A method and system for intelligent control of network transmission terminals

Abstract

This application proposes an intelligent control method and system for network transmission terminals, relating to the field of intelligent control technology for network transmission terminals. The method includes acquiring terminal operation data, external environment data, and user traffic data; performing correlation analysis on the terminal operation data, external environment data, and user traffic data to obtain correlation analysis results to identify evaluation biases in a channel evaluation model; generating calibration parameters for calibrating the channel evaluation model based on the correlation analysis results; applying the calibration parameters to the channel evaluation model to correct the evaluation biases and obtain a channel evaluation result; and adjusting the transmission power based on the channel evaluation result. This application can improve the stability and reliability of network transmission.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology for network transmission terminals, and in particular to an intelligent control method and system for network transmission terminals. Background Technology

[0002] In related technologies, network transmission terminals play a crucial role in modern wireless communication networks, responsible for efficient and stable data transmission in complex wireless environments. To address the inherent uncertainties of wireless channels, such as signal attenuation, multipath effects, and various interferences, these terminals typically integrate advanced intelligent control methods. The core of these methods lies in their ability to assess channel quality in real time and dynamically adjust their transmission parameters based on the assessment results. Among these, adjusting transmission power is a key aspect of ensuring the reliability of the communication link. However, in practical applications, especially in scenarios where the environment and user behavior patterns are constantly evolving, these intelligent control methods may face severe challenges. For example, when the urban environment of the network transmission terminal deployment area undergoes gradual changes, such as the emergence of new building clusters, and when user traffic patterns change significantly, the computational model within the terminal used to assess channel quality begins to show its limitations. This model was initially designed and trained and optimized based on open or minimally obstructed environments. Faced with the complex multipath propagation and deep fading phenomena brought about by new building clusters, the accuracy of the model's assessment of actual channel quality begins to decline. This reduced accuracy of the assessment leads to biases in the information fed back to the transmission power adjustment strategy, making it unable to accurately reflect the current wireless environment. Summary of the Invention

[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes an intelligent control method and system for network transmission terminals, aiming to improve the stability and reliability of network transmission.

[0004] In a first aspect, embodiments of this application provide a method for intelligent control of a network transmission terminal, comprising:

[0005] Acquire terminal operation data, external environment data, and user traffic data;

[0006] The terminal operation data, the external environment data, and the user traffic data are correlated and analyzed to obtain correlation analysis results in order to identify the evaluation bias of the channel evaluation model.

[0007] Based on the correlation analysis results, calibration parameters for calibrating the channel evaluation model are generated;

[0008] The calibration parameters are applied to the channel evaluation model to correct the evaluation bias of the channel evaluation model and obtain the channel evaluation result;

[0009] Based on the channel assessment results, adjust the transmission power.

[0010] According to some embodiments of this application, the step of performing correlation analysis on the terminal operation data, the external environment data, and the user traffic data to obtain correlation analysis results for identifying evaluation biases in the channel evaluation model includes:

[0011] For high real-time data streams identified as industrial IoT, a preset frequency monitoring logic is initiated, and the bit error rate and packet loss rate are monitored within the first preset time window.

[0012] Generate micro-interruption event alarms based on the bit error rate and the packet loss rate;

[0013] Acquire the terminal's geographic location and 3D models of highly reflective buildings from geographic information system data;

[0014] The direction of arrival of the active reflected signal is reversed based on the geographical location and the three-dimensional model;

[0015] Determine whether the direction of arrival of the reflected signal matches the theoretical path from the terminal to a preset high-reflectivity building and then reflected back to the base station;

[0016] Determine whether the reflected signal and the micro-interruption event alarm are synchronized in time;

[0017] A high-risk specular reflection path is obtained when the direction of arrival of the reflected signal matches the theoretical path and the reflected signal is synchronized with the micro-interruption event alarm in time.

[0018] Based on the high-risk specular reflection path, the terminal operation data, the external environment data, and the user traffic data, correlation analysis results are obtained to identify the evaluation bias of the channel evaluation model.

[0019] According to some embodiments of this application, generating calibration parameters for calibrating the channel evaluation model based on the correlation analysis results includes:

[0020] Obtain calibration parameters for high-risk specular reflection paths;

[0021] Based on the calibration parameters and the correlation analysis results, an assessment is conducted to reduce the contribution of the signal components of the high-risk specular reflection path to the overall channel quality, generating calibration parameters for calibrating the channel assessment model.

[0022] According to some embodiments of this application, the high-risk specular reflection path is obtained through the following steps:

[0023] Acquire the spatial direction of arrival and polarization state of the received signal and mark potential specular reflection signals;

[0024] By comparing the arrival direction of the potential specular reflection signal with the geometric reflection path of the high-reflectivity structure in the geographic information system data, high-risk specular reflection paths are marked.

[0025] According to some embodiments of this application, the step of assessing the contribution of the signal components of the high-risk specular reflection path to the overall channel quality based on the calibration parameters and the correlation analysis results, and generating calibration parameters for calibrating the channel assessment model, includes:

[0026] Evaluate the transient interruption characteristics of each of the high-risk specular reflection paths;

[0027] Identify the types of industrial IoT services transmitted through the high-risk specular reflection path, and allocate service priorities according to the latency and reliability requirements of the industrial IoT services;

[0028] Based on the instantaneous interruption characteristics of each of the high-risk specular reflection paths and the service priority, an independent path reliability reduction factor is generated;

[0029] Based on the independent path reliability reduction factor, the calibration parameters, and the correlation analysis results, an assessment is made to reduce the contribution of the signal components of the high-risk specular reflection path to the overall channel quality, and calibration parameters are generated for calibrating the channel assessment model.

