Internet of things card package dynamic configuration method based on intelligent strategy regulation and control

Abstract

The invention belongs to the technical field of digital information transmission, and discloses an Internet of Things card package dynamic configuration method based on intelligent strategy regulation and control, which comprises the following steps: acquiring multi-source operation data of an Internet of Things terminal, and generating a state label with a unified code by constructing a segmented threshold mapping structure and identifying the operation state of the terminal; dynamically generating a strategy configuration result containing recommended network card package parameters, an adjusting elastic boundary and a confidence weight by combining a rule matching process and a trend evolution mechanism based on the state label; identifying a terminal cluster with service business correlation based on a strategy configuration result, generating a group network card package sharing configuration strategy through an equipment generality screening structure, a sharing strategy triggering mechanism and a resource allocation control flow, and synchronously managing a network card package execution state of a terminal in a group; and the user experience and the overall service quality and efficiency of an operator are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital information transmission, and more specifically, to a method for dynamically configuring an Internet of Things card package based on intelligent policy regulation. Background Art

[0002] Patent publication number CN111343100A discloses a speed limiting method, system and device based on the monthly traffic package of the Internet of Things card. The method includes the following steps: dividing the speed-limited traffic interval and the non-speed-limited traffic interval according to the total traffic of the monthly traffic package of the Internet of Things card; dividing the speed-limited traffic interval into multiple traffic ladder intervals; configuring a corresponding bandwidth rate for each traffic ladder interval; if the used traffic value in the current monthly traffic package of the Internet of Things card enters the speed-limited traffic interval, the bandwidth rate of the Internet of Things card is adjusted to the bandwidth rate corresponding to the traffic ladder interval in which the used traffic value in the monthly traffic package of the current Internet of Things card is located. The above patent does not limit the speed of the Internet of Things card in the initial stage as needed, so as not to affect the user experience. In the later stages, the bandwidth rate of the Internet of Things card is gradually increased to control costs, thereby maximizing the cost-effectiveness of the traffic package, and can effectively help the device control large traffic requests.

[0003] The existing method for dynamically configuring IoT card smart packages has the following drawbacks: Existing technologies typically configure packages based on static grouping or manual rules. This makes it difficult to dynamically identify service relevance between terminals based on the actual service needs and policy configurations of terminal devices, resulting in inefficient sharing policy configuration and severe resource waste. Existing sharing policies typically use simple traffic sums or single-point capacity as trigger conditions, lacking a dual assessment of group size and overall capacity requirements. This can lead to delayed or false triggering of sharing policies, or even irrational configurations triggered by extreme terminal traffic. Traditional IoT package allocations typically use static ratios and fail to dynamically adjust based on actual terminal demand. This results in resource constraints on some terminals and idle resources in others, reducing resource utilization.

[0004] Traditional package configuration methods rely on fixed configurations or simple priorities for terminal access frequency bands, making it difficult to dynamically select the optimal frequency band based on real-time network conditions. This results in poor resource utilization and a limited user experience. Existing technologies typically focus solely on static rate calculations, ignoring the discrepancy between actual terminal communication behavior and expected package rates, as well as the additional energy costs associated with frequency band switching. This leads to inaccurate rate allocation, impacting operator revenue and user fairness.

[0005] While frequency band switching facilitates load balancing and signal optimization, the energy consumption, latency, and resource overhead incurred during the switching process are not fully quantified and considered, leading to a lack of comprehensive management and compensation for frequency band switching costs. Access and tariff management for a single terminal cannot meet the requirements for system resource coordination and cost control in a multi-terminal coexistence environment, lacking a dynamic optimization scheduling strategy based on a global perspective. Traditional compensation strategies are static and difficult to respond to changes in terminal behavior and the operating environment. They lack adaptive adjustment capabilities, making it difficult to achieve fair and efficient tariff management.

[0006] In view of this, the present invention proposes a method for dynamic configuration of Internet of Things card packages based on intelligent policy regulation to solve the above problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for dynamically configuring an Internet of Things card package based on intelligent policy regulation, comprising: S1. Collect multi-source operating data of IoT terminals, identify terminal operating status by building a segmented threshold mapping structure, and generate status labels with unified coding; S2. Based on status tags, the system dynamically generates policy configuration results, including recommended network card package parameters, adjusted elasticity boundaries, and confidence weights, through a rule-matching process and trend evolution mechanism. S3. Based on the policy configuration results, identify terminal clusters with service-related business relevance, generate group network card package sharing configuration policies through the device commonality screening structure, sharing policy trigger mechanism and resource allocation control process, and synchronously manage the network card package execution status of terminals in the group; S4. Coordinate the access scheduling of terminals between different communication standards according to the communication frequency band access rules preset in the package sharing configuration strategy, and dynamically adjust the actual tariff amount based on the frequency band switching scheduling mechanism and package execution deviation compensation logic; S5. Collect user feedback data and, in combination with the weight self-adjustment mechanism, identify configuration policies that affect actual fee amount deviations. Through the policy version management process, replace, freeze, or roll back configuration policies that cause actual fee amount deviations, thereby achieving continuous evolution and adaptive adjustment of package configuration logic.

[0008] Preferably, the multi-source operation data includes device behavior data, communication data, service data and historical tariff data.

