A communication data classification management method and system based on AI analysis

Abstract

This invention, applicable to the fields of communication data scheduling and intelligent network management, provides a communication data classification management method and system based on AI analysis. The method comprises: obtaining the original node cache send window values ​​of communication nodes in a target construction area of ​​a target underground tunnel, as well as historical construction and communication data of the target underground tunnel; obtaining from this historical construction and communication data several sub-areas that match the background information of the target construction area and have completed construction, as well as local historical data corresponding to each sub-area; and extracting the construction level values ​​and effective uploaded data utilization rates of the sub-areas at different construction stages. By introducing a feature pattern recognition mechanism based on historical construction and communication data, the present invention enables dynamic and adaptive adjustment of communication parameters (node ​​cache send window values) between construction stages, significantly different from existing management methods that rely on static configuration or empirical settings.

Description

Technical Field

[0001] The present invention belongs to the technical field of communication data scheduling and intelligent network management, and in particular relates to a communication data classification management method and system based on AI analysis. Background Art

[0002] In existing communication data management technologies, most communication node configurations in construction environments still rely on static settings or manual experience to adjust parameters. This is especially true in scenarios with complex communication environments and significant construction interference, such as underground tunnels. This approach has obvious deficiencies in adaptability and real-time performance. Traditional methods often make adjustments based on local network performance only after construction begins, resulting in delayed responses and an inability to effectively resist problems such as changes in channel quality and increased reflection interference caused by evolving construction progress. This further exacerbates the risks of packet loss, delays, or upload failures during data transmission. In addition, most existing communication parameter adjustment methods have failed to form a systematic historical data-driven model, making it difficult to achieve intelligent parameter prediction and configuration across projects and stages. In particular, the node cache send window value, as a key communication scheduling parameter, is still generally configured with fixed values ​​or set empirically, lacking the ability to dynamically optimize as the environment evolves.

[0003] Especially in underground tunnel projects, the physical environment, signal paths, and obstruction conditions in which communication nodes are located vary significantly during each construction phase. This dynamic evolution directly affects communication performance, but existing technologies lack in-depth modeling and response mechanisms for this relationship between "changes in construction level and performance evolution trends." Although some studies have attempted to adopt real-time feedback adjustment methods based on network status, these methods are limited by long data accumulation cycles and large amounts of real-time analysis and computation, making rapid deployment and low-cost implementation difficult. Therefore, how to reasonably preset and dynamically modify communication parameters at each stage of project construction has become a major technical challenge facing the current communication management system. Summary of the Invention

[0004] The purpose of the present invention is to provide a communication data classification management method and system based on AI analysis, aiming to solve the problems raised in the background technology.

[0005] The present invention is implemented as follows: a communication data classification management method based on AI analysis, the method comprising:

[0006] Obtaining an original node cache sending window value of a communication node in a target construction area of ​​a target underground tunnel, as well as historical construction and communication data of the target underground tunnel;

[0007] Obtain several sub-regions that match the background information of the target construction area and have completed construction, as well as local historical data corresponding to each sub-region, from historical construction and communication data, and extract the construction degree values ​​and effective uploaded data utilization rates of the sub-regions at different construction stages;

[0008] Analyze whether there is the following abnormal characteristic pattern: as the construction degree value increases, the utilization rate of effective uploaded data decreases, and the proportion of this trend change relationship in all local historical data exceeds the preset threshold;

[0009] If an abnormal characteristic pattern is determined, determine the reference sub-area and reference historical data that best matches the construction rhythm of the target construction area, and calculate the decrease in the effective uploaded data utilization rate of the reference sub-area at different construction stages compared to the initial construction stage;

[0010] All the decrease amplitudes are converted into a set of correction factors, which are applied to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages to generate a set of corrected window values, which are then applied to parameter scheduling in communication data classification management.

[0011] As a further limitation of the technical solution of the embodiment of the present invention, the step of analyzing whether there is the following abnormal characteristic pattern: as the construction degree value increases, the effective uploaded data utilization rate shows a downward trend, and the proportion of this trend change relationship in the entire local historical data exceeds a preset threshold includes:

[0012] Analyze each set of local historical data to determine whether there is a trend in which the effective uploaded data utilization rate of its communication nodes decreases as the construction level value increases during the construction process of the sub-region;

[0013] Extract all local historical data that meet the above trend change relationship and calculate their proportion in all local historical data;

[0014] When the proportion exceeds a preset threshold, it is determined that an abnormal feature pattern exists in the sub-area that matches the background information of the target construction area.

[0015] As a further limitation of the technical solution of the embodiment of the present invention, if it is determined that an abnormal characteristic pattern exists, the steps of determining a reference sub-area and reference historical data that best matches the construction rhythm of the target construction area, and calculating the decrease in the effective uploaded data utilization rate of the reference sub-area at different construction stages compared to the initial construction stage include:

[0016] Based on the construction timeline, phase division method and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area is determined from several sub-areas with abnormal characteristic patterns, and the local historical data corresponding to the reference sub-area is set as the reference historical data;

[0017] Obtaining the effective uploaded data utilization rate of communication nodes in the reference sub-area at different construction stages and the initial construction stage;

[0018] The decrease in the effective uploaded data utilization rate in each construction phase compared to the initial construction phase is calculated in turn.