[0030] According to some embodiments of this application, the evaluation of the transient interruption characteristics of each of the high-risk specular reflection paths includes:

[0031] Continuously monitor the received signal strength, signal phase, and signal polarization state of the high-risk specular reflection path;

[0032] Real-time acquisition of environmental parameters of the area where the terminal is located;

[0033] Correlation analysis is performed on the received signal strength, the signal phase, the signal polarization state, and the environmental parameters to identify the nonlinear coupling relationship between changes in the environmental parameters and the received signal strength, the signal phase, and the signal polarization state;

[0034] Based on the nonlinear coupling relationship, the detection threshold and evaluation weight for evaluating the instantaneous interruption characteristics of each high-risk specular reflection path are dynamically adjusted to obtain the adjusted detection threshold and evaluation weight.

[0035] Based on the adjusted detection threshold and evaluation weight, the transient interruption characteristics of each of the high-risk specular reflection paths are evaluated.

[0036] According to some embodiments of this application, the step of identifying the type of industrial IoT service transmitted through the high-risk specular reflection path and allocating service priorities based on the latency and reliability requirements of the industrial IoT service includes:

[0037] Identify multiple industrial IoT services transmitted through the high-risk specular reflection path and analyze the dependencies between these multiple industrial IoT services;

[0038] Analyze the competitive relationships among the various Industrial Internet of Things (IIoT) services mentioned above;

[0039] Based on the aforementioned dependencies, the priority of the business that is a prerequisite for the dependency is raised to a level higher than or equal to the priority of the business that depends on that prerequisite, thus obtaining the raised priority.

[0040] Based on the aforementioned competitive relationship, resource allocation is coordinated to obtain a coordinated resource allocation result.

[0041] Based on the improved priority and the coordinated resource allocation results, business priorities are assigned.

[0042] According to some embodiments of this application, the analysis of the dependencies between the various industrial IoT services includes:

[0043] Continuously monitor the data flow and control command execution status of the industrial IoT business;

[0044] Real-time acquisition of device status information, production process configuration information, and historical business dependency graphs in industrial IoT systems;

[0045] The data flow, the execution status of the control commands, the equipment status information, and the production process configuration information are correlated and analyzed to identify the dynamic dependencies between business processes.

[0046] Based on the aforementioned dependencies, a target business dependency graph is obtained;

[0047] Based on the target business dependency graph, analyze the dependencies between the various industrial IoT businesses.

[0048] According to some embodiments of this application, the step of obtaining the target service dependency graph includes one of the following:

[0049] When a new dependency is identified, the historical business dependency graph is updated to obtain the target business dependency graph.

[0050] When the original dependency relationship is identified as invalid, the invalid dependency relationship is removed from the historical business dependency graph to obtain the target business dependency graph.

[0051] Secondly, embodiments of this application provide an intelligent control system for a network transmission terminal, comprising:

[0052] The acquisition module is used to acquire terminal operation data, external environment data, and user traffic data.

[0053] The analysis module is used to perform correlation analysis on the terminal operation data, the external environment data, and the user traffic data to obtain correlation analysis results and identify the evaluation bias of the channel evaluation model.

[0054] The generation module is used to generate calibration parameters for calibrating the channel evaluation model based on the correlation analysis results.

[0055] The correction module is used to apply the calibration parameters to the channel evaluation model to correct the evaluation bias of the channel evaluation model and obtain the corrected channel evaluation result.

[0056] An adjustment module is used to adjust the transmission power based on the corrected channel assessment results.

[0057] The technical solution according to the embodiments of this application has at least the following beneficial effects: This application discloses an intelligent control method for network transmission terminals. By acquiring terminal operation data, external environment data, and user traffic data, and performing correlation analysis on these multi-source data, it can identify the evaluation bias of the channel evaluation model. Based on this, calibration parameters are generated according to the correlation analysis results and applied to the channel evaluation model to correct the model's evaluation bias and obtain a more accurate channel evaluation result. Finally, the transmission power is adjusted according to the corrected channel evaluation result. This method effectively solves the problems in the prior art where the accuracy of the channel evaluation model decreases due to environmental changes and the evolution of user behavior patterns, and the transmission power adjustment strategy is no longer optimal. By introducing multi-source data correlation analysis and model calibration mechanisms, this application can significantly improve the accuracy of channel evaluation, avoid excessive or unnecessary power adjustments, thereby reducing terminal energy consumption, extending equipment life, and reducing radio frequency noise and electromagnetic interference. It effectively solves the specific technical dilemmas of unstable transmission performance, increased energy consumption, and difficult fault diagnosis in the prior art, and achieves more stable, efficient, and reliable network transmission.

[0058] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0059] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0060] Figure 1 A flowchart illustrating a network transmission terminal intelligent control method provided in one embodiment of this application;

[0061] Figure 2 This is a schematic diagram of a network transmission terminal intelligent control system provided in one embodiment of this application. Detailed Implementation

[0062] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0063] It should be noted that the meaning of "multiple" (or "more than") in the description of the embodiments of this application refers to two or more, and "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0064] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: the existence of a alone, the existence of b alone, the existence of c alone, the simultaneous existence of a and b, the simultaneous existence of a and c, the simultaneous existence of b and c, or the simultaneous existence of a, b, and c, where a, b, and c can be single or multiple.

[0065] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0066] Based on the above, this application proposes an intelligent control method and system for network transmission terminals, aiming to improve the stability and reliability of network transmission.

[0067] The intelligent control method for network transmission terminals provided in this application can be applied to terminals, servers, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms; the software can be an application that implements the intelligent control method for network transmission terminals, but is not limited to the above forms.

[0068] This application can be applied to numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices. It should be noted that in various specific embodiments of this invention, when processing is required based on data related to the characteristics of an object (e.g., user attributes or sets of attribute information), permission or consent from the corresponding object is obtained first, and the collection, use, and processing of this data comply with relevant laws and standards. Furthermore, when the embodiments of the present invention need to obtain the attribute information of an object, they will obtain the separate permission or separate consent of the corresponding object through pop-up windows or redirection to a confirmation page. After obtaining the separate permission or separate consent of the corresponding object, they will then obtain the relevant data of the object necessary for the embodiments of the present invention to operate normally.

[0069] See Figure 1 , Figure 1This is a flowchart illustrating a network transmission terminal intelligent control method according to an embodiment of this application. The network transmission terminal intelligent control method provided in this embodiment includes, but is not limited to, steps S110 to S150, which will be described in detail below.

