Communication system based on perception, algorithm model, rich storage and intention arrangement

Abstract

The invention discloses a communication system based on perception, an algorithm model, rich storage and intention arrangement, which realizes real-time acquisition and preprocessing of equipment states, environmental parameters and operation logs through a communication network autonomous domain module, and provides basic guarantee for efficient analysis of data. A perception analysis module is combined with a machine learning algorithm to realize millisecond-level anomaly detection and fault mode identification; a virtual model is constructed through the analogue simulation module, and the operation state and the potential risk of the equipment are predicted, so that the accuracy and the timeliness of fault prevention are improved; unified management of structured, unstructured and multimedia data is supported through the rich media storage module, data standardization processing and compatibility management among multi-device and multi-protocol systems are achieved through the general protocol adaptation module, protocol adaptation complexity is greatly reduced, and the communication system is more intelligent and more reliable while real-time performance, intelligence and adaptability of the communication system are improved. And comprehensive support is provided for stable operation and intelligent operation and maintenance of a future communication network.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication system based on perception, algorithm model, rich storage and intent orchestration. Background Art

[0002] With the rapid development of communication technology, future communication networks will gradually become an important infrastructure to support the digital transformation of thousands of industries. For the next generation of communication networks, communication systems need to meet the needs of diverse business scenarios, requiring not only low latency, high bandwidth and high reliability, but also high deterministic guarantees and intelligent operation and maintenance capabilities. However, the architecture and technology of traditional communication systems are gradually unable to meet these requirements, especially in key areas such as industry, energy, medical care, and transportation. The requirements of communication systems for efficient perception, accurate decision-making, and real-time response are particularly urgent.

[0003] Most existing communication systems are based on fixed protocol stacks and static data processing methods, and mainly rely on manual operation and maintenance and preset rules to monitor and manage network operations. Traditional systems have significant deficiencies in data perception, fault prediction accuracy, and multi-scenario adaptability. Specifically, on the one hand, the data perception capabilities of existing systems are limited, and it is difficult to respond to rapidly changing network states at the microsecond level; on the other hand, the existing algorithm models are insufficient in detecting and predicting fault modes and network hidden dangers in complex scenarios, resulting in the inability to timely discover and effectively handle equipment and network operation risks. In addition, traditional communication systems have weak support for the storage and call of multiple types of data (such as structured and unstructured data), and cannot meet the needs of intelligent analysis of multimedia information (such as video, audio, images, etc.). Finally, due to the diversity of communication equipment and protocols, the system's protocol adaptation and management process is relatively cumbersome, and there is a lack of unified and automated solutions, which further limits the intelligence and operation efficiency of the communication system.

[0004] The existing communication systems have problems such as low efficiency in data perception and analysis, insufficient fault prediction and hidden danger detection capabilities, weak support for storage and call of multiple types of data, and complex protocol adaptation and management. These problems have become technical challenges that need to be solved urgently. Summary of the invention

[0005] The present application provides a communication system based on perception, algorithm model, rich storage and intent orchestration, aiming to solve the problems of low efficiency of data perception and analysis, insufficient fault prediction and hidden danger detection capabilities, weak support for storage and call of multiple types of data, and complex protocol adaptation and management in existing communication systems.

[0006] A communication system based on perception, algorithm model, rich storage and intent orchestration, the system comprising:

[0007] The communication network autonomous domain module is used to define and manage resources, policies and rules within the autonomous domain, and collect raw data of communication devices, including device status information, environmental parameters and operation logs;

[0008] A universal protocol adaptation module, connected to the communication network autonomous domain module, for standardizing the communication protocol, including protocol identification, data format conversion, message parsing and protocol translation;

[0009] A perception and analysis module, connected to the general protocol adaptation module, for receiving standardized data and performing real-time analysis and pattern recognition to identify equipment operating status, abnormal conditions and potential failure modes;

[0010] A simulation module, connected to the perception analysis module, for creating a virtual model based on the received analysis data, simulating the system operation status, predicting the future status of the equipment and evaluating potential risks;

[0011] Rich media storage module, used to store and manage structured data, unstructured data, and multimedia data, including text, audio, video, and images;

[0012] An intention decision control module, connected to the simulation module and the rich media storage module, is used to automatically execute operation and maintenance decisions according to preset rules and real-time analysis results;

[0013] Computing and algorithm modules are used to support the execution of deep learning, machine learning, and data mining algorithms, and provide parallel computing capabilities to optimize model training and reasoning;

[0014] The model orchestration programming module is used to manage and orchestrate various algorithm models and their workflows, and define task dependencies, execution order, and trigger conditions.

