, and is suitable for application scenarios such as communication optimization improvement and privacy protection enhancement in the process of scientific computing-oriented large model fine-tuning and training.

The invention provides a grouping federal recommendation method based on bilateral additive article embedding. A central server constructs a double-layer article embedding characterization structure under a federated learning framework: a client locally maintains user personalized embedding and an article local personalized embedding matrix, and a server generates global group shared article embedding through a dynamic clustering grouping mechanism; superposing global sharing embedding and local personalized embedding by adopting an additive fusion strategy to generate user side personalized article characterization; a progressive course learning scheme is designed, and smooth transition from complete personalization to additive representation is realized by dynamically adjusting a regularization weight coefficient; and a grouping and clustering process is optimized in combination with a knowledge migration strategy, and collaborative knowledge sharing across user groups is promoted. In a client local training stage, a personalized recommendation loss function based on binary cross entropy is constructed, global shared embedding and local embedding parameters are synchronously updated, and a server side updates a global model through grouping federation aggregation. On the premise of protecting user privacy, the problems that in traditional federated recommendation, article characterization is simplified, and personalized perception is insufficient are effectively solved, the accuracy of a recommendation system is remarkably improved, communication overhead is reduced, and the method is suitable for personalized recommendation services of privacy sensitive scenes such as e-commerce and content platforms.

The invention discloses a mobile communication method based on anti-quantum and national cryptographic algorithm mixing and a mobile terminal. The method comprises the steps that the mobile terminal generates a first mixed verification public key and a first mixed signature private key; obtaining a mixed signature certificate; the mobile terminal obtains the mixed ciphertext and the first symmetric key, and the mobile terminal signs the mixed ciphertext to obtain a mixed signature; the validity of the hybrid signature certificate is verified by the server side by using a second hybrid verification public key, the validity of the hybrid signature is verified by a first hybrid verification public key built in the hybrid signature certificate, and the hybrid signature is decrypted by the server side by using a second hybrid decryption private key to obtain a first symmetric key; and the mobile terminal and the server side perform encryption communication by using the first symmetric key. Compared with the prior art, the mobile communication method based on the anti-quantum and national cryptographic algorithm mixture provided by the invention ensures the secure communication between the client and the server, and can maintain the confidentiality and integrity of data even under the threat of quantum computing.

The embodiment of the invention discloses an offline intelligent wearable device NFC service processing scheme. Comprising the following steps: controlling an NFC card reader on the offline intelligent wearable equipment to be in a closed state; receiving an NFC service triggering instruction; entering a specified service application, and automatically switching the NFC card reader from a closed state to an open state through the specified service application; the target information is read from the target NFC machine tool through the NFC card reader in the open state, and the user information of the offline intelligent wearable device is written back to the target NFC machine tool, so that the target NFC machine tool interacts with the server side of the specified service application according to the user information and executes the corresponding target service; according to the target information, short-distance wireless communication connection with the target NFC machine tool is established, an execution result of the target service returned by the target NFC machine tool is received, and the maximum communication distance corresponding to the short-distance wireless communication connection is larger than the NFC communication distance of the target NFC machine tool.

The invention provides a multi-mode medical record intelligent generation and interaction system and method based on a wearable terminal and an MOE architecture, and relates to the technical field of AI large models. The terminal comprises a wearable terminal used for carrying out real-time acquisition and interaction on multi-modal data in a medical scene to obtain multi-modal medical data; the communication network is used for realizing data interaction between the wearable terminal and the server-side AI large model processing system; the server side AI large model processing system is used for performing semantic analysis, structured extraction and medical record document generation based on the multi-modal medical data to obtain an editable first draft of a medical record document; and the wearable terminal is also used for receiving and presenting the first draft of the medical record document returned by the server side AI large model processing system, and editing and confirming the first draft of the medical record document to form a returned draft of the medical record document. The problem that in the prior art, a terminal lacks multi-modal real-time fusion and LLM closed loop, and doctor duplication interaction cannot be achieved is solved.

The invention discloses an Internet of Vehicles federated learning excitation method and system based on a Nash game and a coalition game, and the method comprises the steps: initializing a vehicle client set in an Internet of Vehicles environment, and enabling a server to broadcast global model parameters to all clients; a two-stage Nash game model is established, the first stage is used for determining an alliance set in the vehicle client set, and the second stage is used for bargaining to optimize resource contribution; negotiating and determining an optimal data contribution ratio and an optimal resource contribution ratio based on the server side and the alliance set by using the two-stage Nash game model; and the server allocates rewards to the alliance set according to the optimal data contribution proportion and the optimal resource contribution proportion. The method can adapt to high dynamics of the Internet of Vehicles, and fair and efficient resource allocation is realized.