[0070] Step S110: Obtain terminal operation data, external environment data, and user traffic data;

[0071] Step S120: Perform correlation analysis on terminal operation data, external environment data and user traffic data to obtain correlation analysis results in order to identify the evaluation bias of the channel evaluation model;

[0072] Step S130: Based on the correlation analysis results, generate calibration parameters for calibrating the channel evaluation model;

[0073] Step S140: Apply the calibration parameters to the channel evaluation model to correct the evaluation bias of the channel evaluation model and obtain the channel evaluation result;

[0074] Step S150: Adjust the transmission power based on the channel assessment results.

[0075] It should be noted that "terminal operation data" refers to various data generated by the network transmission terminal during operation, such as CPU utilization, memory usage, battery level, temperature, received signal strength (RSSI), signal-to-noise ratio (SNR), bit error rate (BER), data throughput, connection status, and operating mode. This data reflects the terminal's internal operating status and performance. "External environment data" refers to data related to the physical environment in which the terminal is located, such as geographical location information, weather conditions, the distribution and height of surrounding buildings, vegetation cover, electromagnetic interference sources, and the activity of other wireless devices. This data affects the propagation characteristics of wireless signals. "User traffic data" refers to data traffic information generated by terminal users when using network services, such as upload / download rates, packet size, service type (e.g., video streaming, voice calls, web browsing, IoT data), traffic priority, and user behavior patterns. This data reflects the user's demand for network resources and usage habits. "Channel evaluation model" refers to a mathematical model or algorithm used to predict or estimate the quality of a wireless channel. This model is typically trained based on historical data and pre-defined propagation characteristics, aiming to output assessment results such as channel attenuation, multipath effects, and interference levels based on input parameters (e.g., distance, frequency, environment type). "Assessment bias" refers to the difference between the channel assessment results output by the channel assessment model and the actual channel conditions. When the environment changes or the model parameters become inapplicable, the assessment bias increases, causing the model to fail to accurately reflect the true channel quality. "Calibration parameters" refer to the values ​​or sets of parameters used to adjust or optimize the internal parameters of the channel assessment model. By applying calibration parameters, the model can better adapt to the current environment, reduce assessment bias, and improve assessment accuracy.

[0076] In one embodiment, acquiring terminal operating data can be achieved in several ways. For example, the terminal's internal operating system or firmware can integrate a data acquisition module to periodically read performance indicators such as CPU utilization, memory usage, signal strength, and bit error rate, and store this data in a local log file or report it to the control unit through a specific interface. Another approach is to deploy a lightweight agent program on the terminal, which is responsible for monitoring various operating parameters of the terminal in real time and packaging and sending the data to the central processing server at a preset frequency. Acquiring external environmental data relies on various sensors and external information sources. The terminal can have a built-in GPS module to obtain accurate geographical location information, and combined with pre-stored Geographic Information System (GIS) data, it can infer the distribution of buildings and topography around the terminal. In addition, it can obtain real-time weather data by interfacing with a meteorological service interface. For electromagnetic environment data, the terminal can be equipped with a spectrum sensing module to scan the signal strength and interference levels of surrounding frequency bands. Acquiring user traffic data is usually accomplished by network-side equipment or the terminal itself. For example, base station or core network equipment can monitor the data transmission volume, service type, and QoS (Quality of Service) parameters of each terminal. On the terminal side, applications or operating systems can record user data usage, including upload / download traffic, types of applications accessed, session duration, etc. This data can be aggregated and reported to the control system.

[0077] In one embodiment, the implementation of correlation analysis can be diversified. Preliminary screening and matching can be performed using a pre-defined rule engine. For example, when the bit error rate in terminal operation data continuously increases, while external environment data shows the terminal is located in an area with many high-rise buildings, and user traffic data shows a large amount of high-bandwidth traffic, the system can initially determine that there may be a channel assessment bias. Alternatively, a multiple-input multiple-output (MIMO) neural network model can be constructed, using terminal operation data, external environment data, and user traffic data as input features, and the actual bias of the channel assessment model as the output label for training. Through training with historical data, the model can learn the complex nonlinear relationships between different data dimensions, thereby accurately predicting or identifying potential biases in the channel assessment model when new data inputs are received. If the correlation analysis results indicate that the channel assessment model has a fixed bias pattern in a specific scenario, calibration parameters can also be dynamically generated using optimization algorithms. For example, a loss function can be defined that measures the difference between the channel assessment model's evaluation results and the actual channel conditions before and after applying calibration parameters. Then, through gradient descent or other optimization algorithms, the internal parameters of the channel assessment model can be iteratively adjusted or external calibration factors can be introduced to minimize the loss function. These adjusted parameters or external factors are called calibration parameters. For example, when correlation analysis results indicate that the Doppler effect causes inaccurate channel coherence time estimation by the channel assessment model in high-speed mobile scenarios, a calibration factor can be generated to adjust the Doppler spread parameter. The application of the calibration parameter then depends on its generation method and the structure of the channel assessment model. If the calibration parameter is a simple compensation factor, it can be directly added to or multiplied by the output of the channel assessment model. If the calibration parameter is an internal parameter adjustment of the channel assessment model, it needs to be updated in the model through the model's API or configuration interface. For example, if the calibration parameter adjusts the coefficients of the multipath fading model, the model will use these new coefficients for calculation each time a channel assessment is performed. By applying calibration parameters, the channel assessment model can more accurately reflect the current wireless channel conditions, thus obtaining a corrected channel assessment result. This can be based on the target received signal strength or the target signal-to-noise ratio. If the corrected channel assessment result indicates that the current channel attenuation is large, the terminal will correspondingly increase the transmission power to ensure that the receiver can reach the preset minimum received signal strength. Conversely, if the channel quality is good, the transmission power can be reduced to save energy and reduce interference with other devices. Specific adjustment algorithms can be proportional-integral-derivative (PID) controllers or lookup table-based power control strategies. For example, a power adjustment table can be preset, mapping different channel assessment results (such as signal-to-noise ratio, bit error rate level) to corresponding transmission power levels. When the channel assessment result is "poor signal-to-noise ratio," the terminal will look up the table and select a higher transmission power level for transmission.

