Poor quality cell handling method, apparatus and medium for wireless network

Abstract

This disclosure provides a method, apparatus, and medium for processing poor-quality cells in wireless networks. When applied to a network optimization expert agent, the method includes: compiling several diagnostic sub-tasks and their chained relationships for poor-quality cells in a wireless network; distributing each diagnostic sub-task to a corresponding poor-quality expert agent, so that each poor-quality expert agent obtains its own diagnostic sub-results based on the corresponding diagnostic sub-tasks; sending the diagnostic sub-results to the network optimization expert agent; receiving the diagnostic sub-results sent by each poor-quality expert agent; and obtaining the diagnostic results of the poor-quality cells in the wireless network based on the several diagnostic sub-results and their corresponding chained relationships. The diagnostic results include the root cause of the poor-quality cells and suggestions for handling them. This disclosure employs multi-agent collaboration to provide a network optimization auxiliary tool with a high degree of intelligence and fast processing speed.

Description

Technical Field

[0001] This disclosure relates at least to the field of network technology, and in particular to a method, apparatus and medium for processing poor-quality cells in wireless networks. Background Technology

[0002] Current analyses of poor wireless network quality often rely on traditional data analysis experience and manual intervention by network optimization (RO) personnel. This results in insufficient intelligence in network optimization analysis, with the optimization level depending on the skill level of the optimization staff. Consequently, network optimization work is inefficient, inaccurate, and fails to meet the requirements of large-scale network optimization.

[0003] In particular, the analysis of poor wireless network quality may involve the analysis of data from multiple dimensions. How to achieve efficient multi-dimensional data analysis and aggregation based on intelligence is one of the technical problems to be solved in this field. Summary of the Invention

[0004] The technical problem to be solved by this disclosure is to provide a method, apparatus and medium for processing poor-quality cells in wireless networks, in order to address the above-mentioned shortcomings and improve the efficiency of analysis and optimization of poor-quality cells through automation and intelligent methods.

[0005] In a first aspect, this disclosure provides a method for processing poor-quality cells in a wireless network, the method being applied to a network optimization expert agent, and comprising:

[0006] Several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells are compiled. Each diagnostic sub-task is distributed to the corresponding poor quality expert agent, so that each poor quality expert agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent.

[0007] Receive diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of poor wireless network cells based on several diagnostic sub-results and their corresponding chain relationships. The diagnostic results include the root causes of poor wireless network quality and suggestions for handling poor quality.

[0008] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0009] Furthermore, several diagnostic subtasks and their chained relationships for poor-quality wireless network cells are developed, and each diagnostic subtask is distributed to the corresponding poor-quality expert agent, specifically including:

[0010] The system receives user questions through the first network big model and identifies the user's intent to inquire about cells with poor wireless network quality.

[0011] Based on the statistical results of the indicators of cells with poor wireless network quality, determine the real-time wireless network quality poorness type of the cells with poor wireless network quality;

[0012] Based on the real-time wireless network quality issues, the first network big model uses thought chain question-and-answer prompts to compile several diagnostic sub-tasks and their chain relationships for cells with poor wireless network quality. These diagnostic sub-tasks include at least one of the following: fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis.

[0013] Each diagnostic subtask is assigned to a corresponding quality defect expert agent. The quality defect expert agent includes at least one of the following: fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and hidden fault diagnosis agent.

[0014] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0015] Furthermore, based on several diagnostic sub-results and their corresponding chain relationships, diagnostic results for poor-quality wireless network cells are obtained. These results include the root causes of the poor-quality wireless network cells and recommendations for handling them, specifically including:

[0016] Obtain several diagnostic sub-tasks and their chain relationships, along with the corresponding thought chain question-and-answer prompts and their order. Input the several diagnostic sub-results into the question template of the second network model in order to obtain diagnostic results including the root causes of poor wireless network quality and suggestions for handling poor quality. The diagnostic sub-results include abnormal real-time data on poor wireless network quality cells and their corresponding historical handling experience information.

[0017] Output voice and / or text results of the root causes of poor wireless network quality and suggestions for handling poor quality cells; display poor wireless network quality cells on a map and mark their root causes; generate page instructions to repair the root causes of poor quality based on the suggestions for handling poor quality; and execute the page instructions automatically or when triggered by the user to repair the poor wireless network quality cells.

[0018] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0019] Furthermore, the method also includes a pre-trained large network model, specifically including:

[0020] Collect historical wireless network quality poor types, historical wireless network quality poor cell diagnostic work orders, and wireless network optimization expert knowledge. The wireless network optimization expert knowledge includes historical diagnostic sub-tasks and their chain relationships, as well as historical wireless network quality poor cell diagnostic results. The historical wireless network quality poor cell diagnostic results include the historical root causes of poor quality and historical poor quality handling solutions for the historical wireless network quality poor cells.

[0021] Control the first network model, extract thought chain question and answer prompts from the diagnostic process in the historical wireless network poor quality cell diagnostic work order corresponding to the historical wireless network poor quality type, compile historical diagnostic sub-tasks and their chain relationships under the thought chain question and answer prompts, and fine-tune the parameters of the first network model until the historical diagnostic sub-tasks and their chain relationships are accurately output.

[0022] The historical diagnostic subtasks are distributed to the corresponding poor quality expert agents, so that the poor quality expert agents can determine the wireless network API interface to be called based on the historical diagnostic subtasks, obtain historical poor quality cell data of wireless network through the wireless network API interface, judge the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts, and send the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information to the network optimization expert agent. The API interface is an application programming interface.

[0023] The system receives abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information sent by each poor-quality expert Agent. It controls the second network big model, combines the abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information with chain relationships and preset problem templates, outputs the diagnostic results of historical poor-quality cells of wireless networks, and fine-tunes the parameters of the second network big model until the diagnostic results of historical poor-quality cells of wireless networks are accurately output.

[0024] Secondly, this disclosure provides a method for processing poor-quality cells in a wireless network, the method being applied to a poor-quality expert agent, and comprising:

[0025] Receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent.

[0026] Obtain diagnostic sub-results according to the corresponding diagnostic sub-tasks, and send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network based on the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality.

[0027] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0028] Furthermore, it receives diagnostic sub-tasks distributed by the network optimization expert agent, specifically including:

[0029] Responding to the fact that it is one of the fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent, it receives the corresponding diagnostic sub-tasks distributed by the network optimization expert agent. The diagnostic sub-tasks correspond to one of the fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis, and are the first network big model of the network optimization expert agent. It is compiled by calling the thinking chain question and answer prompt words based on the real-time wireless network quality poor type of the wireless network poor cell.

[0030] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0031] Furthermore, based on the corresponding diagnostic sub-tasks, diagnostic sub-results are obtained and sent to the network optimization expert Agent, specifically including:

[0032] The corresponding diagnostic subtask determines the wireless network API interface to be called, and the real-time data of poor wireless network quality cells is obtained through the corresponding wireless network API interface. The API interface is an application programming interface.

[0033] Based on the knowledge of wireless network quality diagnosis experts, identify abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information, and use the abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information as diagnostic sub-results.

[0034] The diagnostic sub-results are sent to the network optimization expert agent so that the network optimization expert agent can obtain several diagnostic sub-tasks and their chain relationships, corresponding to the thought chain question-and-answer prompts and their order. The agent then inputs the several diagnostic sub-results into the question template of the second network big model in order to obtain diagnostic results including the root causes of poor quality in the poor quality cells of the wireless network and suggestions for handling poor quality.

[0035] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0036] Furthermore, the method also includes an auxiliary pre-trained network large model, specifically including:

[0037] Receive historical diagnostic subtasks distributed by the network optimization expert agent. These historical diagnostic subtasks are generated by the network optimization expert agent after training the first large network model.

[0038] Based on the historical diagnostic subtasks, determine the wireless network API interfaces that need to be called, and obtain historical data on poor-quality wireless network cells through the wireless network API interfaces;

[0039] Based on the knowledge of wireless network quality diagnosis experts, identify historical poor-quality wireless network cell data with anomalies and their corresponding historical processing experience information.

[0040] Abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information are sent to the network optimization expert agent, so that the network optimization expert agent can train the second large network model based on the abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information sent by each poor-quality expert agent.

[0041] Thirdly, this disclosure provides a network optimization expert agent, which includes:

[0042] The compilation and distribution module is used to compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells, and distribute each diagnostic sub-task to the corresponding poor quality expert agent, so that each poor quality expert agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent.

[0043] The receiving and summarizing module, connected to the compilation and distribution module, is used to receive the diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of poor quality wireless network cells based on several diagnostic sub-results and their corresponding chain relationships. The diagnostic results include the root causes of poor quality and suggestions for handling poor quality in the poor quality wireless network cells.

[0044] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0045] Fourthly, this disclosure provides a poor quality expert agent, which includes:

[0046] The receiving task module is used to receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent.

[0047] The task execution module, connected to the task receiving module, is used to obtain diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network according to the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality.

[0048] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0049] Fifthly, this disclosure provides a system for processing poor-quality cells in a wireless network, the system comprising:

[0050] Network optimization expert Agent is used to implement the method for handling poor-quality wireless network cells as described in the first aspect above.

[0051] Several poor-quality expert agents are connected to network optimization expert agents to implement the poor-quality cell processing method for wireless networks as described in the second aspect above.

[0052] Sixthly, this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for processing poor-quality cells in a wireless network as described above.

[0053] This disclosure provides a method, apparatus, and medium for processing poor-quality cells in wireless networks. It employs multi-agent collaboration to provide a highly intelligent and fast network optimization auxiliary tool. Through collaboration between a network optimization expert agent and several poor-quality expert agents, the network optimization expert agents compile diagnostic sub-tasks and their chain relationships. The poor-quality expert agents execute these diagnostic sub-tasks to obtain diagnostic sub-results. Finally, the network optimization expert agents summarize the diagnostic sub-results according to the chain relationships to obtain the root causes of poor-quality cells in the wireless network and suggestions for handling them. This can be used to assist network optimization engineers in completing network optimization work. The efficiency of network optimization is improved by having multiple agents perform diagnoses for each dimension separately, and the accuracy of the final diagnostic result is improved by obtaining the multi-dimensional network poor-quality diagnostic sub-results. Attached Figure Description

[0054] Figure 1 This is a flowchart of a method for processing poor-quality cells in a wireless network according to an embodiment of this disclosure;

[0055] Figure 2 This is a flowchart of another method for processing poor-quality cells in a wireless network according to an embodiment of this disclosure;

[0056] Figure 3 This is a schematic diagram of the structure of a network optimization expert agent according to an embodiment of this disclosure;

[0057] Figure 4 This is a schematic diagram of the structure of a poor quality expert agent according to an embodiment of this disclosure;

[0058] Figure 5 This is a schematic diagram of the structure of a poor-quality wireless network cell processing system according to an embodiment of this disclosure;

[0059] Figure 6 This is a flowchart of another method for processing poor-quality cells in a wireless network according to an embodiment of this disclosure;

[0060] Figure 7 This is an architecture diagram of a poor-quality wireless network cell processing system according to an embodiment of this disclosure;

[0061] Figure 8 This is an example diagram of a poor quality root cause diagnostic chain according to an embodiment of this disclosure;

[0062] Figure 9 This is an example waveform diagram of a cell PRB level according to an embodiment of this disclosure;

[0063] Figure 10 This is a flowchart illustrating the execution method of the execution body according to an embodiment of this disclosure;

[0064] Figure 11 This is a schematic diagram of a page illustrating a quality difference analysis auxiliary tool according to an embodiment of this disclosure;

[0065] Figure 12 This is a page example diagram of a quality difference analysis auxiliary tool according to an embodiment of this disclosure. Detailed Implementation

[0066] To enable those skilled in the art to better understand the technical solutions of this disclosure, the embodiments of this disclosure will be further described in detail below with reference to the accompanying drawings.

[0067] It is understood that the specific embodiments and accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of this disclosure.

[0068] It is understood that, without conflict, the various embodiments and features in the embodiments of this disclosure can be combined with each other.

[0069] It is understood that, for ease of description, only the parts relevant to this disclosure are shown in the accompanying drawings, while parts unrelated to this disclosure are not shown in the drawings.

[0070] It is understood that each module or unit involved in the embodiments of this disclosure may correspond to only one entity structure, or may be composed of multiple entity structures, or multiple modules or units may be integrated into one entity structure.

[0071] It is understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of this disclosure may occur in a different order than that marked in the accompanying drawings.

[0072] It is understood that the flowcharts and block diagrams of this disclosure illustrate the architecture, functions, and operations of possible implementations of systems, apparatuses, devices, and methods according to various embodiments of this disclosure. Each block in a flowchart or block diagram may represent a module, unit, program segment, or code, containing executable instructions for implementing the specified function. Furthermore, each block or combination of blocks in the block diagrams and flowcharts may be implemented using hardware-based devices to implement the specified function, or using a combination of hardware and computer instructions.

[0073] It is understood that the modules and units involved in the embodiments of this disclosure can be implemented by software or by hardware, for example, the modules and units can be located in a processor.

[0074] Example 1:

[0075] like Figure 1 As shown, this disclosure provides a method for processing poor-quality cells in a wireless network. The method is applied to a network optimization expert agent and includes:

[0076] S11. Compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells, distribute each diagnostic sub-task to the corresponding poor quality expert agent, so that each poor quality expert agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent.

[0077] S12. Receive the diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of the poor quality cell of the wireless network based on several diagnostic sub-results and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cell of the wireless network and the suggestions for handling the poor quality.

[0078] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0079] In this embodiment, a multi-agent collaboration approach is adopted to provide a highly intelligent and fast network optimization auxiliary tool. Through collaboration between a network optimization expert agent and several quality-poor expert agents, the network optimization expert agent compiles diagnostic sub-tasks and their chain relationships. The quality-poor expert agents execute the diagnostic sub-tasks to obtain diagnostic sub-results. Finally, the network optimization expert agent summarizes the diagnostic sub-results according to the chain relationships to obtain the root causes of poor network quality and quality-poor handling suggestions for the poor-quality cells. This tool can be used to assist network optimization engineers in completing network optimization work. The efficiency of network optimization is improved by having multiple agents perform diagnoses for each dimension separately, and the accuracy of the final diagnostic result is improved by obtaining the multi-dimensional network quality-poor diagnostic sub-results.

[0080] and Figure 1 The corresponding method applied to the quality-poor expert agent is as follows: Figure 2 As shown, the structure of the corresponding network optimization expert agent is as follows: Figure 3 As shown, the structure of the poor quality expert agent is as follows: Figure 4 As shown, the corresponding module executes the steps labeled accordingly. The following section discusses... Figure 2 The detailed description of the corresponding methods will not be repeated here, as those skilled in the art can easily deduce the actions of the other end based on the correspondence. Furthermore, a wireless network poor-quality cell processing system integrating network optimization expert agents and poor-quality expert agents, such as... Figure 5 As shown, the comprehensive method is as follows: Figure 6 As shown.