[0019] As a further limitation of the technical solution of the embodiment of the present invention, the steps of converting all the decrease amplitudes into a set of correction factors, applying the correction factors to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages, generating a corrected window value set, and applying the corrected window value set to parameter scheduling in communication data classification management include:

[0020] Establish a corresponding relationship between the target construction area and the reference sub-area in terms of construction phase, and associate each construction phase in the target construction area with the closest construction phase in the reference sub-area;

[0021] Convert the effective upload data utilization rate decrease corresponding to each construction stage in the reference sub-area into a correction factor for correcting the original node cache sending window value of the communication node in the corresponding construction stage of the target construction area;

[0022] Arrange the correction factors in sequence according to the order of the construction stages in the target construction area to form a correction factor set;

[0023] The preset window value correction function is called, and based on the correction factor set, the original node cache sending window values ​​of the communication nodes in the target construction area at each construction stage are adjusted one by one to generate a corrected node cache sending window value set, and the window value set is used for parameter scheduling in the communication data classification management process.

[0024] As a further limitation of the technical solution of the embodiment of the present invention, the window value correction function is: ;

[0025] in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

[0026] A communication data classification and management system based on AI analysis, comprising: a data acquisition module, a data analysis module, an abnormal mode determination module, a drop amplitude calculation module, and a window value correction module, wherein:

[0027] a data acquisition module for acquiring original node cache sending window values ​​of communication nodes in a target construction area of ​​a target underground tunnel, and historical construction and communication data of the target underground tunnel;

[0028] The data parsing module is used to obtain several sub-areas that match the background information of the target construction area and have completed construction, as well as the local historical data corresponding to each sub-area, from historical construction and communication data, and extract the construction degree value and effective uploaded data utilization rate of the sub-area at different construction stages;

[0029] The abnormal pattern determination module is used to analyze whether the following abnormal characteristic pattern exists: as the construction degree value increases, the utilization rate of effective uploaded data shows a downward trend, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold;

[0030] A decline amplitude calculation module is used to determine the reference sub-area and reference historical data that best matches the construction rhythm of the target construction area if an abnormal characteristic pattern is determined, and calculate the decline amplitude of the effective uploaded data utilization rate of the reference sub-area in different construction stages compared to the initial construction stage;

[0031] The window value correction module is used to convert all the decline amplitudes into a set of correction factors, which are applied to the original node cache sending window values ​​of communication nodes in the target construction area at different construction stages, to generate a corrected window value set, and apply it to parameter scheduling in communication data classification management.

[0032] As a further limitation of the technical solution of the embodiment of the present invention, the abnormal mode determination module specifically includes:

[0033] A trend change relationship analysis unit is used to analyze each set of local historical data to determine whether there is a trend change relationship in which the effective uploaded data utilization rate of its communication nodes decreases as the construction degree value increases during the construction process of the sub-region;

[0034] A proportion calculation unit is used to extract all local historical data that meet the above trend change relationship and calculate their proportion in all local historical data;

[0035] The abnormal characteristic pattern determination unit is configured to determine that an abnormal characteristic pattern exists in a sub-area that matches the background information of the target construction area when the proportion exceeds a preset threshold.

[0036] As a further limitation of the technical solution of the embodiment of the present invention, the drop amplitude calculation module specifically includes:

[0037] a reference data determination unit for determining, based on the construction timeline, stage division method, and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area from among the sub-areas with abnormal characteristic patterns, and setting the local historical data corresponding to the reference sub-area as the reference historical data;

[0038] A utilization rate extraction unit is used to obtain the effective uploaded data utilization rate of the communication nodes in the reference sub-area at different construction stages and the initial construction stage;

[0039] The reduction amplitude calculation unit is used to sequentially calculate the reduction amplitude of the effective uploaded data utilization rate in each construction phase compared with the initial construction phase.

[0040] As a further limitation of the technical solution of the embodiment of the present invention, the window value correction module specifically includes:

[0041] The construction phase correspondence unit is used to establish a correspondence between the target construction area and the reference sub-area in terms of construction phase, and associate each construction phase in the target construction area with the closest construction phase in the reference sub-area;

[0042] A correction factor conversion unit is used to convert the effective upload data utilization rate reduction amplitude corresponding to each construction stage in the reference sub-area into a correction factor for correcting the original node cache sending window value of the communication node in the corresponding construction stage in the target construction area;

[0043] A set generating unit is used to arrange the correction factors in sequence according to the order of the construction stages in the target construction area to form a correction factor set;

[0044] The window value correction unit is used to call a preset window value correction function and, based on a set of correction factors, adjust the original node cache sending window values ​​of the communication nodes in the target construction area at each construction stage one by one, generate a corrected node cache sending window value set, and use the window value set for parameter scheduling in the communication data classification management process.