[0015] In the above scheme, optionally, the communication network autonomous domain module is used to pre-process the collected raw data through timestamp marking to facilitate efficient processing and analysis by subsequent modules.

[0016] In the above solution, optionally, the general protocol adaptation module includes:

[0017] A protocol identification unit, used to identify the communication protocol type of the connected device;

[0018] A data conversion unit, used to convert non-standardized data into a unified standard format;

[0019] A message processing unit, used to verify and parse received and sent messages;

[0020] Protocol translation unit, used to achieve interoperability between different communication protocols.

[0021] In the above solution, optionally, the perception analysis module includes:

[0022] a data analysis unit for performing pattern recognition based on real-time data;

[0023] Anomaly detection unit, used to identify abnormal behavior and potential failure modes in equipment operation;

[0024] The rule management unit is used to call predefined rules for data association analysis.

[0025] In the above solution, optionally, the simulation module includes:

[0026] A data preprocessing unit, used to clean and format the data provided by the perception analysis module;

[0027] A model building unit for creating a virtual model based on the physical characteristics and historical behavior patterns of the equipment;

[0028] A simulation execution unit, used to simulate the operation behavior of the device under different conditions;

[0029] The risk assessment unit is used to output the predicted equipment health status and potential risks.

[0030] In the above solution, optionally, the rich media storage module supports multi-level storage management, including:

[0031] Structured data storage for storing device status and environmental parameters;

[0032] Unstructured data storage, used to store operation logs;

[0033] Multimedia data storage, used to store audio, video and image data.

[0034] In the above solution, optionally, the intention decision control module includes:

[0035] A rules engine that analyzes input data based on expert knowledge and operational intent;

[0036] A decision execution unit, which is used to automatically generate operation instructions for adjusting equipment parameters, triggering alarms, or generating work orders;

[0037] The optimization unit is used to select the best decision solution according to resource constraints.

[0038] In the above solution, optionally, the computing power and algorithm module supports parallel processing, including:

[0039] Data sharding unit, used to split large-scale data sets into multiple subsets;

[0040] Algorithm execution unit, used to perform computational analysis on shard data;

[0041] Model training unit, used to optimize deep learning and machine learning model parameters;

[0042] Model inference unit, used to make predictions and classifications on new data.

[0043] In the above solution, optionally, the model arrangement programming module includes:

[0044] Task management unit, used to define dependencies between tasks;

[0045] Process control unit, used to set the execution order and trigger conditions of tasks;

[0046] The scheduling management unit is used to adjust the execution status and resource allocation of tasks in real time.

[0047] In the above solution, optionally, the system further includes:

[0048] Data interface module, used to realize data interaction and interface call with external systems;

[0049] The security management module is used to verify permissions and control access to system data and operating processes.

[0050] Compared with the prior art, this application has at least the following beneficial effects:

[0051] Based on further analysis and research of existing technical problems, this application recognizes that the existing communication system has problems such as low efficiency of data perception and analysis, insufficient fault prediction and hidden danger detection capabilities, weak support for multi-type data storage and call, and complex protocol adaptation and management. Through the synergy of multiple modules, it effectively solves the problems of low efficiency of data perception and analysis, insufficient fault prediction and hidden danger detection capabilities, weak support for multi-type data storage and call, and complex protocol adaptation and management in the existing technology. The system realizes real-time collection and preprocessing of equipment status, environmental parameters and operation logs through the communication network autonomous domain module, providing a basic guarantee for efficient data analysis; through the perception and analysis module combined with machine learning algorithms, millisecond-level anomaly detection and fault pattern recognition are realized; through the simulation module, a virtual model is constructed to predict the operating status and potential risks of the equipment, thereby improving the accuracy and timeliness of fault prevention; through the rich media storage module, unified management of structured, unstructured and multimedia data is supported to meet the storage and call requirements of diversified data in complex communication environments; through the general protocol adaptation module, data standardization processing and compatibility management between multiple devices and multi-protocol systems are realized, greatly reducing the complexity of protocol adaptation. In general, the present invention improves the real-time, intelligent and adaptable performance of the communication system while providing comprehensive support for the stable operation and intelligent operation and maintenance of future communication networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A block diagram of the module architecture of a communication system based on perception, algorithm model, rich storage and intent orchestration is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] In the description of this application: unless otherwise specified, the meaning of "plurality" is two or more. The terms "first", "second", "third", etc. in this application are intended to distinguish the objects referred to, and do not have special meanings in terms of technical connotations (for example, they should not be understood as emphasizing the importance or order, etc.). Expressions such as "including", "comprising", "having", etc. also mean "not limited to" (certain units, components, materials, steps, etc.).