[0081] Specifically, this invention provides a method and apparatus for processing poor-quality cells in wireless networks based on multi-agent architecture. "Agent" refers to an AI agent, an intelligent agent driven by a large language model, possessing the ability to autonomously understand, perceive, plan, remember, and use tools, and can automatically execute complex tasks. This embodiment applies multi-agent collaboration to analyze the root causes of poor-quality cells in wireless networks and proposes suggestions for processing and remediation. A poor-quality cell refers to a cell with one or more quality issues in its network communication. These issues involve multiple aspects, such as voice service connection rate, call drop rate, packet loss rate, data service disconnection rate, access success rate, and user-perceived speed, etc.

[0082] Currently, the identification of problem types in poor-quality cells mainly adopts traditional network optimization (or network optimization) methods. Specifically, this involves first collecting key performance indicators (KPIs) for the cell, then network optimization staff using their knowledge and experience to set thresholds for each KPI. Based on these thresholds and the cell's overall KPIs, the type of poor-quality problem is determined. Then, network optimization staff analyze the data, manually troubleshooting various network issues such as coverage, interference, and capacity, and using experience to conduct root cause analysis and develop optimization solutions to improve network performance. However, current practices in wireless network problem root cause analysis largely rely on manual experience combined with various software tools. This approach presents several problems: a shortage of personnel with specialized skills, resulting in a bottleneck in optimization; complex network problems with numerous causes, making problem localization difficult; repetitive problem handling, poor timeliness, and long processing cycles; limitations in network problem discovery, unclear identification of the true pain points, high manpower costs, and low efficiency.

[0083] With the continuous advancement and development of the mobile communication network industry, high-speed and high-quality network applications are placing higher demands on network operations. Network operators aim to "detect problems before users do." Beyond the conventional method of manually analyzing cell KPIs and periodically identifying poor-quality cells according to manually defined rules, designing and applying machine learning-guided methods for locating and analyzing poor-quality cells in mobile communication networks is beneficial for identifying network quality issues. It allows for the prediction of the degree of network quality degradation and whether it will affect user experience. Leveraging machine learning to comprehensively analyze and locate poor-quality cells in mobile communication networks, enabling timely and proactive handling of network problems, and providing users with a better service experience, has significant practical value.

[0084] Therefore, this embodiment proposes a multi-agent intelligent agent concept. Based on a large-model multi-agent architecture, an intelligent agent for analyzing and processing poor-quality wireless cells is constructed. Frontline network optimization engineers present actual poor-quality wireless cell problems in natural language. The overall network optimization expert agent in the multi-agent architecture reads the problem and, as the central hub of the multi-agent system, invokes a long short-term memory framework to understand the intent of the problem. After understanding the intent of analyzing and processing poor-quality wireless cells, it then invokes a thinking framework to plan the poor-quality cell processing task and decompose the task objectives. Finally, the poor-quality expert agent performs diagnostic analysis, which includes multiple dimensions. The single-agent collaboration allows each single agent in each dimension to execute a schedulable toolset to complete sub-tasks, including real-time data acquisition of various key indicators, expert knowledge base cases, and decision-making small models. Through the collaboration between different single agents in the poor quality expert agent, the root cause analysis results and intelligent diagnostic solutions for poor quality cells are obtained. These results are then returned to the network optimization expert agent for aggregation and consideration to generate the final answer, which is then returned to the network optimization engineer. Ultimately, this constructs a full-cycle closed-loop capability for analyzing and processing poor wireless quality cells based on natural language input, improving the intelligence of fault diagnosis, enhancing the work efficiency of front-line network optimization engineers, and reducing network operation and maintenance costs.

[0085] and Figure 5 The corresponding system solution overall design architecture is as follows: Figure 7 As shown, the network optimization expert agent pre-classifies diagnostic results (diagnostic work orders) to train the large language model, making it the network large model in this embodiment. The network optimization expert agent realizes the ability to classify and summarize diagnoses through thinking chains and sub-goal decomposition. It is the central hub for handling poor quality cells in the wireless network. Multiple diagnostic agents (poor quality expert agents) connect to wireless network tools and can call up data such as wireless network indicators, logs, configurations, and reports to realize multi-dimensional diagnosis of poor cell quality problems. It also has the ability to automatically call tools to execute instructions. In practice, users submit questions about poor-quality wireless network services in natural language. The network optimization expert agent reads the question, uses long short-term memory to identify the intent, and then decomposes and plans the task based on this understanding, leveraging its large-scale model thinking chain capabilities. This information is then forwarded to the poor-quality expert agent. Through collaboration among different agents within the poor-quality expert agent, root cause analysis results and intelligent diagnostic solutions are obtained. These are then returned to the network optimization expert agent for aggregation and final answer generation. All information is summarized into two types of instructions: visual instructions and page instructions. The network optimization expert agent executes these instructions and presents the complete answer to the user. The overall design process is as follows: Figure 6 As shown, this will be explained in detail later. Figure 6 Each step in the process.

[0086] In one implementation, S11 involves compiling several diagnostic subtasks and their chain relationships for poor-quality cells in a wireless network, and distributing each diagnostic subtask to the corresponding poor-quality expert Agent. Specifically, this includes:

[0087] The system receives user questions through the first network big model and identifies the user's intent to inquire about cells with poor wireless network quality.

[0088] Based on the statistical results of the indicators of cells with poor wireless network quality, determine the real-time wireless network quality poorness type of the cells with poor wireless network quality;

[0089] Based on the real-time wireless network quality issues, the first network big model uses thought chain question-and-answer prompts to compile several diagnostic sub-tasks and their chain relationships for cells with poor wireless network quality. These diagnostic sub-tasks include at least one of the following: fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis.

[0090] Each diagnostic subtask is assigned to a corresponding quality defect expert agent. The quality defect expert agent includes at least one of the following: fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and hidden fault diagnosis agent.

[0091] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0092] In this embodiment, as Figure 6 As shown, S01: Using the large network model as the interaction entry point, the user's wireless voice service quality issues in the cell are input into the dialog box.

[0093] Users can input their concerns about poor-quality wireless voice services in designated cells into a dialog box using natural language interaction. Table 1 below shows the relevant indicator types and definitions of poor-quality wireless network cells.

[0094] Table 1. Types of relevant indicators and definitions of poor wireless network quality in poor-quality cells.

[0095]

[0096] After defining the poor-quality cells in the table above, users can ask questions about poor-quality cells in the network big model dialog box. The network big model judges the poor-quality cells according to the definitions of different poor-quality types and determines that its poor-quality type is one of "high dropouts", "low call availability", "high interference", "high call dropouts", "high load", "low handover", or "low traffic". For example, the following question can be used: What is the poor quality situation of Shanghai International Exhibition Center West Plaza 1 on October 19? After entering the question into the big model dialog box, step 01 ends and step 02 begins.

[0097] S02: After the user's problem is submitted to the network optimization expert agent through the network big data model, the network optimization expert agent, acting as the central agent, uses long short-term memory to identify and plan the intent of the problem. The user submits a real-world wireless quality poor cell network problem in natural language. The problem is first received by the network optimization expert agent, which, acting as the central agent, uses long short-term memory to identify and plan the intent of the problem, determining which professional fields and related diagnostic agents are involved. Then, it is sent to the poor quality expert agent, which begins to simultaneously call different sub-agents (such as...). Figure 7 The fault agent, parameter agent, coverage agent, quality agent, interference agent, and capacity agent shown are displayed. After completion, step 02 ends and step 03 begins.

[0098] S03: Utilize the network model's thought process capabilities to plan and orchestrate the tasks following intent recognition. The poor-quality expert agent determines which specialized fields and related diagnostic agents need to be used. After the network model receives the intent results from the network optimization expert agent, it sends them to the poor-quality expert agent. This agent, using its thought process capabilities refined through incremental pre-training and SFT instruction fine-tuning, plans the analysis tasks for poor-quality cell problems, primarily decomposing the tasks into the following aspects: 1) Fault diagnosis, 2) Parameter diagnosis, 3) Coverage diagnosis, 4) Quality diagnosis, 5) Interference diagnosis, 6) Capacity diagnosis, and 7) Latent fault diagnosis, etc. Figure 8 As shown, after completion, proceed to step 04.

[0099] S04: The Quality Defect Expert Agent decomposes the quality defect problem into sub-agents at different dimensions. Each sub-agent conducts quality defect analysis through a designated tool and feeds back the analysis and diagnosis results to the Network Optimization Expert Agent for problem diagnosis, classification, and summarization. Note that S04 is the action on the Quality Defect Expert Agent side. After receiving the quality defect problem decomposition task from the Quality Defect Expert, each sub-agent conducts quality defect analysis through a designated tool (such as the API interface (Application Programming Interface) for each sub-task to obtain real-time data), specifically including:

[0100] 1. Fault Agent:

[0101] 1) Use the following tool: base station alarm API interface.

[0102] 2) Analysis process of poor quality issues:

[0103] ① If the conditions in step 1 are met, proceed to step 2, and the fault agent diagnosis ends.

[0104] ② If the conditions in step 1 are not met, proceed to step 3, and the fault agent diagnosis ends.

[0105] Step 1: First, call the base station alarm API interface to check if there are any alarms on the base station that affect normal service communication, such as station outage alarms or cell unavailability alarms.

[0106] Step 2: According to the base station alarm query, the cell was found to be unavailable on [Date] at [Time] in 2024. It is recommended that the maintenance engineer handle the fault and investigate the cause. The fault agent diagnosis is now complete.

[0107] Step 3: After checking the alarms on the base station side, no abnormal alarms were found, and it is not an alarm problem. The case is then transferred to other sub-Agents for processing (other sub-Agents besides the alarm agent). The fault agent diagnosis is now complete.

[0108] Based on the diagnostic analysis results from the API calls made by the faulty Agent, the results are then fed back to the network optimization expert Agent for problem diagnosis, classification, and summarization.

[0109] 2. Parameter Agent:

[0110] 1) Tool for calling: Community CM configuration file API interface (containing timer parameters, handover parameters and other data).

[0111] 2) Analysis process of poor quality issues:

[0112] ① If the conditions in step 1 are met, proceed to step 2, and the Agent parameter diagnosis ends.

[0113] ② If the conditions in step 1 are not met, proceed to step 3, and the Agent parameter diagnosis ends.

[0114] Step 1: First, call the cell CM configuration file API interface to check if there are any unreasonable parameter configuration issues in the cell, such as any parameter exceeding the reasonable range, such as the cell air interface side timer, minimum access level, inter-system handover threshold, etc.

[0115] Step 2: After checking the parameter configuration in the cell's CM configuration file, it was found that the air interface timer T300 configuration of this cell was out of reasonable range, which had a certain impact on the success rate of RRC establishment and data service perception of the cell. It is recommended to reset it to a reasonable range. The parameter Agent diagnosis is now complete.

[0116] Example: Abnormal analysis results in the Agent parameter diagnostic analysis.

[0117] a) T300 = 2000 (reasonable range 0-1000ms), parameter configuration abnormal.

[0118] b) T304 = 500 (reasonable range 500-2000ms), parameter configuration is normal.

[0119] c) N310 = 10 (reasonable range 6-20), parameter configuration is normal.

[0120] d) N311 = 1 (reasonable range 1-2), parameter configuration is normal.

[0121] e) Minimum access level = -126 (reasonable range -110 to -128), parameter configuration is normal.

[0122] f) A2 event RSRP threshold (inter-system data) = -118 (reasonable range -107 to -118), parameter configuration is normal.

[0123] g) Event B2 RSRP threshold 1 (data) = -116 (reasonable range -110 to -118), parameter configuration is normal.

[0124] h) B2 event RSRP threshold 2 (data) = -118 (reasonable range -110 to -118), parameter configuration is normal. Step 3: After checking the parameter configuration in the cell CM configuration file, no abnormal parameter configuration was found. It is not a parameter configuration problem. It is transferred to other sub-Agents for processing (other sub-Agents besides the parameter agent). The parameter agent diagnosis ends.

[0125] Example: In the Agent parameter diagnostic analysis, the analysis results are normal.

[0126] a) T300 = 1000 (reasonable range 0-1000ms), parameter configuration is normal.

[0127] b) T304 = 500 (reasonable range 500-2000ms), parameter configuration is normal.

[0128] c) N310 = 10 (reasonable range 6-20), parameter configuration is normal.

[0129] d) N311 = 1 (reasonable range 1-2), parameter configuration is normal.

[0130] e) Minimum access level = -126 (reasonable range -110 to -128), parameter configuration is normal.

[0131] f) A2 event RSRP threshold (inter-system data) = -118 (reasonable range -107 to -118), parameter configuration is normal.

[0132] g) Event B2 RSRP threshold 1 (data) = -116 (reasonable range -110 to -118), parameter configuration is normal.

[0133] h) B2 event RSRP threshold 2 (data) = -118 (reasonable range -110 to -118), parameter configuration is normal. Based on the diagnostic analysis results of the API interface called by the Agent, the network optimization expert Agent is then fed back to diagnose, classify and summarize the problem.

[0134] 3. Coverage Agent:

[0135] 1) Use the following tool: Community MR Index API interface.

[0136] 2) Analysis process of poor quality issues:

[0137] ① If the conditions in step 1 are met, proceed to step 2, and the Agent diagnosis will end.

[0138] ② If the conditions in step 1 are not met, proceed to step 3, and the Agent diagnosis will end.

[0139] Step 1: First, call the cell MR indicator API interface to check if there are any unreasonable coverage issues in the cell, such as weak coverage ratio, over-coverage rate, near-coverage rate, or any other indicator exceeding the reasonable range.

[0140] Step 2: After checking the MR indicators of the cell, the weak coverage ratio (RSRP <= -105 ratio), or the weak coverage + over-coverage ratio, or the weak coverage + near-coverage ratio of the cell are outside the reasonable range, which has a certain impact on the success rate of RRC establishment and data service perception of the cell. It is recommended to check the cell's engineering parameters and optimize and adjust the RF. The coverage agent diagnosis is now complete.

[0141] Example 1: When only the proportion of weak coverage exceeds a reasonable range

[0142] (1) Weak coverage ratio (RSRP <= -105 ratio) = 12.6% (reasonable range 0-10%), indicator abnormal.

[0143] (2) Total number of sampling points for tadv = 9426

[0144] (3) Number of over-coverage sampling points = 2465

[0145] (4) Over-coverage rate = 16.15% (the reasonable range for TA>1km is 0-25%)

[0146] (5) Number of near-coverage sampling points = 25

[0147] (6) Near coverage rate = 1.28% (reasonable range 0-20%)

[0148] Diagnosis revealed a weak coverage issue in the cell, potentially affecting users with weak signals at the cell's coverage edge, which could impact service experience. Cell parameters were checked.