[0045] As a further limitation of the technical solution of the embodiment of the present invention, the window value correction function is: ;

[0046] in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] By introducing a characteristic pattern recognition mechanism based on historical construction and communication data, this invention enables dynamic and adaptive adjustment of communication parameters (node ​​cache send window values) between construction phases, significantly different from existing management methods that rely on static configuration or empirical settings. Specifically, for scenarios where there is a trend of "increased construction level and decreased communication performance," this invention innovatively constructs a quantitative mapping relationship between the amplitude of the decrease in the reference sub-region and the window value of the target construction area. By adjusting the cache send window configuration stage by stage using correction factors, it effectively improves the scheduling flexibility and stability of data uploads. Furthermore, because the correction factor set can be generated in advance based on historical data before construction begins, the system can implement pre-planned parameters for the construction process, reducing post-operation and maintenance intervention and enhancing the forward-looking and adaptable nature of the scheduling strategy. This mechanism not only has good engineering feasibility but also significantly enhances the intelligent management capabilities of communication nodes in complex construction environments such as underground tunnels, possessing outstanding practical value and promotional prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of a method provided by an embodiment of the present invention;

[0050] Figure 2 A flowchart of determining whether a sub-region has an abnormal characteristic pattern in the method provided in an embodiment of the present invention;

[0051] Figure 3 A flow chart of calculating the decrease in utilization rate of effective uploaded data in the method provided in an embodiment of the present invention;

[0052] Figure 4 A flowchart of correcting the original node cache sending window value in the method provided in an embodiment of the present invention;

[0053] Figure 5 An application architecture diagram of the system provided by an embodiment of the present invention;

[0054] Figure 6 A structural block diagram of an abnormal mode determination module in a system provided by an embodiment of the present invention;

[0055] Figure 7 A structural block diagram of a drop amplitude calculation module in a system provided by an embodiment of the present invention;

[0056] Figure 8 This is a structural block diagram of a window value correction module in a system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0059] Specifically, a communication data classification management method based on AI analysis includes the following steps:

[0060] Step S100: obtaining the original node cache sending window value of the communication node in the target construction area of ​​the target underground tunnel, and the historical construction and communication data of the target underground tunnel.

[0061] In an embodiment of the present invention, step S100 is used to obtain the initial input basis for the classification management of communication data, and clarifies the basic communication parameters and historical environment evolution information that are strongly associated with the target scenario. The object "target underground tunnel" described in this step specifically refers to a closed or semi-closed communication environment in a construction scenario, which is widely present in underground construction projects such as urban subway tunnels, underground integrated pipeline corridors, and mines. Compared with conventional ground communication scenarios, underground tunnels have obvious structural evolution, strong communication reflectivity, and complex construction interference factors. Therefore, the present invention preferably takes underground tunnels as a representative scenario, and on this basis constructs a communication data classification management method that is more universal and requires no correction.

[0062] The "target construction area" refers to the area within the target underground tunnel that is not yet completed and is about to enter the construction phase. Unlike the completed construction area, this area does not yet have the conditions for full communication deployment, and its communication behavior is directly affected by the progress of construction. Therefore, the actual communication parameter settings in this area are often based on preset templates or empirical initial values, which are difficult to dynamically adapt to upcoming environmental changes.

[0063] Communication nodes refer to various network terminal devices deployed within the target construction area and used to collect, upload, or relay communication data, including but not limited to Mesh relay devices, LoRa base stations, and wireless data collection terminals. A communication node can be a single key control point or a group of multiple nodes within a coverage area. The specific number can be determined based on the communication network design logic and is not limited.

[0064] The original node cache sending window value is a key parameter preset by the communication node in the initial state and used to control its data cache sending strategy. This value is usually used to limit the maximum amount of data cache that the node is allowed to release per unit time. It is one of the important scheduling parameters in the wireless communication protocol to control the load release rhythm and avoid instantaneous congestion. In existing communication systems, this parameter can be automatically generated by the communication scheduling module according to the standard template through network configuration, device firmware initialization, or by the communication scheduling module. It is a configuration content that is maturely applied in existing technologies. However, because this value is mostly based on static settings, it does not fully consider the long-term impact of the subsequent construction process on communication stability, link quality and retransmission mechanism, which can easily lead to resource allocation lags or distortion of classification results. Therefore, the present invention proposes to achieve dynamic optimization and adaptive update of this value by introducing a correction mechanism driven by historical data.

[0065] Historical construction and communication data for a target underground tunnel refers to structural evolution records and communication behavior data collected from completed construction areas within the tunnel, comparable in background conditions to the current target construction area. This data can be extracted from construction monitoring systems, environmental sensor networks, historical communication logs, or equipment operation records. This historical construction and communication data includes at least the following types of data: construction phase and time series markers, construction completion indicators, physical information such as structural density or material reflectance coefficients; it should also include communication stability indicators, data upload success rates, number of retransmissions, channel quality feedback, and effective data packet upload ratios for communication node performance during the corresponding phase.