[0055] In one embodiment, Figure 1 As shown, a communication system based on perception, algorithm model, rich storage and intent orchestration is provided, including the following steps:

[0056] The system comprises:

[0057] The communication network autonomous domain module is used to define and manage resources, policies and rules within the autonomous domain, and collect raw data of communication devices, including device status information, environmental parameters and operation logs;

[0058] A universal protocol adaptation module, connected to the communication network autonomous domain module, for standardizing the communication protocol, including protocol identification, data format conversion, message parsing and protocol translation;

[0059] A perception and analysis module, connected to the general protocol adaptation module, for receiving standardized data and performing real-time analysis and pattern recognition to identify equipment operating status, abnormal conditions and potential failure modes;

[0060] A simulation module, connected to the perception analysis module, for creating a virtual model based on the received analysis data, simulating the system operation status, predicting the future status of the equipment and evaluating potential risks;

[0061] Rich media storage module, used to store and manage structured data, unstructured data, and multimedia data, including text, audio, video, and images;

[0062] An intention decision control module, connected to the simulation module and the rich media storage module, is used to automatically execute operation and maintenance decisions according to preset rules and real-time analysis results;

[0063] Computing and algorithm modules are used to support the execution of deep learning, machine learning, and data mining algorithms, and provide parallel computing capabilities to optimize model training and reasoning;

[0064] The model orchestration programming module is used to manage and orchestrate various algorithm models and their workflows, and define task dependencies, execution order, and trigger conditions.

[0065] In this embodiment, the communication network autonomous domain module: This module is the data collection entrance of the entire system, responsible for defining and managing the resources, policies and rules within the autonomous domain. As the smallest unit of the communication network, the autonomous domain can independently manage its internal devices and resources.

[0066] In terms of data collection, the module obtains real-time device status information (such as operating status, performance indicators), environmental parameters (such as temperature, humidity, electromagnetic interference, etc.) and operation logs (such as historical operation records, system event logs, etc.) through sensors, log collection programs and environmental detection tools deployed on the device side.

[0067] To ensure the uniformity and availability of data, the collected data will be timestamped and initially processed through preprocessing procedures, such as data cleaning, redundancy removal, and format conversion, in order to provide high-quality input for subsequent modules.

[0068] Universal protocol adapter module: The universal protocol adapter module is mainly used to solve the problem of protocol incompatibility between different communication devices and systems. Its implementation includes the following parts:

[0069] Protocol identification: Automatically detect the type of communication protocol used by the connected device through device identifiers or protocol signature libraries.

[0070] Data format conversion: Use built-in data conversion rules to convert non-standardized data sent by devices into a unified format that the system can understand.

[0071] Message parsing and protocol translation: The adaptation module uses a rule base and a translation engine to perform protocol translation on data between different devices, ensuring that multi-protocol devices can communicate seamlessly.

[0072] Configuration management: The adaptation module allows users to dynamically adjust protocol adaptation parameters and rules according to actual needs, thereby improving the flexibility of the system.

[0073] Perception and analysis module: The perception and analysis module is the core data processing unit of the system. Its main function is to receive standardized data from the general protocol adaptation module and use machine learning and deep learning algorithms to identify status changes, potential hidden dangers and failure modes in the operation of communication equipment in real time.

[0074] The implementation of this embodiment includes: calling a built-in pattern recognition algorithm, analyzing device operation data, and discovering abnormal conditions, such as device overload, signal interference, link instability, etc.

[0075] Combining historical operating data with set baseline values, the anomaly detection algorithm detects deviations in operating indicators and generates alarm information.

[0076] The rule engine is used to filter and organize the analysis results to provide clear analysis reports for subsequent modules.

[0077] Simulation module: This module is used to build virtual operation models of equipment and systems, simulate different operation scenarios, predict the future status of equipment and evaluate potential risks.