[0149] (1) Station height = 50 meters

[0150] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0151] For weak coverage, optimization is recommended in the following aspects:

[0152] (1) Downtilt Angle Verification: By verifying the electronic and mechanical downtilt angles of the antenna feeder, the downtilt angle configuration is adjusted appropriately to improve downlink weak coverage.

[0153] (2) Azimuth angle verification: The azimuth angle of the site is reasonably planned and adjusted in combination with the coverage area of ​​the base station.

[0154] (3) Power configuration parameter verification: Based on the capabilities of the base station equipment, assess the maximum supported transmit power and modify the power parameters according to the actual situation.

[0155] (4) Base station mounting height verification: Assess the base station coverage area, verify the antenna mounting height, and improve weak downlink coverage by adjusting the antenna feeder mounting height.

[0156] (5) Site planning assessment: Conduct new site assessments for areas with weak coverage, and improve downlink coverage through construction plans.

[0157] Example 2: When both weak coverage and overcoverage exceed reasonable limits.

[0158] (1) Weak coverage ratio (RSRP <= -105 ratio) = 12.6% (reasonable range 0-10%), indicator abnormal.

[0159] (2) Total number of sampling points for tadv = 9426

[0160] (3) Number of over-coverage sampling points = 2465

[0161] (4) Over-coverage rate = 28.85% (the reasonable range for TA>1km is 0-25%), indicating an abnormal index.

[0162] (5) Number of near-coverage sampling points = 25

[0163] (6) Near coverage rate = 1.28% (reasonable range 0-20%)

[0164] The diagnosis revealed an over-coverage issue in the cell. Excessive coverage caused access problems for users at the coverage edge, impacting service experience. The cell's operational parameters are as follows:

[0165] (1) Station height = 50 meters

[0166] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0167] To address over-coverage, we recommend optimizing the process in the following ways:

[0168] (1) Downtilt Angle Verification: By verifying the electronic and mechanical downtilt angles of the antenna feeder, the downtilt angle configuration is adjusted appropriately to improve downlink weak coverage.

[0169] (2) Power configuration parameter verification: Reduce the maximum transmit power of the cell

[0170] (3) Base station mounting platform verification: Assess the base station coverage area, verify the antenna mounting height, and reduce the antenna mounting method to reduce out-of-area coverage.

[0171] (4) Azimuth angle verification: The azimuth angle of the site is reasonably planned and adjusted in combination with the coverage area of ​​the base station.

[0172] Example 3: When both the weak coverage rate and the near coverage rate exceed a reasonable range.

[0173] (1) Weak coverage ratio (RSRP <= -105 ratio) = 12.6% (reasonable range 0-10%), indicator abnormal.

[0174] (2) Total number of sampling points for tadv = 9426

[0175] (3) Number of over-coverage sampling points = 2465

[0176] (4) Over-coverage rate = 16.15% (the reasonable range for TA>1km is 0-25%)

[0177] (5) Number of near-coverage sampling points = 2005

[0178] (6) Near coverage rate = 21.35% (reasonable range 0-20%), indicator abnormal.

[0179] Diagnosis revealed a near-coverage issue in the community, which negatively impacts user experience. Community service parameters were checked.

[0180] (1) Station height = 50 meters

[0181] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0182] For near-coverage, optimization is recommended in the following aspects:

[0183] (1) Effectively control coverage rationality through RF optimization, and adjust azimuth and downtilt angles to avoid the main direction of antenna and feeder obstruction.

[0184] (2) Conduct an on-site inspection to check whether there are billboards or other physical obstructions such as decorative antennas at the site. If necessary, make rectifications.

[0185] (3) Site relocation: For cases where coverage is too close to the site and cannot be improved through rectification due to severe obstruction, site relocation can be carried out.

[0186] (4) Add a new base station. If the coverage of the new base station cannot be optimized and improved, it is recommended to add a new planned base station in the surrounding area to solve the problem. Step 3: After checking the MR index of the cell, no weak coverage, near coverage or over-coverage problems were found. It is not a coverage abnormality problem. It is transferred to other sub-Agents for processing (other sub-Agents besides the coverage Agent). The diagnosis of the coverage Agent ends.

[0187] Example: In the Agent diagnostic analysis, the analysis results are normal.

[0188] (1) Weak coverage ratio (RSRP <= -105 ratio) = 8.28% (reasonable range 0-10%)

[0189] (2) Total number of sampling points for tadv = 9426

[0190] (3) Number of over-coverage sampling points = 2465

[0191] (4) Over-coverage rate = 16.15% (the reasonable range for TA>1km is 0-25%)

[0192] (5) Number of near-coverage sampling points = 25

[0193] (6) Near coverage rate = 1.28% (reasonable range 0-20%)

[0194] Based on the diagnostic analysis results from the API interface of the coverage agent calling the cell MR index, the results are then fed back to the network optimization expert agent for problem diagnosis, classification, and summarization.

[0195] 4. Quality Agent:

[0196] 1) Use the tool: Community KPI performance index API interface.

[0197] 2) Analysis process of poor quality issues:

[0198] ① If the conditions in step 1 are met, proceed to step 2, and the quality agent diagnosis ends.

[0199] ②If the conditions in step 1 are not met, proceed to step 3, and the quality agent diagnosis ends.

[0200] Step 1: First, call the cell KPI performance indicator API interface to check if there are any unreasonable quality issues in the cell, such as the proportion of sampling points with MRO-SINR≥0, CQI good rate, overlapping coverage ratio, MOD30 interference rate, etc., any of which exceed the reasonable range.

[0201] Step 2: After checking the cell's KPI performance indicators, the data for the following eight scenarios are found to be outside the reasonable range: the percentage of sampling points with MRO-SINR≥0, the percentage of sampling points with MRO-SINR≥0 + CQI good rate, the percentage of sampling points with MRO-SINR≥0 + overlapping coverage ratio, the percentage of sampling points with MRO-SINR≥0 + MOD30 interference rate, the percentage of sampling points with MRO-SINR≥0 + CQI good rate + overlapping coverage ratio, the percentage of sampling points with MRO-SINR≥0 + CQI good rate + MOD30 interference rate, and the percentage of sampling points with MRO-SINR≥0 + overlapping coverage ratio + MOD30 interference rate. These scenarios have a certain impact on the cell's RRC establishment success rate and user service perception. It is recommended to check the cell's operating parameters and power, and optimize and adjust RF and parameters. The quality agent diagnosis is now complete.

[0202] Example 1: When only the proportion of sampling points with MRO-SINR ≥ 0 in a cell exceeds a reasonable range

[0203] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0204] (2) CQI excellent rate = 97.19% (reasonable range 95%~100%)

[0205] (3) Overlap coverage ratio (RSRP = -105) = 14.78% (reasonable range 0% to 20%)

[0206] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 3.39% (reasonable range 0%~10%)

[0207] Diagnosis revealed a low proportion of sampling points with SINR ≥ 0 in the cell, indicating a quality issue that impacts service perception. Cell operational parameters were checked.

[0208] (1) Station height = 50 meters

[0209] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0210] To address the low proportion of sampling points with SINR ≥ 0 in a given cell, optimization and adjustments are recommended in the following aspects:

[0211] (1) Neighbor cell optimization: Optimizing the poor SINR quality caused by the neighbor cell problem.

[0212] (2) Overlapping Coverage Optimization

[0213] (3) MOD30 interference optimization

[0214] (4) Troubleshooting and optimization of interference issues

[0215] Example 2: When the proportion of sampling points with MRO-SINR ≥ 0 and the CQI good / good rate in a cell exceed a reasonable range.

[0216] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0217] (2) CQI good rate = 93.21% (reasonable range 95%~100%), indicator abnormal

[0218] (3) Overlap coverage ratio (RSRP = -105) = 14.78% (reasonable range 0% to 20%)

[0219] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 3.39% (reasonable range 0%~10%)

[0220] The diagnosis revealed a low CQI (Quality, Quality, and Achievement) rate in the residential community, indicating quality-related issues that negatively impact business perception. The community's operational parameters were reviewed as follows:

[0221] (1) Station height = 50 meters

[0222] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0223] To address the low CQI (Quality and Quality Index) rate in the community, the following optimization and adjustments are recommended:

[0224] (1) RF optimization for low CQI cells

[0225] (2) Modify the CQI configuration parameters to improve the configuration, taking into account the specific parameters of different manufacturers.

[0226] (Huawei: CQI reliability optimization switch, PUCCH channel optimization enhancement switch, optimized normalized PDSCH power bias)

[0227] (ZTE: Optimizing CQI false detection algorithm, power bias optimization, and PUCCH channel optimization to improve CQI reliability)

[0228] (NOXN: Enable periodic sub-band CQI reporting, optimize NPDSCEPREOFF, and enable CQI adaptive reporting)

[0229] Example 3: When the ratio of MRO-SINR ≥ 0 in a cell plus the overlap coverage ratio exceeds a reasonable range.

[0230] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0231] (2) CQI excellent rate = 96.28% (reasonable range 95%~100%)

[0232] (3) Overlap coverage ratio (RSRP = -105) = 24.78% (reasonable range 0% to 20%), indicator abnormal.

[0233] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 3.39% (reasonable range 0%~10%)

[0234] Diagnosis revealed a high degree of overlap in cell coverage, indicating quality issues that negatively impact service experience. Cell parameters were checked.

[0235] (1) Station height = 50 meters

[0236] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0237] To address the high overlap in coverage between residential areas, we recommend optimizing and adjusting the following aspects:

[0238] (1) Adjust the antenna downtilt angle: optimize the antenna coverage, reduce signal interference, and improve spectrum efficiency.

[0239] (2) Adjust the antenna azimuth angle: improve signal directivity, improve signal quality, enhance system performance, and reduce overlap coverage.

[0240] (3) Adjust the antenna height: improve signal propagation, ensure that the coverage area matches the service cell, and reduce the impact of obstruction.

[0241] (4) Site rectification and relocation: For unreasonable site locations, rooftop rectification or site relocation will be carried out for communities that are interfering with surrounding sites.

[0242] (5) Change the site frequency band: avoid frequency interference, improve data transmission rate, and optimize network performance.

[0243] (6) Adjust cell reference power: balance coverage, optimize signal strength, and reduce interference.

[0244] Example 4: When the proportion of sampling points with MRO-SINR ≥ 0 and the MOD30 interference rate exceed a reasonable range.

[0245] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0246] (2) CQI excellent rate = 96.28% (reasonable range 95%~100%)

[0247] (3) Overlap coverage ratio (RSRP = -105) = 14.78% (reasonable range 0% to 20%)

[0248] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 13.39% (reasonable range 0%~10%), the index is abnormal.

[0249] Diagnosis revealed that cell MOD30 has a high interference rate, indicating a quality issue that negatively impacts service experience. Cell parameters were checked.

[0250] (1) Station height = 50 meters

[0251] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0252] To address the high interference rate of MOD30 in the cell, we recommend optimizing and adjusting the following aspects:

[0253] (1) PCI replanning: Based on the PCI planning data of the entire network, rationally plan the PCI of the cell to ensure PCI reuse distance and MOD value interference.

[0254] (2) Antenna adjustment: Adjust the antennas of cells with the same MOD value to avoid interference between cells with the same MOD value.

[0255] (3) Power modification: Reduce the transmission power of the interfering cell and reduce the level of the interfering signal.

[0256] Example 5: When the percentage of sampling points with MRO-SINR ≥ 0, the CQI good / good rate, and the overlap coverage ratio of a cell exceed a reasonable range.

[0257] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0258] (2) CQI good rate = 92.28% (reasonable range 95%~100%), indicator abnormal

[0259] (3) Overlap coverage ratio (RSRP = -105) = 24.78% (reasonable range 0% to 20%), indicator abnormal.

[0260] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 8.39% (reasonable range 0%~10%), the index is abnormal.

[0261] Diagnosis revealed that the cell had a low CQI good / excellent rate and a high overlap in coverage, indicating quality issues that negatively impact service experience. Cell parameters were checked.

[0262] (1) Station height = 50 meters

[0263] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0264] To address the low CQI good / excellent rate in the community, we recommend optimizing and adjusting the following aspects:

[0265] (1) RF optimization for low CQI cells

[0266] (2) Modify the CQI configuration parameters to improve the configuration, taking into account the specific parameters of different manufacturers.

[0267] (Huawei: CQI reliability optimization switch, PUCCH channel optimization enhancement switch, optimized normalized PDSCH power bias)

[0268] (ZTE: Optimizing CQI false detection algorithm, power bias optimization, and PUCCH channel optimization to improve CQI reliability)

[0269] (NOXN: Enable periodic sub-band CQI reporting, optimize NPDSCEPREOFF, and enable CQI adaptive reporting)

[0270] To address the high overlap in coverage between residential areas, we recommend optimizing and adjusting the following aspects:

[0271] (1) Adjust the antenna downtilt angle: optimize the antenna coverage, reduce signal interference, and improve spectrum efficiency.

[0272] (2) Adjust the antenna azimuth angle: improve signal directivity, improve signal quality, enhance system performance, and reduce overlap coverage.

[0273] (3) Adjust the antenna height: improve signal propagation, ensure that the coverage area matches the service cell, and reduce the impact of obstruction.

[0274] (4) Site rectification and relocation: For unreasonable site locations, rooftop rectification or site relocation will be carried out for communities that are interfering with surrounding sites.

[0275] (5) Change the site frequency band: avoid frequency interference, improve data transmission rate, and optimize network performance.

[0276] (6) Adjust cell reference power: balance coverage, optimize signal strength, and reduce interference.

[0277] Example 6: When the proportion of sampling points with MRO-SINR ≥ 0, the CQI good rate, and the MOD30 interference rate exceed a reasonable range.

[0278] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0279] (2) CQI good rate = 92.28% (reasonable range 95%~100%), indicator abnormal

[0280] (3) Overlap coverage ratio (RSRP = -105) = 14.78% (reasonable range 0% to 20%)

[0281] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 12.39% (reasonable range 0%~10%), the index is abnormal.

[0282] Diagnosis revealed that the cell had a low CQI good rate and a high MOD30 interference rate, indicating quality issues that would negatively impact service experience. Cell parameters were checked.

[0283] (1) Station height = 50 meters

[0284] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0285] To address the low CQI good / excellent rate in the community, we recommend optimizing and adjusting the following aspects:

[0286] (1) RF optimization for low CQI cells

[0287] (2) Modify the CQI configuration parameters to improve the configuration, taking into account the specific parameters of different manufacturers.

[0288] (Huawei: CQI reliability optimization switch, PUCCH channel optimization enhancement switch, optimized normalized PDSCH power bias)

[0289] (ZTE: Optimizing CQI false detection algorithm, power bias optimization, and PUCCH channel optimization to improve CQI reliability)

[0290] (NOXN: Enable periodic sub-band CQI reporting, optimize NPDSCEPREOFF, and enable CQI adaptive reporting)

[0291] To address the high interference rate of MOD30 in the cell, we recommend optimizing and adjusting the following aspects:

[0292] (1) PCI replanning: Based on the PCI planning data of the entire network, rationally plan the PCI of the cell to ensure PCI reuse distance and MOD value interference.