[0078] It should be noted that the solution in this application, by introducing a refined monitoring mechanism for the high real-time data streams of the Industrial Internet of Things (IIoT), can timely capture micro-interruption events in the channel. Simultaneously, by combining the geographical location information of the terminal and the 3D model of highly reflective buildings, potential specular reflection signals are spatially reverse-engineered and verified. More importantly, by correlating spatially matching reflection signals with temporally synchronized micro-interruption event alarms, this application can accurately identify "high-risk specular reflection paths" that not only exist but also actually cause channel quality degradation. This identification method, combining temporal synchronization and spatial matching, allows the evaluation bias of the channel assessment model to be more accurately attributed to specific physical phenomena, thus overcoming the limitation of traditional correlation analysis in accurately identifying the causes of instantaneous channel interruptions in complex environments, and providing more accurate input for subsequent channel calibration. Through the above technical solution, this application can accurately identify high-risk channel interruption paths caused by specular reflection for the high real-time data streams of the Industrial Internet of Things. This significantly improves the accuracy and precision of channel assessment model bias identification, especially in complex multipath environments, effectively distinguishing between general channel fading and instantaneous channel degradation caused by specific physical structures. This provides a solid foundation for generating more accurate calibration parameters, thereby ensuring the transmission reliability and real-time performance of critical industrial IoT services, avoiding improper transmission power adjustments due to inaccurate channel assessment, and ultimately enhancing the intelligent control level of the entire network transmission system.

[0079] In one embodiment, assuming an industrial robot is performing a high-precision assembly task in a smart manufacturing factory environment, its control data stream is identified as a high-real-time data stream of the Industrial Internet of Things (IIoT). When the robot communicates with a base station, complex specular reflections may occur due to the presence of numerous metal walls and large equipment within the factory. The system initiates a preset frequency monitoring logic for the industrial robot's data stream and continuously monitors its bit error rate (BER) and packet loss rate within a first preset time window. Once the BER or packet loss rate is detected to suddenly increase and exceed a preset threshold, the system immediately generates a micro-interruption event alarm, marking a possible channel quality degradation near that time point. Simultaneously, the system acquires the industrial robot's real-time geographical location information and combines it with the 3D model of highly reflective buildings such as metal walls from the factory's geographic information system (GIS) data. Based on this information, the system reverse-engineers the theoretical signal path from the base station to the metal walls and then back to the industrial robot, predicting the arrival direction of the reflected signal. Subsequently, the system compares the actual arrival direction of the received reflected signal with the theoretically predicted arrival direction. For example, if a strong reflected signal from a specific angle is detected, and this angle matches the theoretical path from the base station to the robot via a metal wall, a potential specular reflection is initially identified. Further, the system determines whether the appearance or intensity change of this reflected signal is time-synchronized with a previously generated micro-interruption event alarm. For instance, if the appearance time of the strong reflected signal is highly consistent with the occurrence time of the micro-interruption event alarm, this specific reflection path can be confirmed as a high-risk specular reflection path. Finally, the system inputs the identified high-risk specular reflection path information, along with industrial robot operating data, factory environmental data (such as temperature, humidity, and equipment operating status), and user traffic data, into the correlation analysis model.

[0080] It should be noted that obtaining calibration parameters for high-risk specular reflection paths refers to acquiring calibration data or model parameters related to the characteristics of the high-risk specular reflection paths identified in the aforementioned correlation analysis. These parameters may include, but are not limited to, reflection intensity, delay, attenuation factor, and polarization change, with the aim of quantifying the specific impact of the path on signal transmission quality. These calibration parameters can be pre-stored in the system or dynamically generated through real-time monitoring and analysis of the characteristics of high-risk specular reflection paths. Furthermore, assessing the reduction of the contribution of signal components from high-risk specular reflection paths to overall channel quality based on calibration parameters and correlation analysis results refers to special processing of signal components originating from high-risk specular reflection paths during comprehensive channel quality evaluation. For example, based on the obtained calibration parameters, these signal components can be assigned lower weights, or a reduction factor can be introduced into the channel evaluation model to reduce their positive contribution to the overall channel quality evaluation, or even suppress them as interference sources. The aim is to more accurately reflect the actual channel quality and avoid overly optimistic or inaccurate channel evaluation results due to the existence of high-risk specular reflection paths. This application's solution specifically acquires calibration parameters for high-risk specular reflection paths during calibration parameter generation. Based on these parameters and correlation analysis results, it proactively reduces the contribution of these high-risk path signal components to the overall channel quality assessment. Because high-risk specular reflection paths often bring negative impacts such as signal fading and multipath interference, indiscriminately including their signal components in the overall channel quality assessment could lead to biases in the channel assessment model's judgment of the actual channel conditions. By reducing their contribution, this application can more accurately reflect the true channel quality, thereby making subsequent transmission power adjustments more precise and effective, and solving the aforementioned problem of potentially inaccurate channel assessments.

[0081] In one embodiment, assuming an Industrial Internet of Things (IIoT) terminal is transmitting critical production control data within an industrial park, the system identifies a high-risk specular reflection path caused by a nearby highly reflective building, resulting in micro-interruption events. To more accurately calibrate the channel evaluation model, this application first obtains the calibration parameters of the high-risk specular reflection path. Through historical data analysis or real-time measurement, it is determined that this path, under specific environmental conditions, causes an average signal strength reduction of 5 dB and introduces a 20 ns delay. When generating calibration parameters for calibrating the channel evaluation model, the system, based on these calibration parameters and correlation analysis results, reduces the weight of the signal component of the high-risk specular reflection path in the overall channel quality assessment from the default 1.0 to 0.2, or directly introduces a -5 dB attenuation factor into the channel evaluation model specifically for the signal component of this path. Therefore, even if the signal strength of this path is high, its positive contribution to the overall channel quality is significantly reduced, thus preventing the channel evaluation model from giving an overly optimistic channel quality due to misjudgment. This ensures the accuracy of subsequent transmission power adjustments and effectively improves the reliability of IIoT data transmission.