[0078] The implementation of this embodiment includes: creating a mathematical model of device behavior based on the physical characteristics and historical operation modes of the device.

[0079] Environmental parameters and operating conditions are introduced into the model to simulate the operating performance of the equipment under different load, temperature and other conditions.

[0080] Combined with the simulation results, optimization suggestions are provided to operation and maintenance personnel or automation control modules, such as adjusting operating parameters, replacing components in advance, or preventing faults.

[0081] Rich media storage module: This module supports unified storage of multiple data types, including structured data (such as device status parameters), unstructured data (such as operation logs), and multimedia data (such as audio, video, and images).

[0082] The storage function adopts a layered architecture design, which supports fast query, compressed storage and efficient calling, providing strong data support for data analysis and decision-making modules.

[0083] Intentional decision control module: This module introduces the intelligent decision-making capability of human-machine collaboration into the system by injecting expert knowledge and operation and maintenance intentions. The implementation method includes the collaboration of rule engine, decision optimization algorithm and execution module. This module can automatically generate operation and maintenance decisions based on real-time analysis data, such as adjusting equipment parameters, triggering alarms or generating work orders.

[0084] Computing power and algorithm module: Improve the operating efficiency of the algorithm through parallel computing technology. Large-scale data processing tasks will be divided into multiple subtasks and assigned to different computing nodes for parallel execution.

[0085] The computing power module also supports the training and reasoning of deep learning models, providing powerful computing power for the perception analysis module and simulation module.

[0086] Model orchestration programming module: manage the workflow between modules through the orchestration tool, clarify the dependencies and execution order between tasks. Users can define the trigger conditions and execution rules of tasks through a graphical interface or scripting language to achieve flexible configuration and dynamic adjustment of the system.

[0087] The integrated communication system based on perception, algorithm model, rich storage and intent orchestration proposed in this embodiment can effectively solve the following problems mentioned in the background technology:

[0088] Inefficient data perception and analysis: Through the synergy of the communication network autonomous domain module and the perception and analysis module, microsecond-level data collection and millisecond-level analysis and processing capabilities are achieved. The autonomous domain module is responsible for real-time collection and preprocessing of device status information, while the perception and analysis module uses machine learning algorithms to quickly identify abnormal situations and potential faults.

[0089] Insufficient fault prediction and hidden danger detection capabilities: The simulation module of the present invention uses virtual modeling technology to simulate the operating status of the equipment, predict possible future equipment failures, and evaluate potential risks, providing a basis for early intervention for operation and maintenance personnel.

[0090] Weak support for storage and call of multiple types of data: The rich media storage module enhances the system's support capabilities for diversified data by supporting the storage and management of structured, unstructured and multimedia data, providing comprehensive data resources for subsequent intelligent analysis and decision-making.

[0091] The protocol adaptation and management are highly complex: Through the standardized processing of the general protocol adaptation module, the adaptation problem between multi-vendor equipment and multi-protocol communication systems is solved, the compatibility and consistency of data are ensured, and at the same time, it supports automated protocol configuration management, greatly reducing the complexity of manual operations.

[0092] In summary, the present invention not only solves the problems raised in the background technology through the organic combination of the above modules, but also provides a complete and efficient solution for the intelligent development of communication networks.

[0093] In one embodiment, the communication network autonomous domain module is used to pre-process the collected raw data through time stamp marking to facilitate efficient processing and analysis by subsequent modules.

[0094] In one embodiment, the general protocol adaptation module includes:

[0095] A protocol identification unit, used to identify the communication protocol type of the connected device;

[0096] A data conversion unit, used to convert non-standardized data into a unified standard format;

[0097] A message processing unit, used to verify and parse received and sent messages;

[0098] Protocol translation unit, used to achieve interoperability between different communication protocols.

[0099] In one embodiment, the perception analysis module includes:

[0100] a data analysis unit for performing pattern recognition based on real-time data;

[0101] Anomaly detection unit, used to identify abnormal behavior and potential failure modes in equipment operation;

[0102] The rule management unit is used to call predefined rules for data association analysis.

[0103] In one embodiment, the simulation module includes:

[0104] A data preprocessing unit, used to clean and format the data provided by the perception analysis module;

[0105] A model building unit for creating a virtual model based on the physical characteristics and historical behavior patterns of the equipment;

[0106] A simulation execution unit, used to simulate the operation behavior of the device under different conditions;

[0107] The risk assessment unit is used to output the predicted equipment health status and potential risks.