[0293] (2) Antenna adjustment: Adjust the antennas of cells with the same MOD value to avoid interference between cells with the same MOD value.

[0294] (3) Power modification: Reduce the transmission power of the interfering cell and reduce the level of the interfering signal.

[0295] Example 7: When the percentage of sampling points + overlapping coverage ratio + MOD30 interference rate of a cell with MRO-SINR ≥ 0 exceeds a reasonable range.

[0296] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0297] (2) CQI excellent rate = 98.28% (reasonable range 95%~100%)

[0298] (3) Overlap coverage ratio (RSRP = -105) = 24.78% (reasonable range 0% to 20%), indicator abnormal.

[0299] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 12.39% (reasonable range 0%~10%), the index is abnormal.

[0300] Diagnosis revealed that cell MOD30 has a high interference rate and a high overlap coverage ratio, indicating quality issues that negatively impact service experience. Cell parameters were checked.

[0301] (1) Station height = 50 meters

[0302] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0303] To address the high interference rate of MOD30 in the cell, we recommend optimizing and adjusting the following aspects:

[0304] (1) PCI replanning: Based on the PCI planning data of the entire network, rationally plan the PCI of the cell to ensure PCI reuse distance and MOD value interference.

[0305] (2) Antenna adjustment: Adjust the antennas of cells with the same MOD value to avoid interference between cells with the same MOD value.

[0306] (3) Power modification: Reduce the transmission power of the interfering cell and reduce the level of the interfering signal.

[0307] To address the high overlap in coverage between residential areas, we recommend optimizing and adjusting the following aspects:

[0308] (1) Adjust the antenna downtilt angle: optimize the antenna coverage, reduce signal interference, and improve spectrum efficiency.

[0309] (2) Adjust the antenna azimuth angle: improve signal directivity, improve signal quality, enhance system performance, and reduce overlap coverage.

[0310] (3) Adjust the antenna height: improve signal propagation, ensure that the coverage area matches the service cell, and reduce the impact of obstruction.

[0311] (4) Site rectification and relocation: For unreasonable site locations, rooftop rectification or site relocation will be carried out for communities that are interfering with surrounding sites.

[0312] (5) Change the site frequency band: avoid frequency interference, improve data transmission rate, and optimize network performance.

[0313] (6) Adjust cell reference power: balance coverage, optimize signal strength, and reduce interference.

[0314] Example 8: When the percentage of sampling points with MRO-SINR ≥ 0, the overlap coverage ratio, the MOD30 interference rate, and the CQI good / good rate of a cell exceed a reasonable range.

[0315] (1) The proportion of sampling points with MRO-SINR ≥ 0 was 88.52% (reasonable range 90%~100%), indicating an abnormal index.

[0316] (2) CQI good rate = 92.28% (reasonable range 95%~100%), indicator abnormal

[0317] (3) Overlap coverage ratio (RSRP = -105) = 24.78% (reasonable range 0% to 20%), indicator abnormal.

[0318] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 12.39% (reasonable range 0%~10%), the index is abnormal.

[0319] Diagnosis revealed that the cell had a low CQI good / excellent rate, a high MOD30 interference rate, and a high overlap coverage ratio, indicating quality issues that would negatively impact service experience. Cell parameters were checked.

[0320] (1) Station height = 50 meters

[0321] (2) Mechanical tilt angle = 2 degrees, electronic tilt angle = 0 degrees, azimuth angle = 60 degrees

[0322] To address the low CQI good / excellent rate in the community, we recommend optimizing and adjusting the following aspects:

[0323] (1) RF optimization for low CQI cells

[0324] (2) Modify the CQI configuration parameters to improve the configuration, taking into account the specific parameters of different manufacturers.

[0325] (Huawei: CQI reliability optimization switch, PUCCH channel optimization enhancement switch, optimized normalized PDSCH power bias)

[0326] (ZTE: Optimizing CQI false detection algorithm, power bias optimization, and PUCCH channel optimization to improve CQI reliability)

[0327] (NOXN: Enable periodic sub-band CQI reporting, optimize NPDSCEPREOFF, and enable CQI adaptive reporting)

[0328] To address the high interference rate of MOD30 in the cell, we recommend optimizing and adjusting the following aspects:

[0329] (1) PCI replanning: Based on the PCI planning data of the entire network, rationally plan the PCI of the cell to ensure PCI reuse distance and MOD value interference.

[0330] (2) Antenna adjustment: Adjust the antennas of cells with the same MOD value to avoid interference between cells with the same MOD value.

[0331] (3) Power modification: Reduce the transmission power of the interfering cell and reduce the level of the interfering signal.

[0332] To address the high overlap in coverage between residential areas, we recommend optimizing and adjusting the following aspects:

[0333] (1) Adjust the antenna downtilt angle: optimize the antenna coverage, reduce signal interference, and improve spectrum efficiency.

[0334] (2) Adjust the antenna azimuth angle: improve signal directivity, improve signal quality, enhance system performance, and reduce overlap coverage.

[0335] (3) Adjust the antenna height: improve signal propagation, ensure that the coverage area matches the service cell, and reduce the impact of obstruction.

[0336] (4) Site rectification and relocation: For unreasonable site locations, rooftop rectification or site relocation will be carried out for communities that are interfering with surrounding sites.

[0337] (5) Change the site frequency band: avoid frequency interference, improve data transmission rate, and optimize network performance.

[0338] (6) Adjust cell reference power: balance coverage, optimize signal strength, and reduce interference.

[0339] Step 3: After checking the cell performance KPI indicators, no issues were found such as low proportion of SINR≥0 sampling points, low CQI good rate, high MOD30 interference rate, or high overlap coverage ratio. These issues are not considered quality anomalies and are therefore transferred to other sub-Agents for processing (sub-Agents other than the quality agent). The quality agent diagnosis is now complete.

[0340] Example: In the quality agent diagnostic analysis, the analysis results are normal.

[0341] (1) The proportion of sampling points with MRO-SINR≥0 was 98.52% (reasonable range 90%~100%).

[0342] (2) CQI excellent rate = 97.28% (reasonable range 95%~100%)

[0343] (3) Overlap coverage ratio (RSRP = -105) = 14.78% (reasonable range 0% to 20%)

[0344] (4) Number of MOD30 interference sampling points = 1280, MOD30 interference rate = 2.39% (reasonable range 0%~10%)

[0345] Based on the diagnostic analysis results obtained by the Quality Agent calling the API interface of the cell performance KPI indicators, the results are then fed back to the Network Optimization Expert Agent for problem diagnosis, classification, and summarization.

[0346] 5. Interference Agent:

[0347] 1) Use the following tools: the uplink interference diagnosis AI mini-model API interface already deployed in the system.

[0348] 2) Interference problem analysis process:

[0349] ① If the conditions in step 1 are met, proceed to step 2, and the interference agent diagnosis ends.

[0350] ② If the conditions in step 1 are not met, proceed to step 3, and the interference agent diagnosis ends.

[0351] Step 1: First, call the uplink interference diagnosis AI mini-model API interface that has been deployed in the system to check whether there is an uplink interference problem in the cell, such as the average interference noise power per prb, uplink interference type, or any other indicator that exceeds the reasonable range.

[0352] Step 2: After checking the results of the AI ​​mini-model for uplink interference diagnosis in the cell, the data for five scenarios—average per prb interference noise power, uplink interference type = blocking interference, uplink interference type = narrowband interference, uplink interference type = spurious interference, and uplink interference type = external interference—exceeded the reasonable range. This has a certain impact on the success rate of RRC establishment and user service perception. It is recommended to conduct uplink interference investigation in the cell based on the different interference type results diagnosed by the AI ​​mini-model. The interference agent diagnosis is now complete.

[0353] Example 1: When only the average interference noise power per prb in the cell exceeds the reasonable range

[0354] (1) Average interference noise power per prb (dBm) = -98dBm (reasonable range <= -105dBm), the indicator is abnormal.

[0355] (2) Uplink interference type = None

[0356] Diagnosis revealed uplink interference in the cell, impacting service experience. Hourly PRB-level waveforms for this cell were retrieved (e.g.,...). Figure 9 (As shown, this is just an example).

[0357] It is recommended to conduct a frequency scan on-site to identify the source of interference, eliminate the interference, and then observe whether the wireless connection rate index returns to normal.

[0358] Example 2: When the average interference noise power per prb in the cell exceeds the reasonable range, and the uplink interference type equals blocking interference.

[0359] (1) Average interference noise power per prb (dBm) = -98dBm (reasonable range <= -105dBm), the indicator is abnormal.

[0360] (2) Uplink interference type = blocking interference

[0361] The diagnosis revealed an uplink interference issue in the cell, which is classified as congestion interference and has a certain impact on service perception. The hourly PRB-level waveform diagram for this cell was retrieved.

[0362] We recommend following these steps to troubleshoot interference:

[0363] (1) Install filters of the corresponding frequency band on the base station to reduce interference.

[0364] (2) Increase the isolation between the two systems, such as raising the antenna height of the interference source base station or the interfered base station to change it from horizontal isolation to vertical isolation (generally, the vertical isolation is more than 10dB greater than the horizontal isolation).

[0365] (3) Check the anti-interference capability of the equipment in the affected cell and replace the affected equipment with equipment with stronger anti-blocking capability.

[0366] Example 3: When the average interference noise power per prb in the cell exceeds the reasonable range, and the uplink interference type equals spurious interference.

[0367] (1) Average interference noise power per prb (dBm) = -98dBm (reasonable range <= -105dBm), the indicator is abnormal.

[0368] (2) Uplink interference type = spurious interference

[0369] The diagnosis revealed an uplink interference issue in the cell, which is classified as spurious interference and has a certain impact on service perception. The hourly PRB-level waveform diagram for this cell was retrieved.

[0370] We recommend following these steps to troubleshoot interference:

[0371] (1) By increasing the system isolation between the TD-LTE base station antenna and the interference source base station antenna, the interference can be reduced. Generally, horizontal isolation can be changed to vertical isolation.

[0372] (2) Reduce spurious interference by installing bandpass filters at the base station of the interference source.

[0373] Example 4: When the average interference noise power per prb in the cell exceeds the reasonable range, and the uplink interference type is equal to narrowband interference.

[0374] (1) Average interference noise power per prb (dBm) = -98dBm (reasonable range <= -105dBm), the indicator is abnormal.

[0375] (2) Uplink interference type = narrowband interference

[0376] The diagnosis revealed an uplink interference issue in the cell, which is classified as narrowband interference and has a certain impact on service perception. Hourly PRB-level waveform diagrams for this cell were retrieved.

[0377] We recommend following these steps to troubleshoot interference:

[0378] (1) Frequency hopping technology is used to dynamically change the operating frequency in order to avoid interference sources with fixed frequencies.

[0379] (2) Use direct sequence spread spectrum (DSSS) or frequency hopping spread spectrum (FHSS) techniques to improve the signal’s anti-interference capability.

[0380] (3) Optimize the base station layout and antenna orientation to reduce sensitivity to interference sources.

[0381] (4) Adjust the transmit power and receive sensitivity to adapt to the interference environment.

[0382] Example 5: When the average interference noise power per prb in the cell exceeds the reasonable range, and the uplink interference type equals external interference.

[0383] (1) Average interference noise power per prb (dBm) = -98dBm (reasonable range <= -105dBm), the indicator is abnormal.

[0384] (2) Uplink interference type = external interference

[0385] The diagnosis revealed an uplink interference issue in the cell, which is an external interference that affects service perception. Hourly PRB-level waveforms for this cell were retrieved.

[0386] We recommend following these steps to troubleshoot interference:

[0387] (1) Most of the external interference is persistent, so the source of interference can be found relatively easily by scanning the frequency and can be directly coordinated to shut it down.

[0388] (2) For non-continuous interference sources, find the pattern of interference occurrence based on hourly indicators, and conduct on-site investigations during the period when interference occurs. For example, the time when the interference device is turned on during school exams coincides with the time when students go to and from get off work.

[0389] Some methods and techniques for troubleshooting external interference:

[0390] Method 1: Determine the direction and render the background noise value data into an intuitive heat map to guide field engineers in identifying the location and direction of interference.

[0391] Method 2: Use drones to quickly and effectively locate the position and direction of interference;

[0392] Method 3: Conduct a thorough manual search to determine the final location of the interference;

[0393] Method 4: Three-point positioning method

[0394] Step 1: Identify several high points in the interference area. The first point is usually the building where the interfered base station is located. Use a spectrum analyzer to perform a 360-degree test to find the direction of the strongest interference.

[0395] Step 2: Move forward along the direction of the strongest interference to find the second point, and perform a 360-degree frequency sweep. Compare the interference intensity and direction with the first point to determine the approximate location of the interference source.

[0396] Step 3: Perform frequency sweeps at different high points in sequence until the source of interference is identified.

[0397] Selection criteria: Choose buildings with a wide field of view and high elevation for frequency scanning to facilitate observation of the surrounding environment, check for the presence of antennas, amplifiers, and other devices, and avoid interference that could obstruct the view and affect judgment.

[0398] Step 3: After checking the cell performance KPI indicators, no issues were found regarding the average interference noise power per prb or the type of interference. Since it is not an uplink interference anomaly, the case is transferred to other sub-agents for processing (other sub-agents besides the interference agent). The interference agent diagnosis is now complete.

[0399] Example: In the interference agent diagnostic analysis, the analysis results are normal.

[0400] (1) Average interference noise power per prb (dBm) = -110dBm (reasonable range <= -105dBm)

[0401] (2) Uplink interference type = None

[0402] Based on the diagnostic analysis results of the interference agent calling the API interface of the uplink interference mini-model, the results are then fed back to the network optimization expert agent for problem diagnosis, classification, and summarization.

[0403] 6. Capacity Agent:

[0404] 1) Use the tool: Community KPI performance index API interface.

[0405] 2) Capacity problem analysis process:

[0406] ① If the conditions in step 1 are met, proceed to step 2, and the capacity agent diagnosis ends.

[0407] ②If the conditions in step 1 are not met, proceed to step 3, and the capacity agent diagnosis ends.

[0408] Step 1: First, call the cell KPI performance indicator API interface to check if there is a problem of insufficient capacity load in the cell, such as uplink PRB utilization rate, downlink PRB utilization rate, or any other indicator exceeding the reasonable range.

[0409] Step 2: After checking the KPI performance indicators of the cell, it was found that the uplink PRB utilization rate, uplink PRB utilization rate, or uplink and downlink PRB utilization rate of the cell exceeded the reasonable range. This has a certain impact on the success rate of RRC establishment and user service perception. It is recommended to check the cell's engineering parameters and load balancing parameters and optimize and adjust the RF and parameters. The capacity agent diagnosis is now complete.