[0082] It should be noted that obtaining the spatial direction of arrival and polarization state of the received signal refers to performing spatial spectrum estimation and polarization analysis on the received wireless signal using an array antenna or multi-antenna system equipped on the terminal, to determine the direction from which the signal arrives and the direction of its electromagnetic wave vibration. For example, multiple signal classification (MUSIC) algorithms, rotationally invariant signal parameter estimation (ESPRIT) algorithms, or other beamforming techniques can be used to accurately estimate the spatial direction of arrival of the signal. Simultaneously, by analyzing the polarization components of the received signal, its polarization state can be determined, such as linear, circular, or elliptical polarization. This information helps distinguish between direct, diffuse, and specular reflection signals, thereby initially identifying potential specular reflection signals. Specifically, comparing the direction of arrival of potential specular reflection signals with the geometric reflection paths of high-reflectivity structures in the geographic information system (GIS) data, and marking high-risk specular reflection paths, involves comparing the spatial direction of arrival of the identified potential specular reflection signals with the theoretical geometric reflection paths calculated from the 3D models of high-reflectivity buildings or structures pre-stored in the GIS data. Geographic Information System (GIS) data can contain detailed information such as the height, location, and material properties of urban buildings. Using techniques like ray tracing, the theoretical reflection path from the base station to a highly reflective structure and then to the terminal can be simulated. When the arrival direction of a potential specular reflection signal closely matches a certain theoretical geometric reflection path, it can be marked as a high-risk specular reflection path.

[0083] In one embodiment, a network transmission terminal is assumed to be operating in an urban environment, equipped with an eight-element uniform linear array antenna. When the terminal receives a signal from a base station, its antenna array captures a composite signal, including direct signals and multiple reflected signals. First, the terminal's signal processing unit uses array signal processing techniques, such as subspace decomposition-based algorithms, to process the received signals to estimate the spatial direction of arrival (e.g., azimuth and elevation angles) and corresponding signal polarization state for each arriving signal. For example, if a signal is detected arriving from a specific direction and its polarization state is significantly different from that of a direct signal, the signal may be initially identified as a potential specular reflection signal. Subsequently, the terminal compares the arrival directions of these potential specular reflection signals with pre-loaded geographic information system (GIS) data. This GIS data contains precise 3D models and geometric information of all highly reflective buildings (such as glass curtain wall buildings, metal structure buildings, etc.) within the terminal's area. By calculating the theoretical geometric reflection paths from the base station to these highly reflective buildings and then back to the terminal, a series of expected arrival directions of reflected signals can be obtained. If the arrival direction of a potential specular reflection signal closely matches the arrival direction of one of the theoretical geometric reflection paths within a preset error range, then the potential specular reflection signal is identified as a "high-risk specular reflection path" caused by the highly reflective building. For example, if a potential reflection signal arrives from the northwest at a specific pitch angle, and geographic information system data shows that there is a high-rise glass curtain wall building in that direction, and its geometric reflection path also points in that direction, then the reflection signal is marked as a high-risk specular reflection path.

[0084] It should be noted that transient interruption characteristics refer to the degree and frequency of signal quality degradation or interruption caused by factors such as signal fluctuations, fading, or interference along the specular reflection path within a short period of time. This assessment can be based on continuous monitoring and correlation analysis of received signal strength, signal phase, signal polarization state, and environmental parameters, aiming to quantify the dynamic reliability performance of each specific reflection path. Secondly, data streams can be classified using deep packet inspection (DPI) or based on information such as ports and protocols to determine their respective Industrial IoT services, such as real-time control, sensor data acquisition, and video surveillance. Subsequently, based on the stringent requirements of different services for latency (e.g., millisecond-level response) and reliability (e.g., 99.999% availability), corresponding priorities are assigned to each service; for example, real-time control services are given the highest priority.

[0085] Furthermore, an independent path reliability reduction factor is generated based on the transient interruption characteristics and service priority of each high-risk specular reflection path. This reduction factor is a quantified value used to indicate how the signal component contribution of a specific high-risk specular reflection path should be adjusted during channel assessment. For example, for a path with poor transient interruption characteristics (i.e., unstable) and carrying high-priority services, its reduction factor may be large, meaning that its signal component contribution to the overall channel quality should be significantly reduced; conversely, for a relatively stable path carrying low-priority services, its reduction factor may be small. Finally, by introducing an independent path reliability reduction factor, the signal component contribution of each high-risk specular reflection path can be adjusted in a refined and dynamic manner, thereby enabling the generated calibration parameters to more accurately reflect the actual channel conditions and better serve the performance requirements of different services.

[0086] In one embodiment, assuming an Industrial Internet of Things (IIoT) scenario, such as an automated factory workshop, there exists a network transmission terminal whose communication with a base station is affected by specular reflection signals generated by a large metal structure (e.g., a robotic arm or large equipment on an automated production line). This reflection path has been identified as a high-risk specular reflection path. The system first continuously monitors the received signal strength, signal phase, and signal polarization state of this high-risk specular reflection path, and combines this with environmental parameters such as the real-time location information of mobile devices (e.g., AGVs, forklifts) within the workshop to assess the instantaneous interruption characteristics of the path. For example, when an AGV passes near the reflection path, it may cause instantaneous fading or phase change of the reflected signal, thereby identifying a high risk of instantaneous interruption for this path within a specific time period. Simultaneously, the system identifies the IIoT services transmitted through this high-risk specular reflection path as including: real-time command data for controlling the robotic arm (with extremely high requirements for latency and reliability) and sensor data for monitoring ambient temperature and humidity (with relatively lower requirements for latency and reliability). Based on these requirements, the system allocates robotic arm control command data as high-priority service and environmental monitoring data as low-priority service. Then, based on the assessed transient interruption characteristics and the assigned service priorities, the system generates independent path reliability reduction factors. For example, when the transient interruption characteristics of a high-risk specular reflection path are poor and high-priority robotic arm control commands are being transmitted, the system generates a larger reduction factor to significantly reduce the contribution of the path's signal components to the overall channel quality. Conversely, if the path is relatively stable or only transmits low-priority environmental monitoring data, a smaller reduction factor is generated. Finally, these independent path reliability reduction factors are combined with existing calibration parameters and correlation analysis results to more accurately reduce the contribution of the high-risk specular reflection path's signal components to the overall channel quality, thereby generating calibration parameters for calibrating the channel assessment model. In this way, the channel assessment model can more accurately reflect the actual channel conditions and dynamically adjust transmission power according to the priorities of different services, ensuring the stability and reliability of high-priority services while optimizing overall network resource utilization efficiency.