[0066] Furthermore, the communication data classification management method based on AI analysis also includes the following steps:

[0067] Step S200: Obtain several sub-areas that match the background information of the target construction area and have completed construction, and local historical data corresponding to each sub-area from historical construction and communication data, and extract the construction degree value and effective uploaded data utilization rate of the sub-area at different construction stages.

[0068] In an embodiment of the present invention, step S200 is used to construct a historical sample set with reference value for the target construction area, thereby providing data support for the subsequent identification of communication performance evolution trends and the construction of correction factors. This step first selects several sub-areas that match the background information of the target construction area from the historical construction and communication data of the target underground tunnel. Background information matching refers to the priority selection of sub-areas with high similarity or identical engineering parameters with the target construction area in terms of structural layout, construction process, materials used, layout plan, design width, construction time period, construction team configuration, or communication deployment strategy among multiple candidate sub-areas.

[0069] To achieve this matching process, a set of matching feature vectors can be constructed based on information such as the tunnel project's structural construction BIM model, phased working condition records, material inflow and outflow logs, construction schedules, and communication equipment deployment drawings. For example, several dimensions (such as structural cross-sectional dimensions, typical construction techniques, type of waterproof lining used, and initial number of communication nodes) are defined. Parameters are quantified for the target construction area and each historical sub-area. The degree of matching is calculated using methods such as Euclidean distance, weighted matching score, or cosine similarity. Finally, sub-areas with matching scores above a preset threshold are selected as reference objects. The selected sub-areas must have completed construction and accumulated stable communication data to ensure that the corresponding communication behavior samples possess credible evolutionary characteristics.

[0070] After selecting the reference sub-region, the local historical data of each sub-region at different construction stages are further extracted. Among them, the construction stage can be divided based on time series or key construction nodes, such as the tunnel excavation stage, the primary support completion stage, the waterproof layer laying stage, the secondary lining completion stage, the power and communication pipeline installation stage, etc. For each stage, the information such as the structural completion ratio, structural occupation volume, and steel cage closure rate in the time period are extracted, and these indicators are standardized and used as the "construction degree value". The construction degree value can be obtained by calculation through the construction data in the project progress management system or the BIM model. It is a parameter that has been defined in the engineering management system. In the present invention, it is directly used to describe the evolution process of the spatial structure closure and the density of interference objects.

[0071] At the same time, the communication node's "effective upload data utilization rate" is extracted for each stage. This is the ratio of the total number of valid data packets successfully uploaded to the target communication control node per unit time to the total number of data packets planned to be sent. This metric reflects the node's upload performance under construction interference conditions and is commonly used in communication network operation monitoring and equipment operation logs. It is also a common quality evaluation parameter in existing communication systems. In the present invention, this utilization rate can be extracted from logs generated by communication equipment, protocol layer ACK / NAK feedback records, and statistical data from the upper control platform.

[0072] Furthermore, the communication data classification management method based on AI analysis also includes the following steps:

[0073] Step S300 , analyzing whether the following abnormal characteristic pattern exists: as the construction degree value increases, the utilization rate of the effective uploaded data shows a downward trend, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold.

[0074] Specifically, Figure 2 A flowchart is shown for determining whether a sub-region has an abnormal characteristic pattern.

[0075] The analysis of whether there is the following abnormal characteristic pattern: as the construction degree value increases, the utilization rate of effective uploaded data shows a downward trend, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold specifically includes the following steps:

[0076] Step S301: Analyze each set of local historical data to determine whether there is a trend relationship in which the effective uploaded data utilization rate of its communication nodes decreases as the construction level value increases during the construction process of the sub-region;

[0077] Step S302: extract all local historical data that satisfy the above trend change relationship and calculate their proportion in all local historical data;

[0078] Step S303: When the proportion exceeds a preset threshold, it is determined that an abnormal characteristic pattern exists in the sub-region matching the background information of the target construction area.

[0079] In an embodiment of the present invention, steps S301 to S303 constitute an analysis mechanism for the potential correlation trend between communication performance and structural evolution. The core is to identify whether there is a statistically significant abnormal characteristic pattern, that is, the trend relationship of "as the construction degree value increases, the effective uploaded data utilization rate decreases accordingly". This characteristic pattern is not universal in all construction areas, but tends to appear in sub-areas with specific structural characteristics or construction rhythms, especially those areas where the structural airtightness increases rapidly, the number of reflective surfaces increases significantly, or the construction interference fluctuates violently. In the process of construction evolution in these areas, the complexity of the communication signal propagation path increases sharply, resulting in a decline in link quality and an increase in interference frequency, which causes the upload efficiency of the communication node to be significantly reduced.

[0080] Once this abnormal characteristic pattern appears in large numbers in historical sub-regions that highly match the background information of the target construction area, it means that the target area is likely to show a similar trend of communication performance degradation during its subsequent construction evolution. In other words, this characteristic pattern is an "evolutionary precursor" for communication performance vulnerabilities in a specific environmental category, and has a priori risk identification significance. Therefore, before implementing the present invention, it is necessary to first determine whether the characteristic pattern is widely present in similar areas, so as to decide whether to enable the correction mechanism. If all areas are corrected blindly, not only will redundant calculations be introduced, but the original regional configuration with stable communication may also be destroyed.