[0410] Example: When the uplink PRB utilization rate, downlink PRB utilization rate, or both uplink and downlink PRB utilization rates of a cell exceed the reasonable range.

[0411] (1) Uplink PRB utilization = 72% (reasonable range 0-60%), indicator abnormal (take the moment of maximum 5G total traffic in the entire network over 24 hours) or downlink PRB utilization = 85% (reasonable range 0-80%), indicator abnormal (take the moment of maximum 5G total traffic in the entire network over 24 hours) or uplink PRB utilization = 62% (reasonable range 0-60%) and downlink PRB utilization = 87% (reasonable range 0-80%), indicator abnormal.

[0412] (2)arfcn=636664

[0413] (3) Average number of 5G users = 826

[0414] (4) 5G data traffic (5G uplink traffic + 5G downlink traffic) = 928.5Gb

[0415] The diagnosis revealed that the community is experiencing excessive load, which is negatively impacting service perception. Optimization and adjustments are recommended in the following areas:

[0416] (1) Check the load balancing parameters and configure them appropriately.

[0417] (2) RF optimization and adjustment

[0418] (3) Expansion of cells with the same or different frequencies

[0419] (4) The newly built base station shares the load.

[0420] Step 3: After checking the cell performance KPI indicators, no issues were found where the uplink PRB utilization rate or downlink PRB utilization rate exceeded the reasonable range. This is not a capacity load anomaly issue. The case is then transferred to other sub-Agents for processing (sub-Agents other than the capacity agent). The capacity agent diagnosis is now complete.

[0421] Example: In the capacity agent diagnostic analysis, the analysis results are normal.

[0422] (1) Uplink PRB utilization rate = 12% (reasonable range 0-60%) (taken at the moment of maximum 5G total traffic in the entire network over 24 hours) or downlink PRB utilization rate = 25% (reasonable range 0-80%) (taken at the moment of maximum 5G total traffic in the entire network over 24 hours) or uplink PRB utilization rate = 8% (reasonable range 0-60%) and downlink PRB utilization rate = 17% (reasonable range 0-80%).

[0423] (2)arfcn=636664

[0424] (3) Average number of 5G users = 16

[0425] (4) 5G data traffic (5G uplink traffic + 5G downlink traffic) = 2.8Gb

[0426] Based on the diagnostic analysis results obtained by the capacity agent calling the API interface of the cell KPI performance indicators, the results are then fed back to the network optimization expert agent for problem diagnosis, classification and summarization.

[0427] 7. Switch Agent:

[0428] 1) Use the tool: Community KPI performance index API interface.

[0429] 2) Switching to the problem analysis process:

[0430] ① If the conditions in step 1 are met, proceed to step 2 and the Agent switching diagnosis ends.

[0431] ② If the conditions in step 1 are not met, proceed to step 3 and the Agent switching diagnosis ends.

[0432] Step 1: First, call the cell KPI performance indicator API interface to check if the cell has a high handover failure rate, such as the success rate of handover within the 5G system, the success rate of handover between 5G systems, the success rate of handover between 5G systems, the success rate of handover between 5G frequencies, or the success rate of handover between 5G frequencies, if any of these indicators exceeds a reasonable range.

[0433] Step 2: After checking the KPI performance indicators of the cell, it was found that the success rate of 5G intra-system handover, 5G inter-system handover, 5G intra-frequency handover, or 5G inter-frequency handover in any single case or combination of cases exceeded the reasonable range, which has a certain impact on the user service experience of the cell. It is recommended to check the configuration of neighboring cells and optimize and adjust the RF and handover parameters. The handover agent diagnosis is now complete.

[0434] Example: When any single or combined situation, such as the success rate of handover within a 5G system, the success rate of handover between 5G systems, the success rate of handover within the same frequency, or the success rate of handover between different frequencies, exceeds a reasonable range,

[0435] (1) 5G system intra-system handover success rate = 92.35% (reasonable range 98-100%), indicator abnormal.

[0436] (2) 5G inter-system handover success rate = 96.45% (reasonable range 98-100%), indicator abnormal.

[0437] (3) 5G frequency handover success rate = 97.2% (reasonable range 98-100%), indicator abnormal.

[0438] (4) 5G inter-frequency handover success rate = 91.48% (reasonable range 98-100%), indicator abnormal.

[0439] Diagnosis indicates that the cell's handover success rate was low during this period. It is recommended to check the cell's handover parameter configuration and neighbor cell configuration.

[0440] Neighboring cell configuration verification:

[0441] (1) Verification of missing neighboring cell configuration issues: Timely replenishment of neighboring cells.

[0442] (2) Verification of neighbor cell mismatch issues: This mainly includes verification of external neighbor cell frequency points, PCI, cell identifier, TAC, and PLMN configuration.

[0443] (3) Verification of Neighbor Cell Over-configuration Issues: Based on the model of the community's control board, obtain the maximum number of neighbor cells supported by the main control board, verify the configured and planned neighbor cells, and reasonably add or delete neighbor cells.

[0444] Neighbor class parameter optimization:

[0445] (1) NR_NR Neighbor Cell Self-Configuration Switch (1(ON))

[0446] (2) NR_NR Neighbor cell self-deletion switch (1(ON))

[0447] (3) NR self-deletion cycle (days) 1 (day)

[0448] (4) NR Neighbor Cell Self-Configuration Controlled Mode Switch 0 (OFF)

[0449] (5) In NR neighbor cell self-configuration controlled mode, the switch for allowing unconfirmed neighbor cells to switch off is 0 (OFF).

[0450] (6) NR_EUTRAN Neighbor Cell Self-Configuration Switch 0 (OFF)

[0451] Step 3: After checking the cell performance KPI indicators, no issues were found where the 5G intra-system handover success rate, 5G inter-system handover success rate, 5G intra-frequency handover success rate, or 5G inter-frequency handover success rate exceeded the reasonable range. Therefore, it is not a capacity handover anomaly and is transferred to other sub-Agents for processing (other sub-Agents besides the handover agent). The handover agent diagnosis is now complete.

[0452] Example: During agent switching diagnostic analysis, the analysis results are normal.

[0453] (1) 5G system intra-system handover success rate = 98.35% (reasonable range 98-100%), the indicator is normal.

[0454] (2) 5G inter-system handover success rate = 99.45% (reasonable range 98-100%), the indicator is normal.

[0455] (3) 5G frequency handover success rate = 99.2% (reasonable range 98-100%), the indicator is normal.

[0456] (4) 5G inter-frequency handover success rate = 99.48% (reasonable range 98-100%), the indicator is normal.

[0457] Based on the diagnostic analysis results of the API interface for calling the cell's KPI performance indicators by switching agents, the results are then fed back to the network optimization expert agent for problem diagnosis, classification, and summarization.

[0458] 8. Latent Fault Agent:

[0459] 1) Use the following API interfaces: Cell Fault, MR, and KPI performance metrics.

[0460] 2) Analysis process of latent faults:

[0461] ① If the conditions in step 1 are met, proceed to step 2, and the latent fault agent diagnosis ends.

[0462] ② If the conditions in step 1 are not met, proceed to step 3, and the latent fault agent diagnosis ends.

[0463] Step 1: First, call the base station alarm API interface, cell KPI performance indicator API interface, and cell MR indicator API interface to check whether there are problems such as base station failure, coverage, interference, quality, capacity, and handover in the cell.

[0464] Step 2: After checking the base station alarms, cell KPI performance indicators, and cell MR indicators, the data for any single or combined situation of fault, coverage, interference, quality, capacity, and handover are normal, but the cell is still a poor quality cell. It is recommended to conduct a hidden fault investigation of the cell. The hidden fault agent diagnosis is now complete.

[0465] Example: When querying base station alarms, cell KPI performance indicators, and cell MR indicators, if the data for any single or combined situation such as fault, coverage, interference, quality, capacity, and handover are all normal, it is recommended to check one by one according to the following steps:

[0466] (1) If the RRU software crashes, it is recommended to reset the RRU in the background and then observe the indicators.

[0467] (2) If the problem persists after resetting, the antenna system needs to be inspected on-site, including whether the antenna placement is reasonable, whether the waterproof tape on the antenna feeder head is damaged, and whether the feeder is damaged.

[0468] (3) Replace or exchange the antenna feeder.

[0469] (4) If no hidden fault is found during on-site investigation, it is recommended to use the end-to-end analysis method to obtain relevant indicators of DNS, CDN, core network and SP server side, and analyze them segment by segment to find the problem above the wireless air interface side.

[0470] (5) Check the base station terminal compatibility parameters and change the base station SRS reconfiguration mode switch to 0 (OFF).

[0471] Based on the results of the latent fault agent's calls to the base station alarm API interface, cell KPI performance indicator API interface, and cell MR indicator API interface, the results are then fed back to the network optimization expert agent for problem diagnosis, classification, and summarization, and the latent fault agent diagnosis ends.

[0472] Step 3: After querying the alarms on the base station side, the cell KPI performance indicators, and the cell MR indicators, if any single or combined situation of fault, coverage, interference, quality, capacity, handover, etc., of the cell is found to be abnormal, it does not belong to the latent fault problem, and is transferred to other sub-Agents for processing (other sub-Agents besides the latent fault capsule Agent). The latent fault agent diagnosis ends.

[0473] After completion, proceed to step 05.

[0474] In one embodiment, in S12, the diagnostic results of a poor-quality wireless network cell are obtained based on several diagnostic sub-results and their corresponding chain relationships. The diagnostic results include the root causes of the poor-quality wireless network cell and suggestions for handling the poor quality, specifically including:

[0475] Obtain several diagnostic sub-tasks and their chain relationships, along with the corresponding thought chain question-and-answer prompts and their order. Input the several diagnostic sub-results into the question template of the second network model in order to obtain diagnostic results including the root causes of poor wireless network quality and suggestions for handling poor quality. The diagnostic sub-results include abnormal real-time data on poor wireless network quality cells and their corresponding historical handling experience information.

[0476] Output voice and / or text results of the root causes of poor wireless network quality and suggestions for handling poor quality cells; display poor wireless network quality cells on a map and mark their root causes; generate page instructions to repair the root causes of poor quality based on the suggestions for handling poor quality; and execute the page instructions automatically or when triggered by the user to repair the poor wireless network quality cells.

[0477] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0478] In this embodiment, as Figure 6 S05 shown: The network optimization expert agent comprehensively summarizes and processes the diagnostic results, obtaining two types of instructions: visual instructions and page instructions. The executor executes the instructions after receiving them (e.g., Figure 7 (As shown), the complete answer is finally presented to the user.

[0479] like Figure 7 As shown, the Network Optimization Expert Agent also acts as a central induction mechanism. Based on the thought process and combined with pre-stored knowledge (which can be implemented in the form of a Network Optimization Expert Knowledge Base), it calls preset problem templates. For example, based on the following diagnostic sub-results, it summarizes the root causes of poor network quality in the poor-quality cells. The second network model of the Network Optimization Expert Agent has been pre-trained and can summarize the root causes of poor quality based on the diagnostic sub-results and their order filled in in the problem template. In the same way, effective suggestions for handling poor quality can be summarized, such as prioritizing the repair of certain abnormal causes that cause poor network quality or setting their repair order.

[0480] In such Figure 7In the overall architecture shown, the executor is responsible not only for using external tools but also for executing client commands. After identifying and planning the user's intent, the network optimization expert agent identifies and plans the processing results, determines the type and object of the operation involved, extracts the unique identifier of the object from the command library, and processes all data according to the format agreed upon with the command execution SDK (Software Development Kit), resulting in two types of command data: page commands and visual commands, such as... Figure 10 As shown.

[0481] Page instructions are the specific steps that the intelligent agent plans to perform on behalf of the user on the existing system page. After the client receives the user's instructions, it uses the page instructions to execute the SDK tool ("to complete the user's query / operation goal through a question"). According to the received page instructions, the SDK tool executes them in the order of the instructions. There is an error correction mechanism to ensure the correct and complete execution of the instructions. If the instruction execution fails, the error correction mechanism will use a backup plan to correct the error and execute the instruction, ensuring that every instruction can be executed completely.

[0482] Visualization commands are used to visualize the content of large-scale intelligent agent planning and summarization on the client side. The client extracts the corresponding components from the visualization component library according to the visualization type, inputs the visualization content, and forms a user-friendly visualization page.

[0483] After all instructions are executed, the client provides feedback to the network optimization expert agent, including the operation results and any anomalies. The network optimization expert agent learns from this feedback and continuously optimizes the accuracy of operation steps and intent recognition. Simultaneously, this process ensures that the entire workflow complies with data protection regulations and business compliance requirements, protecting user data security. All instruction types and operation targets are stored in the instruction database.

[0484] Command Library: This includes command operation types, operable objects, and visual components. The operable objects in the command library contain all operable objects on the system pages. Initially, all operable objects were stored in the command library using automated tools to assist manual input. Each operable object has a precise semantic mapping with the business operation process or function name, facilitating intent recognition by the intelligent agent.

[0485] After the network optimization expert agent plans the operation steps (detailed down to each specific action), it performs intent recognition based on specific terms within those steps, queries the command library to retrieve the unique object identifier corresponding to that semantic meaning, assembles it into a data format agreed upon with the client, and returns it to the client. Upon receiving the command data, if the command is a page command, the client will locate the target object on the page based on the unique object identifier and execute the corresponding action on the target object according to the agent's operation type.

[0486] The instruction library contains two main categories of instructions: page instructions and visual instructions.

[0487] Page instructions are instructions given by the agent to the user to operate on the page. The entire instruction consists of agent operation type (required) + agent operation object (required) + agent content (optional). Among them, agent operation type and agent operation object are required, while agent content is optional.

[0488] Proxy operation types include: 1) Click: Using the mouse to click a specific object on the webpage, including left-click, right-click, and double-click. 2) Text Fill: Entering text in an editable control, mimicking keyboard behavior. 3) Copy. 4) Cut. 5) Paste. 6) Get Text: Retrieving the content of the manipulated object. 7) Get Object: Retrieving the manipulated object. 8) Scroll: Scrolling a control item vertically or horizontally, mimicking the behavior of a mouse wheel, making obscured content visible. 9) Drag and Drop: Mimicking the user's behavior of dragging components. 10) Jump: Mimicking the user's page jump, including jumping to the current page, opening a new tab, and opening a new window. 11) Page Matching: Matching to determine if the current page is the target page. 12) Screenshot: Taking a screenshot of the current screen / browser view interface. 13) Confirmation Prompt: Can be used to provide secondary confirmation prompts before performing some irreversible or high-risk operations as a protection mechanism.