[0108] In one embodiment, the rich media storage module supports multi-level storage management, including:

[0109] Structured data storage for storing device status and environmental parameters;

[0110] Unstructured data storage, used to store operation logs;

[0111] Multimedia data storage, used to store audio, video and image data.

[0112] In one embodiment, the intention decision control module includes:

[0113] A rules engine that analyzes input data based on expert knowledge and operational intent;

[0114] A decision execution unit, which is used to automatically generate operation instructions for adjusting equipment parameters, triggering alarms, or generating work orders;

[0115] The optimization unit is used to select the best decision solution according to resource constraints.

[0116] In one embodiment, the computing power and algorithm module supports parallel processing, including:

[0117] Data sharding unit, used to split large-scale data sets into multiple subsets;

[0118] Algorithm execution unit, used to perform computational analysis on shard data;

[0119] Model training unit, used to optimize deep learning and machine learning model parameters;

[0120] Model inference unit, used to make predictions and classifications on new data.

[0121] In one embodiment, the model arrangement programming module includes:

[0122] Task management unit, used to define dependencies between tasks;

[0123] Process control unit, used to set the execution order and trigger conditions of tasks;

[0124] The scheduling management unit is used to adjust the execution status and resource allocation of tasks in real time.

[0125] In one embodiment, the system further comprises:

[0126] Data interface module, used to realize data interaction and interface call with external systems;

[0127] The security management module is used to verify permissions and control access to system data and operating processes.

[0128] In one embodiment, the next generation of future communication networks will carry the scenario requirements of thousands of industries, and low latency, high bandwidth, and high certainty guarantee will become the embodiment of the differentiated value capabilities of communication networks. The rapid perception, accurate prediction model, diversified storage, and injection of expert intent of the communication system all need to evolve from the architecture and system. From the perspective of risk prediction and prevention of hidden dangers in future communication networks, fault handling and disaster recovery, network robustness assessment and improvement, network security operation guarantee, and network capability opening, real-time data reporting and processing at the microsecond level, millisecond-level rapid perception and analysis of protocols, and effective intent collaborative decision-making at the second level are all required.

[0129] This embodiment adds a human-machine collaborative intention injection scenario. By pre-setting the orchestration design state medium, expert knowledge and operation and maintenance intentions can be injected in real time and regularly to predict and observe the health status of the simulation operation; a communication algorithm model is added to automatically determine the operation trend through machine learning and KPI recommendations, adapt the anomaly detection baseline, quickly match the fault scenario, and realize self-learning and self-training of fault rules; storage types are enhanced to support the management and call of structured, unstructured and other text, images and rich media information; the universal adaptation containerization and unified management of communication protocols are enhanced, and automatic parsing, translation and execution are performed.

[0130] This embodiment realizes the intelligence of the communication system through components such as universal adapter containers, computing power modules, intent orchestration, rich storage, and algorithm models. Modeling is performed based on data factors such as alarms, performance, logs, and environments from the dimensions of single boards, network elements, and networks. Algorithm analysis and matching are performed through computing power modules to complete machine learning ML and KPI anomaly recommendations, achieve real-time, accurate, and deterministic alarm correlation analysis, software and hardware fault prediction and prevention, and network experience assurance, and improve the efficiency, reliability, and security of data processing and operation and maintenance.

[0131] Among them, intelligent computing capabilities: realize data parsing, analysis, processing, and verification of preset models, and have the self-closed-loop behavior of the network autonomous domain for analysis, simulation, decision-making, scheduling, and execution;

[0132] Operation and maintenance algorithm: Telecom operation and maintenance model and training framework, supporting model training generation and feature engineering modeling and verification;

[0133] Orchestration design state medium, which is responsible for developing and arranging programming to inject expert knowledge and operation and maintenance intentions into the system;

[0134] The universal protocol adapter container manages the universal protocols of different manufacturers and devices in a unified containerized manner, and supports the automation of the command parsing and translation governance process.

[0135] The detailed workflow of this embodiment is described as follows:

[0136] Through the close collaboration of multiple modules, the whole process from data collection, analysis, simulation to decision-making and execution is automated and intelligent. This comprehensive solution helps to improve the efficiency, reliability and flexibility of the system while reducing operating costs and risks.