The invention discloses a federal learning data aggregation method for predicting traffic flow, a medium and equipment, and the method comprises the steps: setting a relation between a cloud server end and a vehicle end, and building a corresponding system; collecting vehicle data through the system for local training and data preprocessing; the vehicle side encrypts the preprocessed data through a homomorphic encryption technology and uploads the data to the cloud server side; carrying out weight distribution on the encrypted and uploaded data based on a kernel ridge regression technology; the encrypted update quantity is aggregated based on the adjusted weight, so that the update contribution to the global model is more similar, and the more reliable vehicle update quantity plays a more important role in the prediction process; and the aggregated encryption update quantity is decrypted, pre-training model updating is completed, finally, the server in the training wheel screens the weights of all vehicles, a part of vehicles with too low weights are eliminated, certain residual data similarity is ensured, and data aggregation of residual data is easier.

The invention discloses a dynamic scheduling optimization method for a low-altitude logistics distribution network, and the method comprises the steps: a server side builds a multi-source sensing network through satellite remote sensing, an unmanned plane airborne sensor and ground traffic monitoring, fuses meteorological data, airspace control data and order distribution data which are collected in real time, and generates a four-dimensional space-time grid map; based on the four-dimensional space-time grid map, the server side adopts a TD3-GA hybrid intelligent algorithm to carry out path planning of the logistics distribution network; the edge calculation end generates an optimal scheduling scheme of the unmanned aerial vehicle group through a multi-objective optimization function based on the global path; and the server side performs security risk assessment on the optimal scheduling scheme by using a Bayesian network model, and dynamically adjusts a space-time routing strategy of the unmanned aerial vehicle cluster according to an assessment result. According to the method, in a large-scale unmanned aerial vehicle concurrent scheduling scene, the scheduling efficiency can be effectively improved, the response time delay is reduced, the risk prediction accuracy is improved, and the timeliness and safety of a low-altitude distribution network are remarkably improved.

The invention discloses a federated learning method based on quantum transfer learning, relates to the cross technical field of quantum computing and federated learning, and aims to solve the problem of performance bottleneck of traditional federated learning under non-independent identically distributed data. The local model of each client is fused with a classic convolutional layer and a quantum convolutional layer; during local training, firstly, classic features are extracted by a classic convolutional layer, and then the classic features are coded and input into a quantum convolutional layer to generate quantum features; and the two features are spliced and then classified, and model parameters are uploaded. And the server side adopts a FedAvg algorithm to aggregate parameters so as to update the global model. According to the method, a transfer learning mechanism is introduced to optimize the initialization of the quantum layer, and an experimental result shows that the classification performance and robustness of the method are remarkably superior to those of a traditional federated CNN model on a non-IID image data set, and the method is particularly suitable for a privacy protection cooperative computing scene.

The invention discloses a federal large model knowledge collaborative training method supporting multi-modal heterogeneous clients, which comprises the following steps: each client receives a model initialization parameter issued by a central server, and applies adaptive differential privacy noise to independently train a heterogeneous lightweight model based on local private data; updating the model to which the noise is applied and uploading a modal identifier of the model to a central server side; after model updating and modal identification of each client are received, based on a modal perception weighted consensus fusion mechanism, knowledge of each client is fused to update a global large model; and the central server side issues the updated presentation layer parameters of the global large model to the client side for initialization of the next round of local training. According to the method, on the premise that a public data set or specific task setting is not needed, comprehensive compatibility of data isomerism, client dynamic participation, model diversity and privacy protection requirements is achieved, and the adaptability, stability and knowledge utilization efficiency of large model federation training are remarkably improved.