[0489] The objects to be manipulated by the agent include: 1) All operable component objects on the system page: radio buttons, checkboxes, input boxes, dates, maps, charts, icons, tables, buttons, etc. 2) Page scrollbars: horizontal scrollbars and vertical scrollbars.

[0490] Proxy content: This consists of data used to populate and assign values ​​to the proxy operation object. The proxy operation types, proxy operation objects, and proxy content can be freely combined according to business needs.

[0491] Visualization instructions include visualization type and visualization content.

[0492] Visualization types include: 1) Chart commands, which can be further divided into: bar charts, line charts, line graph bar charts, pie charts, cube charts, etc.; 2) Tables, which can be further divided into: no pagination, front-end pagination, and remote pagination; supporting filtering, with filter types including single selection, multiple selection, text, date, labels, etc.; ... 3) Single selection. 4) Multiple selection. 5) Buttons. 6) Lists. 7) Text. 8) Download. 9) Images...

[0493] Visualized content is generally transmitted in JSON format, and the specific data fields and hierarchical relationships are related to the visualization type.

[0494] The specific page composition is as follows Figure 11 As shown, the page presentation is as follows Figure 12 As shown.

[0495] In one embodiment, the method further includes a pre-trained large network model, specifically comprising:

[0496] Collect historical wireless network quality poor types, historical wireless network quality poor cell diagnostic work orders, and wireless network optimization expert knowledge. The wireless network optimization expert knowledge includes historical diagnostic sub-tasks and their chain relationships, as well as historical wireless network quality poor cell diagnostic results. The historical wireless network quality poor cell diagnostic results include the historical root causes of poor quality and historical poor quality handling solutions for the historical wireless network quality poor cells.

[0497] Control the first network model, extract thought chain question and answer prompts from the diagnostic process in the historical wireless network poor quality cell diagnostic work order corresponding to the historical wireless network poor quality type, compile historical diagnostic sub-tasks and their chain relationships under the thought chain question and answer prompts, and fine-tune the parameters of the first network model until the historical diagnostic sub-tasks and their chain relationships are accurately output.

[0498] The historical diagnostic subtasks are distributed to the corresponding poor quality expert agents, so that the poor quality expert agents can determine the wireless network API interface to be called based on the historical diagnostic subtasks, obtain historical poor quality cell data of wireless network through the wireless network API interface, judge the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts, and send the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information to the network optimization expert agent. The API interface is an application programming interface.

[0499] The system receives abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information sent by each poor-quality expert Agent. It controls the second network big model, combines the abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information with chain relationships and preset problem templates, outputs the diagnostic results of historical poor-quality cells of wireless networks, and fine-tunes the parameters of the second network big model until the diagnostic results of historical poor-quality cells of wireless networks are accurately output.

[0500] In this embodiment, the large network model used is a large language model trained and fine-tuned for the quality difference analysis task, such as... Figure 7 As shown, it includes two main functions: task decomposition and result aggregation. Both functions are based on the thinking chain capability of the large network model. The thinking chain capability is mainly achieved through incremental pre-training and SFT (Supervised Fine-Tuning) instruction fine-tuning.

[0501] Incremental pre-training: New knowledge is learned from a large-scale "unlabeled" text corpus through training. More than ten types of first-line business corpora and data, including wireless network air interface side, core network side, base station transmission side, bearer network, IP (Internet Protocol) network, backbone network, complaint handling, road testing, service quality analysis, network maintenance, etc., are collected and organized. Text purification is performed, PPL (Perplexity) is calculated to filter low-quality text, forming a network private professional knowledge dataset. The basic model is then incrementally pre-trained to infuse network expert knowledge.

[0502] SFT instruction fine-tuning: For the specific task of analyzing poor voice service quality cells, the system cleans and integrates labeled data such as voice service quality poor quality cell diagnosis work orders, voice service fault diagnosis knowledge graphs, and voice service performance indicators (KPIs) and perception indicators (KQIs) into specialized private domain data (network private professional knowledge datasets) for the voice service quality poor quality cell analysis process. This includes the responses to voice service diagnosis work orders, forming a structured voice service quality poor quality diagnosis process text. Subsequently, this text is used in conjunction with a large model and Prompt Engineering to form a thought chain capability. Based on intent understanding, the task is decomposed into sub-objectives, enabling the large model to possess ReAct-Agent capabilities for analyzing, diagnosing, perceiving, and making flexible decisions regarding voice service quality poor quality cell issues.

[0503] Among them, the keywords for forming the thinking chain ability (the thinking chain prompts of the first network big model) and the problem templates used by the second network big model for summarizing can come from the diagnostic work order. The diagnostic work order refers to the work order formed by human repair of poor quality communities in the past. The big language model is trained to learn human experience in the diagnostic work order, so that it has the ability to imitate humans to analyze and judge poor quality faults.

[0504] More specifically, the following empirical knowledge is used to design large-scale network models:

[0505] I. Experience in judging and handling poor-quality residential areas:

[0506] 1. Approach to handling poor-quality residential areas

[0507] First, define relevant indicators and design KPI monitoring templates on the network management system according to the assessment requirements. Then, filter out problematic cells based on call statistics data. Next, classify and handle different problems, track and record the process, and track and compare the results after the process is completed. The loop can only be closed after the problem is solved.

[0508] In LTE (Long Term Evolution) networks, the common approach to handling poor-quality cells involves addressing the following six categories. In daily operations, priority is given to voice / data service performance analysis, resource assessment, network structure analysis, and neighbor cell parameter verification, as shown in Table 2 below:

[0509] Table 2 Examples of Poor Quality Community Analysis and Processing Types

[0510]

[0511] 2. Suggested thresholds for judging poor-quality residential areas:

[0512] 2.1) RRC (Radio Resource Control) establishment failure: The number of RRC connection establishment failures is greater than 50, and the RRC connection establishment success rate is less than 95%;

[0513] 2.2) E-RAB (Evolved Radio Access Bearer) establishment failure: The number of E-RAB establishment failures is greater than 50, and the E-RAB establishment success rate is less than 95%;

[0514] 2.3) Drop rate: If the number of abnormal releases of the UE Context (User Equipment Context) is greater than 50, the drop rate is greater than 5%.

[0515] 2.4) Handover success rate: The number of cell handover failures is greater than 300, and the handover success rate is less than 80%;

[0516] 2.5) Capacity Resources: Sites involved in key areas.

[0517] Condition 1: The average utilization rate of downlink PRB (Physical Resource Block) is greater than 50%, the average number of effective RRC connections is greater than 30, and the downlink busy-hour throughput of the cell is greater than 5G (gigabytes, GB).

[0518] Condition 2: The average uplink PRB utilization is greater than 50%, the average number of effective RRC connections is greater than 30, and the cell uplink busy-hour throughput is greater than 1 G (gigabytes, GB).

[0519] Condition 3: The maximum number of valid RRC connections is greater than 200;

[0520] Note: Take the average data of the busiest time of the system in the community over 7 days, and it is acceptable if any one of the conditions is met.

[0521] 3. Suggestions for parameter verification in poor-quality residential areas are shown in Table 3 below:

[0522] Table 3. Parameter Verification Table for Poor-Quality Communities

[0523]

[0524] 4. Problem identification approach and steps in poor-quality residential communities:

[0525] Poor-quality cells typically fall into three main categories: radio-side parameter configuration issues, core network-side configuration issues, and channel environment influences. Therefore, when encountering a problematic cell, the following steps can be generally followed for troubleshooting.

[0526] 4.1) Confirm whether the network-wide indicators have deteriorated. If the network-wide indicators have deteriorated, it is necessary to check whether there are any network changes or upgrades in the operation and alarms.

[0527] 4.2) If the indicators of some sites deteriorate, affecting the overall network indicators, it is necessary to find the TOP poor quality cells (the worst quality cells).

[0528] 4.3) Query the top cells (the worst quality cells) with low RRC connection establishment success rate, low ERAB establishment success rate, low handover success rate, and high drop rate, and the time periods in which the top cells are most frequently affected (the time periods in which the poor quality occurs most frequently).

[0529] 4.4) Check the alarms of the TOP cell site, check the status of the board, the status of the RRU (Remote Radio Unit), the cell operation status, and whether the OMC (Operation & Management Center) operation configuration is normal.

[0530] 4.5) Extract CHR logs (log files reflecting problems from base stations), analyze the channel quality and SRS (Sounding Reference Signal) SINR (Signal to Interference plus Noise Ratio) during access to see if they are poor (weak coverage), and whether there are TOP users (users with the worst SINR).

[0531] 4.6) Perform targeted standard signaling tracing and interference detection analysis on TOP cell sites. If no anomalies are found in the standard signaling and interference detection, return the one-click logs, standard port tracing, and interference detection results to the developers for analysis.

[0532] 4.7) Based on the analysis results of all the above data, front-line network optimization engineers will analyze and eliminate each problem one by one, formulate optimization solutions based on expert experience, and then arrange RF optimization personnel and tower engineers to implement the optimization solutions until the problem of poor quality TOP cells is finally resolved.

[0533] II. Experience in designing machine learning algorithms:

[0534] 1. Algorithm level

[0535] Machine learning-driven localization of poor-quality cells in mobile communication networks requires, at the algorithmic level, integration of network service perception KQI data with comprehensive analysis of service perception profiles. Machine learning-driven algorithms effectively improve the accuracy of these profiles, constructing a machine learning-driven system for locating poor-quality cells and providing comprehensive user perception assessment and analysis. User profiles are created using quantitative data such as telephone follow-ups, user experience, and network satisfaction, and iteratively optimized. This process gradually identifies user groups and poor-quality content, further optimizing the location of poor-quality cells and analyzing the root causes of poor user satisfaction, ultimately providing optimization solutions for these cells. Machine learning pushes the analysis results to network maintenance units and integrates with network data analysis, including network performance indicators, network complaint data, coverage and drive test quality data, equipment failure rates, and impact duration, facilitating the mapping of basic customer perception elements and network elements.

[0536] 1.1) Business perception-related feature profile

[0537] The characteristic profile of service perception is mainly based on service models and time models. Mobile communication network services mainly include streaming media, web browsing, and instant messaging. Streaming media services primarily include video and games. Poor perception scenarios mainly include inability to access the internet, video buffering, game buffering, and slow webpage loading, which have the highest frequency of occurrence in current network complaints, accounting for approximately 90% of all complaints. Different services often correspond to different perception factors W. The time model implements hourly quantitative comprehensive scoring of user perception at different stages of the internet user's service process to accurately describe the specific situation of user perception.

[0538] 1.2) Basic Quantization Algorithm for Mobile Communication Network Sensing Based on Machine Learning

[0539] By conducting correlation analysis on user sample data through methods such as online KQI and online surveys, and using machine learning as the basic algorithm to effectively construct the indicator set, the user business perception coupling is realized, which leads to the initial formation of the profile model. Data verification analysis and model optimization are carried out to finally form a complete perception profile model, enabling quantitative assessment and effective identification of root causes of problems under customer perception.

[0540] 2. Identify users with poor quality.

[0541] 2.1) Identify poor-quality services

[0542] For business scenarios such as game lag, video lag, inability to access the internet, and slow webpage loading, perception indicators are determined separately to quantitatively define poor-quality services. For services related to inability to access the internet, control plane indicators are used for rapid definition. For video lag-related perception-level services, corresponding algorithms are introduced manually; for example, playback lag caused by insufficient data loading rate in the client buffer contains many perception indicators. Guided by logistic regression machine learning, the thresholds output by the iterative algorithm are calibrated, and combined with these thresholds, service identification is performed based on user perception. For example, if the actual number of video lags is 0.3 times / minute, the actual video playback waiting time is more than 5 seconds, and the total video lag time accounts for 30%, it can be determined that a user has experienced a poor user experience.

[0543] 2.2) User Perception Rating Model

[0544] Based on the basic model of perceived quality quantification for wireless internet access, the actual number of perceived differences experienced by current network users per unit time is obtained. Different dimensions of scoring are applied to different types of services with varying quality, resulting in the final user perception score. The formula is as follows:

[0545] S = ∑(K1W1 + K2W2 + K3W3)

[0546] S represents the user perception score; K1 represents the web browsing service score, and W1 represents the weight of the web browsing service; K2 represents the streaming media service score, and W2 represents the weight of the streaming media service; K3 represents the instant messaging service score, and W3 represents the instant messaging service weight.

[0547] Statistical analysis reveals that user perception varies, with most users falling into an intermediate state—meaning average or good perception—which aligns with a normal or chi-square distribution. In this typical distribution pattern, the focus is on the probability distribution. The horizontal axis represents the total number of failures per user across various scenarios at the hourly level, while the vertical axis represents the actual percentage of users corresponding to each failure. The vertical axis is a one-dimensional variable X, representing the probability of the total number of failures at the hourly level. By analyzing the probability density function curve of the random variable X, we can see that the number of failures can be divided into different intervals, each corresponding to a different user perception score. We can project user scores onto five different intervals to obtain the perception scores of current users. For example, we can filter and output users with perception scores below 3 as having poor perception quality.

[0548] 3. Identifying poor-quality residential areas

[0549] 3.1) Basic Model for Community Perception Scoring

[0550] By combining the number of poor-quality services received from users, and aggregating the users' frequently visited communities, and inputting the actual number of poor-quality services received by different users in each community, we can obtain the actual number of perceived poor services received by each user and the total number of such services. This allows us to understand that the proportion of users with perceived poor service is higher than X (users with perceived poor service). Statistical analysis of the number of users with perceived poor service Ki and the total number of users Mi in each community yields the proportion of users with perceived poor service as Zi = Ki / Mi.

[0551] 3.2) The proportion of users with poor perception can be divided into five different intervals.

[0552] The perception scores of users in each corresponding interval differ. Based on the perception algorithm, the cell score is effectively mapped to 5 different intervals to obtain the cell score, that is, 5 points is the best perception and 0 points is the worst perception.

[0553] 3.3) Identifying poor-quality residential areas

[0554] Based on this model, cell scores are obtained, and cells with low perception scores below 3 are filtered out as poor-quality cells.

[0555] 4. Identify the root causes of poor-quality communities and develop solutions.

[0556] Network quality and user experience are interdependent. Evaluating user experience can be comprehensively assessed based on factors such as network availability, stable connection speed, and high data rate. When a cell exhibits issues like low access speed, high dropout rate, or low data rate, it is considered a poor-quality cell negatively impacting user experience. While these poor-quality cells can be identified through metrics like dropout rate, wireless access success rate, and data rate, the diverse and often multifaceted causes affecting user experience make pinpointing the root causes of poor-quality cells extremely difficult for frontline personnel. Currently, the primary approach relies on expert experience, categorizing network problems into types such as coverage, interference, capacity, and faults. This is combined with data from Configuration Management (CM), Performance Management (PM), and Measurement Reports (MR) to develop root cause analysis algorithms for identifying the root causes of poor-quality cells.