[0081] To ensure objectivity and statistical significance, step S302 introduces the "percentage of all local historical data" as a quantitative judgment criterion. This percentage indicates the proportion of all analyzed historical subregions that actually exhibit the abnormal trend, providing reliable data support. The "preset threshold" described in step S303 is the boundary condition for determining whether an abnormality exists.

[0082] This preset threshold can be set in a variety of ways. For example, based on empirical findings, when the proportion of abnormal trend sub-areas exceeds 60%, the risk of communication performance degradation increases significantly. It can also be based on historical project data and statistical learning methods to select the optimal balance point between misjudgment and missed judgment as the threshold. In addition, the data evolution process under different construction conditions can be simulated to determine the minimum proportion value that causes communication instability.

[0083] By following these steps and determining whether abnormal characteristic patterns exist before implementing corrections, the system’s sensitivity and robustness to abnormal trends can be effectively improved, ensuring that the correction mechanism is only activated when absolutely necessary, and avoiding misadjustment interference in non-sensitive areas.

[0084] Furthermore, the communication data classification management method based on AI analysis also includes the following steps:

[0085] In step S400, if it is determined that an abnormal characteristic pattern exists, a reference sub-area and reference historical data that best match the construction rhythm of the target construction area are determined, and the decrease in the effective uploaded data utilization rate of the reference sub-area in different construction stages compared to the initial construction stage is calculated.

[0086] Specifically, Figure 3 A flow chart for calculating the effective upload data utilization rate reduction is shown.

[0087] If an abnormal characteristic pattern is determined, the reference sub-area and reference historical data that best match the construction rhythm of the target construction area are determined, and the decrease in the effective uploaded data utilization rate of the reference sub-area at different construction stages compared to the initial construction stage is calculated. Specifically, the following steps are included:

[0088] Step S401: Based on the construction timeline, phase division method, and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area is determined from among the sub-areas with abnormal characteristic patterns, and the local historical data corresponding to the reference sub-area is set as the reference historical data;

[0089] Step S402, obtaining the effective uploaded data utilization rate of the communication nodes in the reference sub-area at different construction stages and the initial construction stage;

[0090] Step S403 , sequentially calculating the decrease in the effective uploaded data utilization rate in each construction phase compared to the initial construction phase.

[0091] In this embodiment of the present invention, step S401 constructs a target construction rhythm vector by integrating information such as the target construction area's construction timeline, phase division, and construction progress curve through time series matching analysis and rhythm similarity measurement. Subsequently, historical construction rhythm feature vectors are extracted from sub-regions with abnormal characteristic patterns. These feature vectors are then compared against the target rhythm vector using dynamic time warping (DTW) or a similarity scoring function. The sub-region with the highest similarity score is selected as the reference sub-region, and its local historical data is set as the reference historical data. This approach ensures that the reference sub-region maintains a high degree of consistency with the target region in terms of construction advancement methods and node deployment evolution paths, thereby improving the adaptability and accuracy of subsequent corrections.

[0092] In step S403, the effective uploaded data utilization rate at each construction stage of the reference sub-area is compared with the utilization rate during the initial construction stage to obtain the utilization rate decline rate for each stage. This decline rate quantitatively characterizes the attenuation trend of communication performance due to changes in the construction environment, reflecting the combined impact of factors such as reflection interference and channel quality changes on communication nodes at different stages. This provides a realistic and quantifiable reference for correcting communication parameters within the target construction area corresponding to the reference sub-area.

[0093] Furthermore, the communication data classification management method based on AI analysis also includes the following steps:

[0094] Step S500 converts all the decrease amplitudes into a set of correction factors, applies them to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages, generates a corrected window value set, and applies it to parameter scheduling in communication data classification management.

[0095] Specifically, Figure 4 A flow chart of correcting the original node cache sending window value is shown.

[0096] The steps of converting all the drop amplitudes into a set of correction factors, applying them to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages, generating a set of corrected window values, and applying them to parameter scheduling in communication data classification management are as follows:

[0097] Step S501: establishing a corresponding relationship between the target construction area and the reference sub-area in terms of construction phases, and associating each construction phase in the target construction area with the closest construction phase in the reference sub-area;

[0098] Step S502: Convert the effective upload data utilization rate decrease values ​​corresponding to each construction stage in the reference sub-area into correction factors for correcting the original node cache sending window values ​​of the communication nodes in the target construction area at the corresponding construction stage.

[0099] Step S503, arranging the correction factors in sequence according to the order of the construction phases in the target construction area to form a correction factor set;

[0100] Step S504, calling the preset window value correction function, and based on the correction factor set, adjusting the original node cache sending window value of the communication node in the target construction area at each construction stage one by one, generating a corrected node cache sending window value set, and using the window value set for parameter scheduling in the communication data classification management process.