The invention relates to the technical field of industrial Internet of Things and wireless sensor networks, in particular to a large-scale industrial fault diagnosis system and method based on Bluetooth MESH, and efficient monitoring and fault early warning of industrial equipment are realized through low-power-consumption sensor nodes, edge computing, a Bluetooth Mesh network and a clustered Mesh tree architecture. The system comprises a client, a server, a gateway, a cluster head and a cluster node, the cluster node comprises a sensor node and a relay node, a lightweight neural network model is built in the sensor node, and edge reasoning and fault detection can be realized locally. Stable communication and real-time cooperation of large-scale nodes are ensured through a decentralized architecture of Bluetooth Mesh, establishment of a dynamic cluster head chain and optimization of an RSSI threshold value. And a low-power-consumption mechanism combining event driving and fixed polling is adopted, so that the energy consumption of the system is remarkably reduced. In-cluster communication loads are reduced through a clustering architecture, and the communication efficiency and the anti-interference capability of the system are improved in combination with channel separation of Mesh and BLE protocols.

The invention relates to the technical field of medical instrument intelligent review, and discloses a medical instrument registration intelligent review system and method based on a large model, and the system comprises a user side and a cloud server side. The user side comprises an enterprise user side and an evaluation mechanism side; a file generation module, a file submission module and a form filling module are arranged in the enterprise user side; an intelligent review module and an intelligent distribution module are arranged in the cloud server side; the intelligent review module comprises a form review unit and a technical review unit; the form review unit is used for intelligently obtaining a form review result; the technical review unit is used for intelligently acquiring a technical review result; and the review institution end is used for rechecking the form review result and the technical review result by reviews and forming a final review result. According to the method, the declaration material can be intelligently generated, the review efficiency, accuracy and scientificity are improved by utilizing the large model, review decision can be assisted, the labor cost is reduced, the regulation adaptability is high, and the intelligent development of medical instrument registration review is promoted.

The invention discloses a method and a system for realizing remote ADB debugging of Android equipment based on Frpc, and belongs to the technical field of remote debugging of mobile equipment. Aiming at the problems of complex configuration, low security, poor network penetration capability and the like of a traditional scheme, a system architecture including dynamic configuration management, bidirectional security authentication and intelligent tunnel maintenance is constructed by embedding an Frpc client in an equipment end. An encryption tunnel is established between an equipment end and a server end through TLS1.3, ADB flow reverse proxy transmission is achieved, configuration safety is guaranteed by adopting a dynamic token and HMAC-SHA256 encryption, and hot loading update and certificate two-way verification are supported; and an exponential backoff reconnection strategy is designed, and the network outage recovery capability is improved. According to the scheme, the configuration complexity is reduced by 83% in a public network environment, the connection speed is increased by 5 times, the data leakage risk is reduced by 97%, seamless debugging across NAT networks is supported, and a safe and efficient solution is provided for remote maintenance of Internet of Things equipment and intelligent terminals.

The invention belongs to the field of intelligent transportation, and provides a federated learning method and system based on bidirectional feedback knowledge distillation and differential privacy. Dynamic collaborative optimization of a client and a server is realized by designing a bidirectional feedback mechanism, and a knowledge distillation framework is designed at the client, so that the dynamic collaborative optimization of the client and the server is realized. Local model features in a non-independent identically distributed data environment are respectively extracted, traffic data privacy is protected in combination with an adaptive differential privacy algorithm, knowledge distillation and privacy protection strength is dynamically adjusted by means of a bidirectional feedback mechanism, the balance capability of model performance and privacy protection is remarkably improved, and the service life of a server is prolonged. According to the method, a self-adaptive knowledge distillation fusion strategy based on bidirectional feedback is designed, global model optimization is carried out, prediction precision and communication efficiency in a complex traffic scene are further improved, user privacy leakage is effectively avoided, and precision of a federal learning model and adaptability to the complex traffic scene are improved.

The invention provides a communication segmentation learning system and method for adaptive channel compression, and a medium, and relates to the technical field of segmentation learning and communication compression. The system comprises a plurality of clients, a server side and a channel compression device arranged between the clients and the server side, the channel compression device comprises a channel sensitivity modeling module, a rate distortion adaptive compression module and a cross-client fair coordination module. In a channel sensitivity modeling module, channel importance is dynamically evaluated through intermediate layer activation value fusion; differentiated quantization and compression strategies are designed on the basis of sensitivity scores in a rate-distortion self-adaptive compression module, so that key information is reserved while communication overhead is reduced; and meanwhile, the cross-client fair coordination module realizes balanced distribution of communication resources among multiple clients through a fairness regularization and dual optimization mechanism, so that the influence of excessive compression on global convergence is avoided. According to the method, the communication efficiency and the training stability of segmentation learning in a complex heterogeneous environment are remarkably improved.