[0557] 4.1) Coverage-based root cause localization algorithm

[0558] Wireless network coverage issues are caused by unreasonable network coverage. In general, in macro base station scenarios, network coverage problems can be divided into weak coverage, over-coverage, and overlapping coverage. Overlapping coverage and weak coverage problems can be located by jointly using indicators such as RSRP (Reference Signal Received Power), TA (Time Advanced), and overlapping coverage rate.

[0559] ① Overcoverage Root Cause Analysis

[0560] The large-scale deployment of 4G (fourth-generation mobile communication technology) networks has matured, but the coverage areas of planned sites vary across different scenarios, such as urban areas, suburbs / counties, and towns / rural areas. Urban sites are densely packed, suburban / county sites have larger spacing, and town / rural sites are planned for wide coverage. Therefore, when developing root cause analysis algorithms, different TA (Target Acquisition) thresholds need to be set according to different coverage scenarios. Urban sites are generally planned to cover within 1km. Considering the differences in the wireless environment within urban areas, a TA > 1km ratio > 25% is used as a prerequisite for determining over-coverage in urban areas. Suburban / county sites, due to their larger spacing, should not have excessively long coverage areas, but also not too short. A TA > 2.5km ratio > 25% is used as a prerequisite for determining over-coverage in suburban / county areas. While town / rural sites have wider coverage, considering wireless propagation loss—signal attenuation increases with distance and signal penetration weakens—the coverage distance should not be too wide. A TA > 3.5km ratio > 25% is used as a prerequisite for determining over-coverage in town / rural areas.

[0561] ② Overcoverage Root Cause Analysis

[0562] Coverage too close refers to signal coverage being significantly lower than the planned coverage area. For example, poor-quality cells may have excessively large antenna downtilt angles, obstructions in the coverage direction, or hidden faults in the antenna feeder system, all leading to coverage being too close. This can be comprehensively assessed using indicators such as TA (Target Aspect Ratio) and RSRP (Recovery Support Ratio). Using a single indicator to determine the root cause of poor signal quality has low accuracy. Therefore, by combining indicators such as TA and MR (Mean Mitigation Ratio) with the inter-cell spacing and average cell coverage distance for different scenarios, root cause localization algorithms can be developed for weak coverage, overlapping coverage, over-coverage, and coverage being too close, as shown in Table 4 below.

[0563] Table 4. Criteria for Judging Coverage Issues

[0564]

[0565]

[0566] ③ Optimize solution output

[0567] For coverage issues such as weak or excessive coverage of macro base stations, assuming the base station and antenna system are functioning normally and the parameters are set reasonably, there are generally five solutions: adjusting antenna height, adjusting antenna downtilt and azimuth angles, adjusting RS (Reference Signal) power, relocating the site, and adding RRUs. Among these, adjusting antenna downtilt and azimuth angles is the preferred solution.

[0568] 4.2) Interference-type root cause localization algorithm

[0569] Wireless network interference problems generally include external interference and internal interference. In the root cause algorithm proposed in this paper, interference problems are identified by using the average interference noise power per PRB, which is an indicator of interference intensity.

[0570] ① External interference

[0571] Generally referred to as external interference, it is caused by external interference sources. External interference is continuous and persistent. When judging external interference, it is necessary to exclude situations where interference increases during the day due to factors such as user activity. Generally, the interference index at 03:00, the off-peak time for business, is considered for evaluation.

[0572] ②Interference within the system

[0573] This refers to interference caused by co-channel interference within the system, GPS (Global Positioning System) malfunctions, data configuration errors, etc. The time period for interference within the system is not fixed and is greatly affected by factors such as user behavior; generally, an interference index covering the entire day is considered. The root cause localization algorithms for external and internal system interference are shown in Table 5 below:

[0574] Table 5. Basis for Locating Interference Problems

[0575]

[0576] ③ Optimize solution output

[0577] For interference-related issues, assuming the base station and antenna system are functioning normally and the parameters are set appropriately, there are generally five solutions: troubleshooting external interference sources, adjusting antenna height, adjusting antenna downtilt and azimuth angles, and adjusting RS power. Among these, troubleshooting external interference sources should be given priority.

[0578] 4.3) Root Cause Localization Algorithm for Poor Quality

[0579] Wireless network quality indicators directly characterize network quality. For example, CQI (Channel Quality Indicator) represents channel quality, and MCS (Modulation and Coding Scheme) represents resource scheduling. In macrocell scenarios, poor network quality is generally caused by weak coverage, over-coverage, and overlapping coverage. In root cause analysis based on expert experience, CQI, MCS, and TA (Transmission and Coding Scheme) indicators can be jointly used to pinpoint quality issues.

[0580] ① Quality difference analysis based on coverage type

[0581] For poor quality issues, the coverage decision conditions for urban, suburban / county, and township / rural sites are consistent with the coverage analysis process. In the root cause localization algorithm for poor quality issues, CQI, MCS, TA index, average cell coverage distance, and average inter-site spacing under different scenarios are combined to locate the root causes of poor quality cells, as shown in Table 6 below:

[0582] Table 6. Basis for Coverage Type Quality Poor Analysis

[0583]

[0584] ②Optimize solution output

[0585] Solving quality issues is similar to solving coverage issues. Under the premise of ensuring that the base station and antenna feeder system are working properly and the parameters are set reasonably, optimization can be achieved by adjusting the antenna height, antenna downtilt angle and azimuth angle, and RS power.

[0586] 4.4) Capacity-based root cause localization algorithm

[0587] High load directly affects various network indicators. In the root cause algorithm based on expert experience, load-related problems such as inter-carrier traffic load imbalance and inter-sector traffic load imbalance are jointly located by indicators that characterize the load, such as PRB utilization and RS power.

[0588] ① Load imbalance analysis

[0589] Under the same RRU, if two logical cells with the same bandwidth but different frequency points have the same maximum power, and their PRB utilization difference is greater than 20%, then there is a load imbalance. In reality, under the same physical cell, two logical cells with the same coverage may not have the same maximum power due to interference avoidance and other factors, but under normal circumstances, the difference will not exceed 3dB. Based on practical considerations, PRB utilization and RS power are used together to locate inter-carrier load balancing issues. This method is also applicable to inter-sector load balancing decisions.

[0590] Based on indicators such as PRB utilization, CCE (Control Channel Element) utilization, number of users, traffic, RS power, and cell merging, root cause algorithms were developed for issues such as uneven traffic load between carriers, uneven traffic load between sectors, large number of users, insufficient bandwidth, high load in merged logical cells, uneven traffic load between license-limited sectors, and high CCE utilization, as shown in Table 7 below.

[0591] Table 7. Basis for Quality Defect Analysis of Volume Category

[0592]

[0593]

[0594] ②Optimize solution output

[0595] To address the high load issue of macro base stations, assuming the base station and antenna feeder system are functioning normally and the parameters are set reasonably, there are generally five solutions: cell expansion, cell splitting and merging, inter-carrier load balancing, inter-site load balancing, and expansion license. Among these, inter-carrier load balancing and inter-site load balancing are preferred.

[0596] 5. Optimization effect verification

[0597] The effectiveness of the root cause localization algorithm was verified by conducting an application experiment in the pilot area.

[0598] This embodiment utilizes the definition of poor-quality cells in 4G / 5G wireless networks and constructs an intelligent agent for analyzing and processing poor-quality cells based on a large-model multi-agent architecture. This method can be designed on the basis of cells in any network or even higher-evolved networks. This solution can be applied in a poor-quality analysis assistant product, and the application effect is as follows: Figure 12 As shown.

[0599] Example 2:

[0600] like Figure 2 As shown, this disclosure provides a method for processing poor-quality cells in a wireless network. The method is applied to a poor-quality expert agent and includes:

[0601] S21. Receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent.

[0602] S22. Obtain the diagnostic sub-results according to the corresponding diagnostic sub-tasks, and send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network based on the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality.

[0603] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0604] In one embodiment, receiving the diagnostic subtask distributed by the network optimization expert Agent in S21 specifically includes:

[0605] Responding to the fact that it is one of the fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent, it receives the corresponding diagnostic sub-tasks distributed by the network optimization expert agent. The diagnostic sub-tasks correspond to one of the fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis, and are the first network big model of the network optimization expert agent. It is compiled by calling the thinking chain question and answer prompt words based on the real-time wireless network quality poor type of the wireless network poor cell.

[0606] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0607] In one embodiment, step S22 involves obtaining a diagnostic sub-result based on the corresponding diagnostic sub-task and sending the diagnostic sub-result to the network optimization expert Agent, specifically including:

[0608] The corresponding diagnostic subtask determines the wireless network API interface to be called, and the real-time data of poor wireless network quality cells is obtained through the corresponding wireless network API interface. The API interface is an application programming interface.

[0609] Based on the knowledge of wireless network quality diagnosis experts, identify abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information, and use the abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information as diagnostic sub-results.

[0610] The diagnostic sub-results are sent to the network optimization expert agent so that the network optimization expert agent can obtain several diagnostic sub-tasks and their chain relationships, corresponding to the thought chain question-and-answer prompts and their order. The agent then inputs the several diagnostic sub-results into the question template of the second network big model in order to obtain diagnostic results including the root causes of poor quality in the poor quality cells of the wireless network and suggestions for handling poor quality.

[0611] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0612] In one embodiment, the method further includes an auxiliary pre-trained network large model, specifically comprising:

[0613] Receive historical diagnostic subtasks distributed by the network optimization expert agent. These historical diagnostic subtasks are generated by the network optimization expert agent after training the first large network model.

[0614] Based on the historical diagnostic subtasks, determine the wireless network API interfaces that need to be called, and obtain historical data on poor-quality wireless network cells through the wireless network API interfaces;

[0615] Based on the knowledge of wireless network quality diagnosis experts, identify historical poor-quality wireless network cell data with anomalies and their corresponding historical processing experience information.

[0616] Abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information are sent to the network optimization expert agent, so that the network optimization expert agent can train the second large network model based on the abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information sent by each poor-quality expert agent.

[0617] Example 3:

[0618] like Figure 3 As shown, this disclosure provides a device for processing poor-quality cells in a wireless network. The device is a network optimization expert agent, which includes:

[0619] The compilation and distribution module 11 is used to compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells, and distribute each diagnostic sub-task to the corresponding poor quality expert Agent, so that each poor quality expert Agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert Agent.

[0620] The receiving and summarizing module 12 is connected to the compilation and distribution module 11. It is used to receive the diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of the poor quality cell of the wireless network based on several diagnostic sub-results and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cell of the wireless network and the poor quality handling suggestions.

[0621] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0622] In one embodiment, the compilation and distribution module 11 specifically includes:

[0623] The intent recognition unit is used to receive user questions through the first network big model and identify the user's query intent regarding poor wireless network quality cells in the user questions;

[0624] The type determination unit, connected to the intent recognition unit, is used to determine the real-time wireless network quality poorness type of the wireless network poorness cell based on the statistical results of the indicators of the wireless network poorness cell.

[0625] The compilation unit, connected to the type judgment unit, is used to compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells by calling the thought chain question and answer prompts based on the real-time poor wireless network quality type through the first network big model. The several diagnostic sub-tasks include at least one of the following: fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and hidden fault diagnosis.

[0626] The distribution unit, connected to the compilation unit, is used to distribute each diagnostic subtask to the corresponding poor quality expert agent. The poor quality expert agent includes at least one of the following: fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent.

[0627] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0628] In one embodiment, the receiving and summarizing module 12 includes:

[0629] The aggregation unit is specifically used to obtain several diagnostic sub-tasks and their chain relationships, corresponding to the thought chain question-and-answer prompts and their order. The several diagnostic sub-results are then input into the question template of the second network model in order to obtain diagnostic results including the root causes of poor wireless network quality and suggestions for handling poor quality. The diagnostic sub-results include abnormal real-time data of poor wireless network quality cells and their corresponding historical handling experience information.

[0630] The output unit, connected to the aggregation unit, is used to output voice and / or text results of the root causes of poor wireless network quality and suggestions for handling poor quality cells. It displays the poor wireless network quality cells on a map and marks their root causes of poor quality. It generates page instructions to repair the root causes of poor quality based on the suggestions for handling poor quality, and executes the page instructions automatically or when triggered by the user to repair the poor wireless network quality cells.

[0631] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0632] In one embodiment, the network optimization expert agent further includes a training module, specifically comprising:

[0633] The data acquisition unit is used to collect historical wireless network quality poor types, historical wireless network quality poor cell diagnostic work orders, and wireless network optimization expert knowledge. The wireless network optimization expert knowledge includes historical diagnostic sub-tasks and their chain relationships, as well as historical wireless network quality poor cell diagnostic results. The historical wireless network quality poor cell diagnostic results include the historical root causes of the poor quality of the historical wireless network quality poor cells and historical poor quality handling solutions.

[0634] The first training unit, connected to the acquisition unit, is used to control the first large network model. It extracts thought chain question-and-answer prompts from the diagnostic process in the diagnostic work order of the historical poor wireless network cell corresponding to the historical poor wireless network quality type. It compiles historical diagnostic sub-tasks and their chain relationships under the thought chain question-and-answer prompts and fine-tunes the parameters of the first large network model until it accurately outputs the historical diagnostic sub-tasks and their chain relationships.

[0635] The distribution unit is also connected to the first training unit and is used to distribute historical diagnostic subtasks to the corresponding poor quality expert agents, so that the poor quality expert agents can determine the wireless network API interface to be called based on the historical diagnostic subtasks, obtain historical poor quality cell data of wireless networks through the wireless network API interface, judge the abnormal historical poor quality cell data of wireless networks and its corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts, and send the abnormal historical poor quality cell data of wireless networks and its corresponding historical processing experience information to the network optimization expert agent. The API interface is an application programming interface.

[0636] The second training unit, connected to the distribution unit, receives abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information sent by each poor-quality expert agent. It controls the second network big model to combine the abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information with chain relationships and preset problem templates, outputs the diagnostic results of historical poor-quality cells of wireless networks, and fine-tunes the parameters of the second network big model until the diagnostic results of historical poor-quality cells of wireless networks are accurately output.

[0637] Example 4:

[0638] like Figure 4 As shown, this disclosure provides a device for processing poor-quality cells in a wireless network. The device is a poor-quality expert agent, which includes:

[0639] The receiving task module 21 is used to receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent.

[0640] The execution task module 22 is connected to the receiving task module 21. It is used to obtain the diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network according to the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality.

[0641] Among them, Agent is an intelligent agent. Network optimization expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and poor quality expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis.

[0642] In one embodiment, the task receiving module 21 is specifically used for:

[0643] Responding to the fact that it is one of the fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent, it receives the corresponding diagnostic sub-tasks distributed by the network optimization expert agent. The diagnostic sub-tasks correspond to one of the fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis, and are the first network big model of the network optimization expert agent. It is compiled by calling the thinking chain question and answer prompt words based on the real-time wireless network quality poor type of the wireless network poor cell.