[0101] The window value correction function is: ;

[0102] in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

[0103] In this embodiment of the present invention, step S501 establishes a construction phase mapping relationship, matching each construction phase to be revised in the target construction area with completed phases in the reference sub-area for which communication performance evolution records exist. This matching can be based on dimensions such as structural completion ratio, phase start and end times, operation process type, and construction density. Weighted similarity analysis or time alignment algorithms (such as dynamic time warping) are used to achieve optimal correspondence between phases. This ensures that each target construction phase can find a reference phase in the historical data with the most similar structural conditions and communication characteristics, thereby establishing a set of structure-communication behavior mapping relationships.

[0104] In step S502, for each reference construction phase, the magnitude of the decrease in effective upload data utilization compared to the initial phase is extracted and converted into a correction factor for the communication node buffer send window value at the corresponding construction phase in the target construction area. The correction logic employed by this invention uses the "performance degradation magnitude" of the reference sub-region as a derivation basis to feedforward-enhance the "parameter input strength" of the target area. In other words, the greater the degradation in upload performance in the reference sub-region at a given phase, the more complex the communication environment under these structural conditions. Therefore, the buffer send window configuration of the nodes in the target area should be appropriately tightened in the corresponding phase to reduce the risk of link congestion and improve communication stability, thereby offsetting potential interference risks. This mechanism establishes a correction model that "infers input enhancement based on performance deviation." This proactive optimization of communication resource allocation based on historical response results is achieved without the need for real-time monitoring of communication performance, and is one of the key technical innovations of this invention.

[0105] In step S503, all calculated correction factors are arranged according to the phase order of the target construction area to construct a correction factor set for use in subsequent phases. In step S504, the system calls a preset window value correction function and, in combination with this correction factor set, individually corrects the original node cache send window value at each phase. Ultimately, a corrected window value set is formed and used in the parameter scheduling phase of actual communication data classification management, improving scheduling accuracy and system stability.

[0106] The correction method adopted by the present invention has significant novelty and beneficial effects. On the one hand, existing technologies mostly rely on real-time monitoring of the operating status of communication equipment or static configuration templates, and lack feedforward perception of potential communication risks brought about by structural evolution; on the other hand, conventional optimization strategies are often only based on responsive adjustments based on network-side load balancing or signal feedback, making it difficult to perform proactive configuration in advance. The present invention, by constructing a performance-parameter mapping relationship driven by historical data, achieves advance optimization of key communication scheduling parameters before the evolution of structural interference is fully apparent, effectively improving the robustness and adaptability of the system, and is particularly suitable for construction scenarios where communication conditions evolve drastically, such as underground tunnels.

[0107] The window value correction logic employed in this invention uses a correction factor based on the decrease in the effective upload data utilization rate of the reference sub-region during each construction phase compared to the initial phase. The more significant the decrease, the smaller the algebraic value of the correction factor (tending toward negative values), and the corresponding larger the window value adjustment. This ensures that the send rate is proactively reduced when communication performance deteriorates, thereby improving link stability and communication reliability. This function incorporates an adjustable weight coefficient to flexibly control the correction magnitude to adapt to the scheduling requirements of different project scenarios or communication strategies.

[0108] Specifically, when the effective upload data utilization rate during a construction phase falls below the initial level of the reference sub-region, indicating a decrease in communication performance, the system automatically tightens the node cache send window configuration in the target region during that phase. By reducing the window value, the system reduces data transmission pressure and the risk of network congestion, thereby improving communication stability and transmission reliability during the classified scheduling process.

[0109] Further, Figure 5 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0110] Among them, in another preferred embodiment provided by the present invention, a communication data classification management system based on AI analysis includes:

[0111] The data acquisition module 100 is used to acquire the original node cache sending window value of the communication node in the target construction area of ​​the target underground tunnel, and the historical construction and communication data of the target underground tunnel.

[0112] Furthermore, the communication data classification management system based on AI analysis also includes:

[0113] The data analysis module 200 is used to obtain several sub-areas that match the background information of the target construction area and have completed construction and the local historical data corresponding to each sub-area from historical construction and communication data, and extract the construction degree value and effective uploaded data utilization rate of the sub-area at different construction stages.

[0114] Furthermore, the communication data classification management system based on AI analysis also includes:

[0115] The abnormal pattern determination module 300 is used to analyze whether the following abnormal characteristic pattern exists: as the construction degree value increases, the effective uploaded data utilization rate shows a downward trend, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold.

[0116] Specifically, Figure 6 FIG. 3 is a structural block diagram of the abnormal mode determination module 300 in the system provided by an embodiment of the present invention.

[0117] In a preferred embodiment of the present invention, the abnormal mode determination module 300 specifically includes:

[0118] The trend change relationship analysis unit 301 is used to analyze each set of local historical data to determine whether there is a trend change relationship in which the effective uploaded data utilization rate of the communication node decreases as the construction level value increases during the construction process of the sub-region;

[0119] A proportion calculation unit 302 is used to extract all local historical data that meet the above trend change relationship and calculate their proportion in all local historical data;

[0120] The abnormal characteristic pattern determination unit 303 is configured to determine that an abnormal characteristic pattern exists in the sub-region matching the background information of the target construction region when the proportion exceeds a preset threshold.