[0087] It should be noted that continuous monitoring of the received signal strength, signal phase, and signal polarization state of high-risk specular reflection paths refers to the continuous sampling and measurement of signals from specific high-risk specular reflection paths by the receiver of a network transmission terminal or base station. Received signal strength reflects the signal's energy level, signal phase carries sensitive information about signal propagation distance and environmental changes, and signal polarization state is closely related to the physical characteristics of the reflecting surface and the signal propagation direction. Continuous monitoring of these parameters provides the foundational data for subsequent correlation analysis. Real-time acquisition of environmental parameters in the terminal's location can be understood as collecting real-time information related to the signal propagation environment through various sensors or external data sources. For example, these environmental parameters may include, but are not limited to, temperature, humidity, rainfall, wind speed, atmospheric pressure, and the presence of moving obstacles (such as vehicles or pedestrians). Changes in these environmental parameters may directly or indirectly affect the signal propagation characteristics, thereby affecting the instantaneous interruption characteristics of the specular reflection path. Correlation analysis is performed on received signal strength, signal phase, signal polarization state, and environmental parameters to identify the nonlinear coupling relationship between changes in environmental parameters and received signal strength, signal phase, and signal polarization state. Specifically, this involves using machine learning algorithms (e.g., neural networks, support vector machines, or decision trees) or statistical analysis methods to deeply mine multidimensional data. The aim is to reveal how subtle changes in environmental parameters nonlinearly affect the received signal quality and characteristics. For example, increased humidity may lead to accelerated signal attenuation, or micro-vibrations on highly reflective building surfaces at specific wind speeds may cause rapid signal phase drift. Identifying this nonlinear coupling relationship is crucial for achieving dynamic adaptive assessment. Finally, based on adjusted detection thresholds and assessment weights, the instantaneous interruption characteristics of each high-risk specular reflection path are evaluated. This means that the final judgment and quantification of instantaneous interruptions are based on dynamically adjusted assessment standards that better reflect current environmental conditions. This assessment can more accurately reflect the true reliability of high-risk specular reflection paths, avoiding misjudgments or omissions.

[0088] In one embodiment, an Industrial Internet of Things (IIoT) terminal is located near a highly reflective building, and its data stream is transmitted through a high-risk specular reflection path. To accurately assess the transient interruption characteristics of this path, the system first continuously monitors the received signal strength, signal phase, and signal polarization state of the path. Simultaneously, environmental sensors deployed near the terminal acquire real-time environmental parameters such as temperature, humidity, and wind speed. When a sudden increase in ambient humidity is detected, the system activates a correlation analysis module. This module, based on a pre-trained machine learning model, identifies a non-linear coupling relationship between increased humidity and decreased received signal strength. Specifically, the model may find that when humidity exceeds a certain threshold, signal attenuation accelerates at a faster rate. Based on this identified non-linear coupling relationship, the system dynamically adjusts the detection threshold for assessing transient interruption characteristics. For example, in low humidity conditions, a 5dB decrease in signal strength might be considered a potential interruption; however, in high humidity environments, due to generally increased signal attenuation, the system might adjust the detection threshold to only consider an 8dB decrease in signal strength as a potential interruption to avoid false alarms. Simultaneously, the evaluation weights are adjusted accordingly. For example, in high wind speeds, the weight of signal phase fluctuations may be increased because wind can cause slight movements of the reflective surface, resulting in drastic phase changes. Based on these dynamically adjusted detection thresholds and evaluation weights, the system assesses the transient interruption characteristics of high-risk specular reflection paths. For instance, if in a high-humidity, high-wind-speed environment, the received signal strength drops by 7 dB and the signal phase fluctuates drastically, the system can accurately determine that the path is experiencing a transient interruption event based on the adjusted thresholds and weights. This dynamically adjusted evaluation method enables the system to more accurately identify and quantify transient interruptions.

[0089] It's important to clarify that identifying multiple Industrial IoT (IIoT) services transmitted via high-risk specular reflection paths and analyzing their dependencies means the system needs to accurately identify all IIoT service instances currently in use or likely to use this high-risk specular reflection path for data transmission. These services might include real-time control command transmission, sensor data acquisition, video surveillance streaming, and device status reporting. After identifying these services, further analysis of their logical relationships is needed—that is, whether the completion of one service depends on the output or status of another. This dependency can be an upward data flow dependency or a control logic dependency, aiming to ensure the correct order and integrity of service execution. Analyzing the competitive relationships between multiple IIoT services can be understood as assessing potential conflicts arising when different IIoT services share limited network resources (e.g., bandwidth, processing power, transmission time slots). For example, when multiple high-real-time services simultaneously need to transmit data via the same high-risk specular reflection path, they will compete for limited channel resources. The purpose is to identify these points of competition to provide a basis for subsequent resource allocation coordination, avoiding resource contention that could lead to performance degradation. Resource allocation coordination based on competition, resulting in a coordinated resource allocation outcome, refers to the system's optimized allocation of shared resources according to preset strategies or algorithms after identifying competitive relationships between services. For example, scheduling algorithms based on priority, fairness, or maximizing overall system throughput can be used to allocate corresponding bandwidth, latency guarantees, and other resources to each competing service. The aim is to minimize resource conflicts and improve resource utilization efficiency while satisfying service dependencies. Therefore, allocating service priorities based on the improved priorities and coordinated resource allocation results means that the final service priority allocation comprehensively considers the priorities adjusted for service dependencies and the results of resource competition coordination. This final priority guides network transmission terminals in how to schedule and process data streams from different services during actual operation, aiming to ensure the smooth execution of critical services while optimizing overall system performance.