[0137] Communication network autonomous domain data collection: As the starting point of the system, the autonomous domain is responsible for defining and managing its internal resources, policies and rules. As the smallest unit in the communication network system, the autonomous domain is responsible for collecting raw data from various communication devices, including but not limited to device status information, environmental parameters, operation logs, etc.

[0138] Universal protocol adaptation: When the autonomous domain needs to interact with other systems or services, the universal protocol adaptation container comes into play. It is responsible for receiving the raw data collected by the autonomous domain, standardizing and converting the protocols and data formats within the autonomous domain, ensuring data compatibility and consistency between different devices and systems, thereby achieving seamless communication between different systems and facilitating efficient processing and analysis of subsequent modules.

[0139] Protocol Identification: First identify the communication protocols used by different devices or services connected to the system.

[0140] Data format conversion: The adapter converts the data received from the device into a unified format so that other modules within the system can understand and process it.

[0141] Message processing: The adapter processes the messages sent and received by the device to ensure the correctness and integrity of the messages.

[0142] Protocol conversion: When devices use different communication protocols, the adapter can perform protocol conversion so that devices with different protocols can communicate with each other.

[0143] Configuration Management: The Generic Protocol Adapter supports configuration management, allowing the adapter parameters to be adjusted as needed.

[0144] Perception analysis: Receive standardized data and perform real-time analysis and pattern recognition to identify device operating status, abnormal conditions, and potential failure modes. Invoke deep learning and machine learning algorithms to intelligently analyze data and provide real-time monitoring and analysis of abnormal information on system performance, user behavior, network status, etc.

[0145] Simulation: Based on the data provided by the perception analysis module, the simulation module creates a virtual model to simulate the operation of the system, simulate the equipment behavior under different operating conditions, evaluate the potential impact of different decision-making plans, predict the future status of the equipment, and evaluate potential risks, thereby providing strong support for decision makers.

[0146] Data input and preprocessing: Receive data from the perception analysis module, which may include device status, environmental parameters, historical operation records, etc. During the data preprocessing stage, the module will clean and convert the input data.

[0147] Model building: Based on the preprocessed data, the module builds a virtual model of the device or system taking into account the physical characteristics, operating environment and historical behavior patterns of the device.

[0148] Simulation execution: simulate the behavior of the equipment under different conditions, such as temperature changes, load increases, etc. Predict the future state of the equipment, assess potential failure risks, or optimize the operating parameters of the equipment.

[0149] Result analysis and feedback: Extract key information, such as the health status and performance indicators of the equipment, and feed it back to the intention decision control module as the basis for decision-making, such as adjusting equipment parameters, triggering alarms, or generating work orders.

[0150] Rich media storage: Responsible for storing and managing large amounts of multimedia data, such as video, audio, and images, providing data support for advanced analysis. This data can be used for video surveillance analysis, device sound anomaly detection, etc., to enhance the system's intelligent perception capabilities.

[0151] Intentional decision control: Automatically perform maintenance operations and make decisions based on preset rules and real-time analysis results, such as adjusting equipment parameters, triggering alarms, or generating work orders. This module considers various factors, such as business goals, resource constraints, risk preferences, etc., to determine the optimal decision.

[0152] Computing algorithm: The computing algorithm module provides powerful computing resources to support the operation of complex algorithms, such as deep learning, machine learning, etc. It ensures that the system can process large-scale data, perform advanced analysis, and improve the accuracy of decision-making and the overall intelligence level of the system.

[0153] Algorithm execution: The module supports the execution of multiple algorithms, including but not limited to deep learning, machine learning, data mining, etc.

[0154] Parallel processing: To speed up the execution of algorithms, the computing algorithm module supports parallel processing, which can split large data sets into multiple subsets and execute them simultaneously on different computing nodes, and finally summarize the results, significantly shortening the processing time.

[0155] Model training and reasoning: For deep learning and machine learning algorithms, the computing algorithm module supports model training and reasoning. The training phase uses a large number of data sets and computing resources to optimize model parameters and improve the model's predictive ability. The reasoning phase uses the trained model to predict or classify new data, supporting real-time decision-making and intelligent analysis.

[0156] Model orchestration programming: Manage and orchestrate various algorithm models, and orchestrate workflows to form a coherent, automated workflow. Programming languages ​​and tools are used to define the dependencies, execution order, and trigger conditions between tasks, thereby improving the flexibility and efficiency of system intelligent operation and maintenance.