[0121] Furthermore, the communication data classification management system based on AI analysis also includes:

[0122] The decline amplitude calculation module 400 is used to determine the reference sub-area and reference historical data that best match the construction rhythm of the target construction area if an abnormal characteristic pattern is determined to exist, and calculate the decline amplitude of the effective uploaded data utilization rate of the reference sub-area in different construction stages compared to the initial construction stage.

[0123] Specifically, Figure 7 FIG. 4 is a structural block diagram of a drop amplitude calculation module 400 in a system provided by an embodiment of the present invention.

[0124] In a preferred embodiment of the present invention, the drop amplitude calculation module 400 specifically includes:

[0125] The reference data determination unit 401 is configured to determine, based on the construction timeline, phase division method, and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area from among the sub-areas with abnormal characteristic patterns, and set the local historical data corresponding to the reference sub-area as the reference historical data;

[0126] The utilization rate extraction unit 402 is used to obtain the effective upload data utilization rate of the communication nodes in the reference sub-area at different construction stages and the initial construction stage;

[0127] The reduction amplitude calculation unit 403 is used to sequentially calculate the reduction amplitude of the effective uploaded data utilization rate in each construction phase compared with the initial construction phase.

[0128] Furthermore, the communication data classification management system based on AI analysis also includes:

[0129] The window value correction module 500 is used to convert all the decline amplitudes into a set of correction factors, apply them to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages, generate a corrected window value set, and apply it to parameter scheduling in communication data classification management.

[0130] Specifically, Figure 8 FIG. 5 shows a structural block diagram of a window value correction module 500 in a system provided by an embodiment of the present invention.

[0131] In a preferred embodiment of the present invention, the window value correction module 500 specifically includes:

[0132] The construction phase correspondence unit 501 is used to establish a correspondence between the target construction area and the reference sub-area in terms of construction phase, and associate each construction phase in the target construction area with the closest construction phase in the reference sub-area;

[0133] The correction factor conversion unit 502 is used to convert the effective upload data utilization rate reduction amplitude corresponding to each construction stage in the reference sub-area into a correction factor for correcting the original node cache sending window value of the communication node in the target construction area corresponding to the construction stage;

[0134] A set generating unit 503 is configured to sequentially arrange the correction factors according to the order of the construction phases in the target construction area to form a correction factor set;

[0135] The window value correction unit 504 is used to call a preset window value correction function, and based on the correction factor set, adjust the original node cache sending window value of the communication node in the target construction area at each construction stage one by one, generate a corrected node cache sending window value set, and use the window value set for parameter scheduling in the communication data classification management process.

[0136] The window value correction function is: ;

[0137] in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

[0138] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0139] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0140] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0142] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A communication data classification management method based on AI analysis, characterized in that: The method comprises: Obtaining an original node cache sending window value of a communication node in a target construction area of ​​a target underground tunnel, as well as historical construction and communication data of the target underground tunnel; Obtain several sub-regions that match the background information of the target construction area and have completed construction, as well as local historical data corresponding to each sub-region, from historical construction and communication data, and extract the construction degree values ​​and effective uploaded data utilization rates of the sub-regions at different construction stages; Analyze whether there is the following abnormal characteristic pattern: as the construction degree value increases, the utilization rate of effective uploaded data decreases, and the proportion of this trend change relationship in all local historical data exceeds the preset threshold; If an abnormal characteristic pattern is determined, determine the reference sub-area and reference historical data that best matches the construction rhythm of the target construction area, and calculate the decrease in the effective uploaded data utilization rate of the reference sub-area at different construction stages compared to the initial construction stage; Convert all the drop amplitudes into a set of correction factors, apply them to the original node cache sending window values ​​of the communication nodes in the target construction area at different construction stages, generate a set of corrected window values, and apply them to parameter scheduling in communication data classification management; Establish a corresponding relationship between the target construction area and the reference sub-area in terms of construction phase, and associate each construction phase in the target construction area with the closest construction phase in the reference sub-area; Convert the effective upload data utilization rate decrease corresponding to each construction stage in the reference sub-area into a correction factor for correcting the original node cache sending window value of the communication node in the corresponding construction stage of the target construction area; Arrange the correction factors in sequence according to the order of the construction stages in the target construction area to form a correction factor set; The preset window value correction function is called, and based on the correction factor set, the original node cache sending window values ​​of the communication nodes in the target construction area at each construction stage are adjusted one by one to generate a corrected node cache sending window value set, and the window value set is used for parameter scheduling in the communication data classification management process.

2. The communication data classification management method based on AI analysis according to claim 1 is characterized in that: The steps of analyzing whether there is the following abnormal characteristic pattern: as the construction degree value increases, the utilization rate of effective uploaded data decreases, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold include: Analyze each set of local historical data to determine whether there is a trend in which the effective uploaded data utilization rate of its communication nodes decreases as the construction level value increases during the construction process of the sub-region; Extract all local historical data that meet the above trend change relationship and calculate their proportion in all local historical data; When the proportion exceeds a preset threshold, it is determined that an abnormal feature pattern exists in the sub-area that matches the background information of the target construction area.