[0090] It's important to clarify that continuously monitoring the data flow and control command execution status of Industrial IoT (IIoT) operations refers to the system uninterruptedly tracking the data packet transmission paths, data volume, transmission latency, and the sending, receiving, and execution feedback of key control commands between IIoT devices. The purpose is to obtain the real-time context of business operations, providing foundational data for subsequent dependency identification. Real-time acquisition of device status information, production process configuration information, and historical business dependency maps within the IIoT system can be understood as the system collecting real-time data on device operation status (e.g., online status, fault codes, load rate) through various channels such as sensors, SCADA systems, and MES systems; the current production line's process flow settings; and previously accumulated business dependency models. The aim is to provide multi-dimensional and comprehensive contextual information to aid in understanding business behavior. Correlation analysis of data flow, control command execution status, device status information, and production process configuration information identifies dynamically changing dependencies between businesses. Specifically, this involves using big data analytics and machine learning technologies to cross-reference and identify patterns in the massive amounts of data acquired through real-time monitoring. For example, when a device malfunctions, observing changes in its data flow and control command execution status, and combining this with production process configuration, helps determine which upstream or downstream businesses will be affected, thereby identifying new or changing dependencies. The goal is to uncover hidden, dynamically changing business relationships from complex data. Further, based on these dependencies, a target business dependency graph is obtained. This involves integrating the identified dynamically changing dependencies into the existing business dependency model to form an up-to-date business dependency graph reflecting the current system state. The aim is to provide a clear and visually representable business dependency structure, facilitating subsequent analysis and decision-making. Finally, based on the target business dependency graph, the dependencies between various industrial IoT businesses are analyzed. This means that based on the updated business dependency graph, the system can more accurately assess the degree of mutual influence, critical paths, and potential bottlenecks between different businesses. The goal is to provide a precise basis for the rational allocation of business priorities and the optimization of resource configuration.

[0091] It's important to clarify that "new dependencies" refer to interactions between business processes not yet recorded in the current historical business dependency graph, discovered during continuous monitoring of data flow and control command execution status in the Industrial Internet of Things (IIoT) business, combined with correlation analysis of equipment status information and production process configuration information. New dependencies arise when a new production task is introduced or a new data exchange link is established between existing devices. The "historical business dependency graph" can be understood as a graph-like data structure pre-stored by the system, reflecting known dependencies between IIoT businesses. This graph dynamically evolves with business operations and environmental changes, serving as the foundation for analyzing business dependencies. "Update the historical business dependency graph" means adding newly identified dependencies to the historical business dependency graph to ensure its integrity and real-time performance. This can be represented by adding new nodes or edges to the graph. "Existing dependencies becoming invalid" means that during business operations, due to factors such as production process adjustments, equipment shutdowns, communication link interruptions, or changes in business logic, some dependencies already recorded in the historical business dependency graph no longer hold true or have practical significance. For example, when a piece of equipment is replaced or a production step is canceled, its associated dependencies become invalid. "Removing invalid dependencies from the historical business dependency graph" means deleting dependencies that are no longer valid from the historical business dependency graph to avoid performing business dependency analysis based on outdated information, thereby improving the accuracy of the analysis. This can be achieved by deleting the corresponding nodes or edges in the graph.

[0092] See Figure 2 , Figure 2 This is a schematic diagram of a network transmission terminal intelligent control system provided in one embodiment of this application. The network transmission terminal intelligent control system 200 includes:

[0093] The acquisition module 210 is used to acquire terminal operating data, external environment data, and user traffic data.

[0094] Analysis module 220 is used to perform correlation analysis on terminal operation data, external environment data and user traffic data, and obtain correlation analysis results to identify the evaluation bias of the channel evaluation model.

[0095] The generation module 230 is used to generate calibration parameters for calibrating the channel evaluation model based on the correlation analysis results.

[0096] The correction module 240 is used to apply the calibration parameters to the channel evaluation model to correct the evaluation bias of the channel evaluation model and obtain the corrected channel evaluation result.

[0097] The adjustment module 250 is used to adjust the transmission power based on the corrected channel assessment results.

[0098] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0099] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, 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. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0100] The foregoing has provided a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined in this application.

Claims

1. A network transmission terminal intelligent control method, characterized in that, The method comprises the following steps: acquiring terminal running data, external environment data, and user traffic data; performing correlation analysis on the terminal running data, the external environment data, and the user traffic data to obtain a correlation analysis result to identify evaluation deviation of a channel evaluation model; generating a calibration parameter for calibrating the channel evaluation model according to the correlation analysis result; applying the calibration parameter to the channel evaluation model to correct the evaluation deviation of the channel evaluation model to obtain a channel evaluation result; adjusting transmission power according to the channel evaluation result; the correlation analysis on the terminal running data, the external environment data, and the user traffic data to obtain a correlation analysis result to identify evaluation deviation of a channel evaluation model comprises: starting preset frequency monitoring logic and monitoring bit error rate and packet loss rate within a first preset time window for high real-time data flow identified as industrial Internet of Things; generating a micro-interruption event alarm according to the bit error rate and the packet loss rate; acquiring geographical position of a terminal and a three-dimensional model of a high-reflective building in geographic information system data; performing reverse deduction on a direction of arrival of an active reflective signal according to the geographical position and the three-dimensional model; judging whether the direction of arrival of the reflective signal is consistent with a theoretical path from the terminal to a preset high-reflective building and then to a base station; judging whether the reflective signal is synchronized in time with the micro-interruption event alarm; when the direction of arrival of the reflective signal is consistent with the theoretical path and the reflective signal is synchronized in time with the micro-interruption event alarm, obtaining a high-risk mirror reflection path; obtaining a correlation analysis result to identify evaluation deviation of a channel evaluation model according to the high-risk mirror reflection path, the terminal running data, the external environment data, and the user traffic data.