[0157] In this embodiment, the efficiency of data perception and analysis is improved, and KPI anomalies, signaling link establishment, microbursts, retransmissions, routing table entries and messages are processed and analyzed at the microsecond level; the stability and health of equipment and networks are improved, and risks, hidden dangers and faults such as chip board failure, link traffic congestion, and single-point network exit of network elements are discovered and monitored at the millimeter level; the efficiency of network security operation and maintenance is improved, and simulation, switching, dialing, disaster recovery and switching, etc., are coordinated and decided by effective human intentions to achieve an autonomous self-healing closed loop at the second level.

[0158] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A communication system based on perception, algorithm model, rich storage and intent orchestration, characterized in that: The system comprises: The communication network autonomous domain module is used to define and manage resources, policies and rules within the autonomous domain, and collect raw data of communication devices, including device status information, environmental parameters and operation logs; A universal protocol adaptation module, connected to the communication network autonomous domain module, for standardizing the communication protocol, including protocol identification, data format conversion, message parsing and protocol translation; A perception and analysis module, connected to the general protocol adaptation module, for receiving standardized data and performing real-time analysis and pattern recognition to identify equipment operating status, abnormal conditions and potential failure modes; A simulation module, connected to the perception analysis module, for creating a virtual model based on the received analysis data, simulating the system operation status, predicting the future status of the equipment and evaluating potential risks; Rich media storage module, used to store and manage structured data, unstructured data, and multimedia data, including text, audio, video, and images; An intention decision control module, connected to the simulation module and the rich media storage module, is used to automatically execute operation and maintenance decisions according to preset rules and real-time analysis results; Computing and algorithm modules are used to support the execution of deep learning, machine learning, and data mining algorithms, and provide parallel computing capabilities to optimize model training and reasoning; The model orchestration programming module is used to manage and orchestrate various algorithm models and their workflows, and define task dependencies, execution order, and trigger conditions.

2. The system according to claim 1, characterized in that: The communication network autonomous domain module is used to pre-process the collected raw data through time stamp marking to facilitate efficient processing and analysis by subsequent modules.

3. The system according to claim 1, characterized in that: The general protocol adaptation module includes: A protocol identification unit, used to identify the communication protocol type of the connected device; A data conversion unit, used to convert non-standardized data into a unified standard format; A message processing unit, used to verify and parse received and sent messages; Protocol translation unit, used to achieve interoperability between different communication protocols.

4. The system according to claim 1, characterized in that: The perception analysis module includes: a data analysis unit for performing pattern recognition based on real-time data; Anomaly detection unit, used to identify abnormal behavior and potential failure modes in equipment operation; The rule management unit is used to call predefined rules for data association analysis.

5. The system according to claim 1, characterized in that: The simulation module comprises: A data preprocessing unit, used to clean and format the data provided by the perception analysis module; A model building unit for creating a virtual model based on the physical characteristics and historical behavior patterns of the equipment; A simulation execution unit, used to simulate the operation behavior of the device under different conditions; The risk assessment unit is used to output the predicted equipment health status and potential risks.

6. The system according to claim 1, characterized in that: The rich media storage module supports multi-level storage management, including: Structured data storage for storing device status and environmental parameters; Unstructured data storage, used to store operation logs; Multimedia data storage, used to store audio, video and image data.

7. The system according to claim 1, characterized in that: The intention decision control module includes: A rules engine that analyzes input data based on expert knowledge and operational intent; A decision execution unit, which is used to automatically generate operation instructions for adjusting equipment parameters, triggering alarms, or generating work orders; The optimization unit is used to select the best decision solution according to resource constraints.

8. The system according to claim 1, characterized in that: The computing power and algorithm modules support parallel processing, including: Data sharding unit, used to split large-scale data sets into multiple subsets; Algorithm execution unit, used to perform computational analysis on shard data; Model training unit, used to optimize deep learning and machine learning model parameters; Model inference unit, used to make predictions and classifications on new data.

9. The system according to claim 1, characterized in that: The model arrangement programming module includes: Task management unit, used to define dependencies between tasks; Process control unit, used to set the execution order and trigger conditions of tasks; The scheduling management unit is used to adjust the execution status and resource allocation of tasks in real time.

10. The system according to claim 1, characterized in that: The system further comprises: Data interface module, used to realize data interaction and interface call with external systems; The security management module is used to verify permissions and control access to system data and operating processes.

Images