3. The communication data classification management method based on AI analysis according to claim 2 is characterized in that: If an abnormal characteristic pattern is determined, the steps of determining a reference sub-area and reference historical data that best matches the construction rhythm of the target construction area, and calculating the decrease in the effective uploaded data utilization rate of the reference sub-area at different construction stages compared to the initial construction stage include: Based on the construction timeline, phase division method and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area is determined from several sub-areas with abnormal characteristic patterns, and the local historical data corresponding to the reference sub-area is set as the reference historical data; Obtaining the effective uploaded data utilization rate of communication nodes in the reference sub-area at different construction stages and the initial construction stage; The decrease in the effective uploaded data utilization rate in each construction phase compared to the initial construction phase is calculated in turn.

4. The communication data classification management method based on AI analysis according to claim 1 is characterized in that: The window value correction function is: ； in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

5. A communication data classification management system based on AI analysis, characterized in that: The system includes: a data acquisition module, a data analysis module, an abnormal mode determination module, a drop amplitude calculation module and a window value correction module, wherein: a data acquisition module for acquiring original node cache sending window values ​​of communication nodes in a target construction area of ​​a target underground tunnel, and historical construction and communication data of the target underground tunnel; The data parsing module is used to obtain several sub-areas that match the background information of the target construction area and have completed construction, as well as the local historical data corresponding to each sub-area, from historical construction and communication data, and extract the construction degree value and effective uploaded data utilization rate of the sub-area at different construction stages; The abnormal pattern determination module is used to analyze whether the following abnormal characteristic pattern exists: as the construction degree value increases, the utilization rate of effective uploaded data shows a downward trend, and the proportion of this trend change relationship in all local historical data exceeds a preset threshold; A decline amplitude calculation module is used to determine the reference sub-area and reference historical data that best matches the construction rhythm of the target construction area if an abnormal characteristic pattern is determined, and calculate the decline amplitude of the effective uploaded data utilization rate of the reference sub-area in different construction stages compared to the initial construction stage; The window value correction module is used to convert all the drop amplitudes into a set of correction factors, which are applied to the original node cache sending window values ​​of communication nodes in the target construction area at different construction stages. The corrected window value set is generated and applied to the parameter scheduling in the communication data classification management; The window value correction module specifically includes: The construction phase correspondence unit is used to establish a correspondence between the target construction area and the reference sub-area in terms of construction phase, and associate each construction phase in the target construction area with the closest construction phase in the reference sub-area; A correction factor conversion unit is used to convert the effective upload data utilization rate reduction amplitude corresponding to each construction stage in the reference sub-area into a correction factor for correcting the original node cache sending window value of the communication node in the corresponding construction stage in the target construction area; A set generating unit is used to arrange the correction factors in sequence according to the order of the construction stages in the target construction area to form a correction factor set; The window value correction unit is used to call a preset window value correction function and, based on a set of correction factors, adjust the original node cache sending window values ​​of the communication nodes in the target construction area at each construction stage one by one, generate a corrected node cache sending window value set, and use the window value set for parameter scheduling in the communication data classification management process.

6. The communication data classification management system based on AI analysis according to claim 5 is characterized in that: The abnormal mode determination module specifically includes: A trend change relationship analysis unit is used to analyze each set of local historical data to determine whether there is a trend change relationship in which the effective uploaded data utilization rate of its communication nodes decreases as the construction degree value increases during the construction process of the sub-region; A proportion calculation unit is used to extract all local historical data that meet the above trend change relationship and calculate their proportion in all local historical data; The abnormal characteristic pattern determination unit is configured to determine that an abnormal characteristic pattern exists in a sub-area that matches the background information of the target construction area when the proportion exceeds a preset threshold.

7. The communication data classification management system based on AI analysis according to claim 6 is characterized in that: The decline amplitude calculation module specifically includes: a reference data determination unit for determining, based on the construction timeline, stage division method, and construction progress characteristics of the target construction area, a reference sub-area that best matches the construction rhythm of the target construction area from among the sub-areas with abnormal characteristic patterns, and setting the local historical data corresponding to the reference sub-area as the reference historical data; A utilization rate extraction unit is used to obtain the effective uploaded data utilization rate of the communication nodes in the reference sub-area at different construction stages and the initial construction stage; The reduction amplitude calculation unit is used to sequentially calculate the reduction amplitude of the effective uploaded data utilization rate in each construction phase compared with the initial construction phase.

8. The communication data classification management system based on AI analysis according to claim 7 is characterized in that: The window value correction function is: ； in, Refers to the target construction area The modified node cache sending window value in the construction phase, Refers to the initial node cache sending window value, Refers to the reference sub-area and the target construction area The effective upload data utilization rate of the construction phase corresponding to each construction phase, Refers to the effective upload data utilization rate during the initial construction phase of the reference sub-area. Refers to the target construction area Correction factors for each construction phase, is the adjustment coefficient of the correction factor, and Greater than 0.

Images