[0644] Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

[0645] In one embodiment, the task execution module 22 specifically includes:

[0646] The data acquisition unit is used to determine the wireless network API interface to be called according to the corresponding diagnostic subtask, and to acquire real-time data of poor wireless network quality cells through the corresponding wireless network API interface. The API interface is an application programming interface.

[0647] The anomaly judgment unit, connected to the data acquisition unit, is used to judge the abnormal real-time poor quality cell data of wireless network and its corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts, and to take the abnormal real-time poor quality cell data of wireless network and its corresponding historical processing experience information as the diagnostic sub-result.

[0648] The result sending unit, connected to the anomaly judgment unit, is used to send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain several diagnostic sub-tasks and their chain relationship corresponding to the thought chain question and answer prompts and their order, and input several diagnostic sub-results into the question template of the second network big model in order to obtain diagnostic results including the root cause of poor quality of the poor quality cell in the wireless network and the poor quality handling suggestions;

[0649] The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

[0650] In one embodiment, the quality-poor expert agent also assists in the pre-training of a large network model, specifically including:

[0651] The task receiving module 21 is also used to receive historical diagnostic subtasks distributed by the network optimization expert agent. The historical diagnostic subtasks are generated after the network optimization expert agent trains the first large network model.

[0652] The data acquisition unit is also used to determine the wireless network API interface to be called based on the historical diagnostic subtasks, and to acquire historical data of poor-quality wireless network cells through the wireless network API interface;

[0653] The anomaly detection unit is also used to determine the historical poor quality cell data of wireless networks and their corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts.

[0654] The result sending unit is also used to send abnormal historical poor quality cell data of wireless networks and their corresponding historical processing experience information to the network optimization expert agent, so that the network optimization expert agent can train the second network large model based on the abnormal historical poor quality cell data of wireless networks and their corresponding historical processing experience information sent by each poor quality expert agent.

[0655] Example 5:

[0656] like Figure 5 As shown, this disclosure provides a device for processing poor-quality cells in a wireless network. The device is a system for processing poor-quality cells in a wireless network, and the system includes:

[0657] The network optimization expert Agent described in Example 3 is used to implement the poor quality wireless network cell processing method described in Example 1.

[0658] As described in Example 4, several poor quality expert agents are connected to network optimization expert agents to implement the poor quality cell processing method for wireless networks as described in Example 2.

[0659] Example 6:

[0660] Embodiment 6 of this disclosure provides a computer-readable storage medium storing a computer program. When the computer program is run by a processor, it implements the poor quality cell processing method for wireless networks as described in Embodiment 1 or 2, or the poor quality cell processing apparatus for wireless networks as described in any of Embodiments 3-5.

[0661] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program units, or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), DVD or other optical disc storage, cartridges, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer.

[0662] Additionally, this disclosure may provide a computer device including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the poor-quality wireless network cell processing method as described in Embodiment 1 or 2. This computer device may be a poor-quality wireless network cell processing apparatus as described in any of Embodiments 3-5.

[0663] The memory is connected to the processor. The memory can be flash memory, read-only memory or other types of memory. The processor can be a central processing unit or a microcontroller.

[0664] Embodiments 1-6 of this disclosure provide a method, apparatus, and medium for processing poor-quality cells in a wireless network. They employ multi-agent collaboration to provide a highly intelligent and fast network optimization auxiliary tool. Through collaboration between a network optimization expert agent and several poor-quality expert agents, the network optimization expert agent compiles diagnostic sub-tasks and their chain relationships. The poor-quality expert agents execute these diagnostic sub-tasks to obtain diagnostic sub-results. Finally, the network optimization expert agents summarize the diagnostic sub-results according to the chain relationships to obtain the root causes of poor-quality cells in the wireless network and suggestions for handling them. This can be used to assist network optimization engineers in completing network optimization work. The efficiency of network optimization is improved by having multiple agents perform diagnoses for each dimension separately, and the accuracy of the final diagnostic result is improved by obtaining the multi-dimensional network poor-quality diagnostic sub-results.

[0665] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this disclosure, and this disclosure is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this disclosure, and these modifications and improvements are also considered to be within the scope of protection of this disclosure.

Claims

1. A method for processing poor-quality cells in a wireless network, characterized in that, The method is applied to a network optimization expert agent and includes: Several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells are compiled. Each diagnostic sub-task is distributed to the corresponding poor quality expert agent, so that each poor quality expert agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent. Receive diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of poor wireless network cells based on several diagnostic sub-results and their corresponding chain relationships. The diagnostic results include the root causes of poor wireless network quality and suggestions for handling poor quality. Among them, Agent is an intelligent agent, Network Optimization Expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and Poor Quality Expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis. This includes developing several diagnostic sub-tasks and their chained relationships for poor-quality wireless network cells, and distributing each diagnostic sub-task to the corresponding poor-quality expert agent, specifically including: The system receives user questions through the first network big model and identifies the user's intent to inquire about cells with poor wireless network quality. Based on the statistical results of the indicators of cells with poor wireless network quality, determine the real-time wireless network quality poorness type of the cells with poor wireless network quality; Based on the real-time wireless network quality issues, the first network big model uses thought chain question-and-answer prompts to compile several diagnostic sub-tasks and their chain relationships for cells with poor wireless network quality. These diagnostic sub-tasks include at least one of the following: fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis. Each diagnostic subtask is assigned to a corresponding quality defect expert agent. The quality defect expert agent includes at least one of the following: fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and hidden fault diagnosis agent. Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

2. The method according to claim 1, characterized in that, The diagnostic results for poor-quality wireless network cells are obtained based on several diagnostic sub-results and their corresponding chain relationships. These results include the root causes of the poor-quality wireless network and recommendations for resolving the issue. Specifically, they include: Obtain several diagnostic sub-tasks and their chain relationships, along with the corresponding thought chain question-and-answer prompts and their order. Input the several diagnostic sub-results into the question template of the second network model in order to obtain diagnostic results including the root causes of poor wireless network quality and suggestions for handling poor quality. The diagnostic sub-results include abnormal real-time data on poor wireless network quality cells and their corresponding historical handling experience information. Output voice and / or text results of the root causes of poor wireless network quality and suggestions for handling poor quality cells; display poor wireless network quality cells on a map and mark their root causes; generate page instructions to repair the root causes of poor quality based on the suggestions for handling poor quality; and execute the page instructions automatically or when triggered by the user to repair the poor wireless network quality cells. The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

3. The method according to claim 2, characterized in that, The method also includes a pre-trained large network model, specifically including: Collect historical wireless network quality poor types, historical wireless network quality poor cell diagnostic work orders, and wireless network optimization expert knowledge. The wireless network optimization expert knowledge includes historical diagnostic sub-tasks and their chain relationships, as well as historical wireless network quality poor cell diagnostic results. The historical wireless network quality poor cell diagnostic results include the historical root causes of poor quality and historical poor quality handling solutions for the historical wireless network quality poor cells. Control the first network model, extract thought chain question and answer prompts from the diagnostic process in the historical wireless network poor quality cell diagnostic work order corresponding to the historical wireless network poor quality type, compile historical diagnostic sub-tasks and their chain relationships under the thought chain question and answer prompts, and fine-tune the parameters of the first network model until the historical diagnostic sub-tasks and their chain relationships are accurately output. The historical diagnostic subtasks are distributed to the corresponding poor quality expert agents, so that the poor quality expert agents can determine the wireless network API interface to be called based on the historical diagnostic subtasks, obtain historical poor quality cell data of wireless network through the wireless network API interface, judge the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information based on the knowledge of wireless network poor quality diagnosis experts, and send the abnormal historical poor quality cell data of wireless network and its corresponding historical processing experience information to the network optimization expert agent. The API interface is an application programming interface. The system receives abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information sent by each poor-quality expert Agent. It controls the second network big model, combines the abnormal historical poor-quality cell data of wireless networks and their corresponding historical processing experience information with chain relationships and preset problem templates, outputs the diagnostic results of historical poor-quality cells of wireless networks, and fine-tunes the parameters of the second network big model until the diagnostic results of historical poor-quality cells of wireless networks are accurately output.

4. A method for processing poor-quality cells in a wireless network, characterized in that, The method is applied to a poor-quality expert agent and includes: Receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent. Obtain diagnostic sub-results according to the corresponding diagnostic sub-tasks, and send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network based on the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality. Among them, Agent is an intelligent agent, Network Optimization Expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and Poor Quality Expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis. The diagnostic sub-tasks distributed by the network optimization expert agent include: Responding to the fact that it is one of the fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent, it receives the corresponding diagnostic sub-tasks distributed by the network optimization expert agent. The diagnostic sub-tasks correspond to one of the fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis, and are the first network big model of the network optimization expert agent. It is compiled by calling the thinking chain question and answer prompt words based on the real-time wireless network quality poor type of the wireless network poor cell. Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

5. The method according to claim 4, characterized in that, Based on the corresponding diagnostic sub-task, obtain the diagnostic sub-results and send them to the network optimization expert Agent. Specifically, this includes: The corresponding diagnostic subtask determines the wireless network API interface to be called, and the real-time data of poor wireless network quality cells is obtained through the corresponding wireless network API interface. The API interface is an application programming interface. Based on the knowledge of wireless network quality diagnosis experts, identify abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information, and use the abnormal real-time wireless network quality poor cell data and their corresponding historical processing experience information as diagnostic sub-results. The diagnostic sub-results are sent to the network optimization expert agent so that the network optimization expert agent can obtain several diagnostic sub-tasks and their chain relationships, corresponding to the thought chain question-and-answer prompts and their order. The agent then inputs the several diagnostic sub-results into the question template of the second network big model in order to obtain diagnostic results including the root causes of poor quality in the poor quality cells of the wireless network and suggestions for handling poor quality. The second network big model is a large language model used to summarize the root causes of poor quality and suggestions for handling poor quality cells in wireless networks after the first network big model is trained. The problem templates include questions to obtain the root causes of poor quality based on summarizing several abnormal real-time wireless network poor quality cell data and questions to obtain suggestions for handling poor quality based on summarizing several historical processing experiences.

6. The method according to claim 5, characterized in that, The method also includes an auxiliary pre-trained network large model, specifically including: Receive historical diagnostic subtasks distributed by the network optimization expert agent. These historical diagnostic subtasks are generated by the network optimization expert agent after training the first large network model. Based on the historical diagnostic subtasks, determine the wireless network API interfaces that need to be called, and obtain historical data on poor-quality wireless network cells through the wireless network API interfaces; Based on the knowledge of wireless network quality diagnosis experts, identify historical poor-quality wireless network cell data with anomalies and their corresponding historical processing experience information. Abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information are sent to the network optimization expert agent, so that the network optimization expert agent can train the second large network model based on the abnormal historical poor-quality wireless network cell data and their corresponding historical processing experience information sent by each poor-quality expert agent.

7. A network optimization expert agent, characterized in that, The network optimization expert agent includes: The compilation and distribution module is used to compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells, and distribute each diagnostic sub-task to the corresponding poor quality expert agent, so that each poor quality expert agent can obtain its own diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent. The receiving and summarizing module, connected to the compilation and distribution module, is used to receive the diagnostic sub-results sent by each poor quality expert Agent, and obtain the diagnostic results of poor quality wireless network cells based on several diagnostic sub-results and their corresponding chain relationships. The diagnostic results include the root causes of poor quality and suggestions for handling poor quality in the poor quality wireless network cells. Among them, Agent is an intelligent agent, Network Optimization Expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and Poor Quality Expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis. The compilation and distribution module specifically includes: The intent recognition unit is used to receive user questions through the first network big model and identify the user's query intent regarding poor wireless network quality cells in the user questions; The type determination unit, connected to the intent recognition unit, is used to determine the real-time wireless network quality poorness type of the wireless network poorness cell based on the statistical results of the indicators of the wireless network poorness cell. The compilation unit, connected to the type judgment unit, is used to compile several diagnostic sub-tasks and their chain relationships for poor wireless network quality cells by calling the thought chain question and answer prompts based on the real-time poor wireless network quality type through the first network big model. The several diagnostic sub-tasks include at least one of the following: fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and hidden fault diagnosis. The distribution unit, connected to the compilation unit, is used to distribute each diagnostic subtask to the corresponding poor quality expert agent. The poor quality expert agent includes at least one of the following: fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent. Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

8. A poor quality expert agent, characterized in that, The poor quality expert agent includes: The receiving task module is used to receive diagnostic subtasks distributed by the network optimization expert agent. The diagnostic subtasks are the diagnostic subtasks corresponding to the poor quality expert agent in the chain relationship of several diagnostic subtasks of poor quality cells of wireless network compiled by the network optimization expert agent. The task execution module, connected to the task receiving module, is used to obtain diagnostic sub-results according to the corresponding diagnostic sub-tasks and send the diagnostic sub-results to the network optimization expert agent, so that the network optimization expert agent can obtain the diagnostic results of the poor quality cells of the wireless network according to the several diagnostic sub-results sent by several poor quality expert agents and the corresponding chain relationship. The diagnostic results include the root cause of the poor quality of the poor quality cells of the wireless network and the suggestions for handling the poor quality. Among them, Agent is an intelligent agent, Network Optimization Expert Agent is an intelligent agent with expert knowledge of wireless network optimization, and Poor Quality Expert Agent is an intelligent agent with expert knowledge of poor wireless network quality diagnosis. The task receiving module is specifically used for: Responding to the fact that it is one of the fault diagnosis agent, parameter diagnosis agent, coverage diagnosis agent, quality diagnosis agent, interference diagnosis agent, capacity diagnosis agent, handover diagnosis agent, and latent fault diagnosis agent, it receives the corresponding diagnostic sub-tasks distributed by the network optimization expert agent. The diagnostic sub-tasks correspond to one of the fault diagnosis, parameter diagnosis, coverage diagnosis, quality diagnosis, interference diagnosis, capacity diagnosis, handover diagnosis, and latent fault diagnosis, and are the first network big model of the network optimization expert agent. It is compiled by calling the thinking chain question and answer prompt words based on the real-time wireless network quality poor type of the wireless network poor cell. Among them, the first network big model is a big language model used to compile several diagnostic sub-tasks and their chain relationships after being trained based on the wireless network optimization problem. The thought chain question and answer prompt words are obtained in advance based on the diagnostic process in the historical wireless network poor quality cell diagnostic work order.

9. A system for processing poor-quality wireless network cells, characterized in that, The system includes: Network optimization expert Agent, used to implement the method for handling poor-quality cells in wireless networks as described in any one of claims 1-3; Several poor quality expert agents are connected to network optimization expert agents to implement the poor quality cell processing method for wireless networks as described in any one of claims 4-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for processing poor-quality cells in a wireless network as described in any one of claims 1-3 or 4-6.

Images