2. The method of claim 1, wherein, the generation of a calibration parameter for calibrating the channel evaluation model according to the correlation analysis result comprises: acquiring a calibration parameter of a high-risk mirror reflection path; reducing evaluation of a signal component of the high-risk mirror reflection path on contribution to overall channel quality according to the calibration parameter and the correlation analysis result to generate a calibration parameter for calibrating the channel evaluation model.

3. The method of claim 2, wherein, the high-risk mirror reflection path is obtained through the following steps: acquiring a spatial direction of arrival of a received signal and a signal polarization state and marking a potential mirror reflection signal; comparing the direction of arrival of the potential mirror reflection signal with a geometric reflection path of a high-reflective structure in the geographic information system data to mark a high-risk mirror reflection path.

4. The method of claim 3, wherein, the reduction of evaluation of a signal component of the high-risk mirror reflection path on contribution to overall channel quality according to the calibration parameter and the correlation analysis result to generate a calibration parameter for calibrating the channel evaluation model comprises: evaluating instantaneous interruption characteristics of each high-risk mirror reflection path; identifying a type of industrial Internet of Things service transmitted through the high-risk mirror reflection path and assigning a service priority according to a requirement of the industrial Internet of Things service on latency and reliability; generating independent path reliability reduction factors according to the instantaneous interruption characteristics of each of the high-risk mirror reflection paths and the service priorities; reducing the evaluation of the signal component of the high-risk mirror reflection path on the overall channel quality contribution according to the independent path reliability reduction factors, the calibration parameters and the correlation analysis results, and generating calibration parameters for calibrating the channel evaluation model.

5. The method of claim 4, wherein, The evaluation of the instantaneous interruption characteristics of each of the high-risk mirror reflection paths comprises: continuously monitoring the received signal strength, signal phase and signal polarization state of the high-risk mirror reflection path; real-time acquisition of the environmental parameters of the area where the terminal is located; correlation analysis of the received signal strength, signal phase, signal polarization state and environmental parameters to identify the nonlinear coupling relationship between the changes in the environmental parameters and the received signal strength, signal phase and signal polarization state; according to the nonlinear coupling relationship, dynamically adjusting the detection threshold and evaluation weight for evaluating the instantaneous interruption characteristics of each of the high-risk mirror reflection paths to obtain the adjusted detection threshold and evaluation weight; based on the adjusted detection threshold and evaluation weight, evaluating the instantaneous interruption characteristics of each of the high-risk mirror reflection paths.

6. The method of claim 4, wherein, The identification of the type of industrial internet of things service transmitted through the high-risk mirror reflection path and the allocation of service priority according to the requirements of the industrial internet of things service on delay and reliability comprises: identifying multiple industrial internet of things services transmitted through the high-risk mirror reflection path and analyzing the dependency relationship between the multiple industrial internet of things services; analyzing the competition relationship between the multiple industrial internet of things services; according to the dependency relationship, the priority of the service as the prerequisite for dependency is raised to be higher than or equal to the priority of the service dependent on the prerequisite to obtain the raised priority; according to the competition relationship, resource allocation coordination is carried out to obtain the coordinated resource allocation result; according to the raised priority and the coordinated resource allocation result, the service priority is allocated.

7. The method of claim 6, wherein, The analysis of the dependency relationship between the multiple industrial internet of things services comprises: continuously monitoring the data flow direction and execution state of the control instruction of the industrial internet of things service; real-time acquisition of device state information, production process configuration information and historical service dependency graph in the industrial internet of things system; correlation analysis of the data flow direction, the execution state of the control instruction, the device state information and the production process configuration information to identify the dynamically changing dependency relationship between services; according to the dependency relationship, a target service dependency graph is obtained; according to the target service dependency graph, the dependency relationship between the multiple industrial internet of things services is analyzed.

8. The method of claim 7, wherein, The step of obtaining the target service dependency graph comprises one of: when a new dependency relationship is identified, updating the historical service dependency graph to obtain the target service dependency graph; when an existing dependency relationship is identified to be invalid, removing the invalid dependency relationship from the historical service dependency graph to obtain the target service dependency graph.

9. A network transmission terminal intelligent control system, characterized by, comprises: An acquisition module is configured to acquire terminal running data, external environment data, and user traffic data; An analysis module is configured to perform correlation analysis on the terminal running data, the external environment data, and the user traffic data to obtain a correlation analysis result to identify evaluation deviation of a channel evaluation model; Further configured to, for a high real-time data stream identified as an industrial Internet of Things, start preset frequency monitoring logic and monitor a bit error rate and a packet loss rate within a first preset time window; Generate a micro-interruption event alarm according to the bit error rate and the packet loss rate; Acquire a geographical position of a terminal and a three-dimensional model of a high-reflective building in geographic information system data; Backwardly deduce a direction of arrival of an active reflective signal according to the geographical position and the three-dimensional model; Determine whether the direction of arrival of the reflective signal coincides with a theoretical path from the terminal to a preset high-reflective building and then to a base station; Determine whether the reflective signal is synchronized in time with the micro-interruption event alarm; When the direction of arrival of the reflective signal coincides with the theoretical path and the reflective signal is synchronized in time with the micro-interruption event alarm, a high-risk mirror surface reflection path is obtained; Obtain a correlation analysis result according to the high-risk mirror surface reflection path, the terminal running data, the external environment data, and the user traffic data to identify evaluation deviation of a channel evaluation model; A generation module is configured to generate a calibration parameter for calibrating the channel evaluation model according to the correlation analysis result; A correction module is configured to apply the calibration parameter to the channel evaluation model to correct evaluation deviation of the channel evaluation model to obtain a corrected channel evaluation result; An adjustment module is configured to adjust transmission power according to the corrected channel evaluation result.

Images