[0009] Preferably, the method for generating a status label with a unified code includes: The collected multi-source operation data is normalized by standard deviation and mapped into potential variables. Based on the preset resting potential threshold and action potential threshold, the potential state intervals corresponding to biological neurons are set, including the resting state interval, the threshold readiness state interval, and the action state interval. Each interval corresponds to a different terminal operation state. All acquired potential variables are accumulated to obtain the total potential value at the current moment, and potential mutation detection logic similar to the neuronal depolarization-repolarization process is introduced. The total potential change rate within a continuous time period is defined, and a total potential change rate threshold is preset. The terminal operation state is further annotated by combining the total potential value and potential state interval. The annotated terminal operation state includes depolarization state, repolarization state, hyperpolarization state, and threshold readiness state. The definition of depolarization state means that if the total potential value is greater than or equal to the preset action potential threshold, and the total potential change rate exceeds the preset total potential change rate threshold, it is determined that a potential mutation has occurred and the terminal has entered the action state interval; the definition of repolarization state means that if the total potential value is less than or equal to the resting potential threshold, and the total potential change rate is lower than the negative preset total potential change rate threshold, it is determined that the terminal has returned to the resting state interval; The hyperpolarized state means that if the total potential value drops and remains, and continues to be lower than the resting potential threshold, the terminal is judged to be in an abnormally low activity state interval; the threshold readiness state means that if the total potential value is greater than the resting potential threshold and less than the action potential threshold, the terminal is judged to be in the threshold readiness state interval; based on the currently identified terminal operating state, a uniformly coded state label is generated.

[0010] Preferably, the method for generating the policy configuration result includes: The preset rule base matches the corresponding policy rules based on the status tag. The matching process includes matching based on the current status tag, fuzzy matching, and default matching to obtain preliminary recommended network card package parameters, elastic boundary range, and confidence weight. Compare the change trends of the current state label with the terminal's historical state label, extract the label migration trajectory and evolution trend, and predict the state evolution direction and speed within the next n periods of time. Combined with the terminal's state label change trend, identify evolution characteristics and behavior patterns to assist in correcting the current terminal's prediction results. Based on the prediction results, the parameters of the preliminary recommended package are adjusted, and the elastic boundary range is expanded or contracted. Based on the changing trend of the terminal's historical status label, the number and average interval of the terminal's label switching within a given m time period are counted to obtain the trend confidence, and the confidence weight is increased or decreased based on the trend confidence. A dynamic weighted fusion mechanism is adopted to fuse the current state label matching results with the current terminal's prediction results, and output dynamic policy configuration results. The dynamic policy configuration results include recommended package parameters, adjusted elastic boundary range and revised confidence weight.

[0011] Preferably, the method for generating the group network card package sharing configuration strategy includes: For each terminal, a set of corresponding service feature vectors is extracted based on its policy configuration results, and the service similarity between any two terminals is calculated using cosine similarity. Based on the service feature vectors, the DBSCAN clustering algorithm is used to cluster different terminals according to their service similarity to form different preliminary terminal clusters. Screen the preliminary terminal clusters and retain the terminal clusters that meet the preset condition that the number of preliminary terminal clusters is greater than or equal to the preset minimum number threshold for the formation of preliminary terminal clusters; adopt a sharing policy trigger mechanism and define a policy trigger judgment function to determine whether to generate a group network card package sharing configuration policy; Perform resource allocation on the terminal cluster that generates the group network card package sharing configuration strategy, calculate the sharing allocation coefficient of each terminal cluster; calculate the sharing package quota allocated to each terminal group based on the sharing allocation coefficient; and finally generate the group network card package sharing configuration strategy.

[0012] Preferably, the network card package execution status includes package activation status, traffic usage status, access frequency band status, shared resource occupancy status, traffic deviation status, abnormal event status and policy version status.

[0013] Preferably, the method for coordinating access scheduling of terminals between different communication standards includes: The preset communication frequency band access rules include the definition of the accessible frequency band set, frequency band access priority, switching scheduling conditions, and the correspondence between the frequency band charges within the package. Based on the communication frequency band access rules preset by the package sharing configuration strategy, the accessible frequency band set is defined for each terminal, and the access preference weight of each terminal in different frequency bands, the network signal strength of the frequency band, and the communication load of the frequency band are collected; The access priority score of each terminal for each frequency band is calculated through the access scoring function, and the frequency band with the highest access priority score is selected to determine the frequency band as the terminal's priority access frequency band; during the frequency band access process, the terminal's cumulative communication behavior is continuously tracked, and the accumulated communication behavior includes the actual package charges , expected package charges , frequency band switching times and the additional energy consumption generated by a single switch ; Calculate terminal package execution deviation based on cumulative communication behavior ; Based on the execution deviation of the terminal package, the terminal compensation weight and package unit price coefficient are introduced to calculate the dynamic actual fee amount of the terminal ; For terminal clusters, the dynamic actual tariff amount of each terminal in the terminal group is counted and summed to determine the total actual tariff amount of the terminal; based on the access priority score of each terminal for each frequency band and the total actual tariff amount of the terminal, the access scheduling of the terminal between different communication standards is coordinated.

[0014] Preferably, the method for dynamically adjusting the actual fee amount includes: The communication frequency band access rules preset in the package sharing configuration strategy include dynamic monitoring and evaluation of the signal quality, communication load, data transmission requirements and frequency band capacity in the package sharing resource pool under different operating states of the terminal, and the establishment of different levels of frequency band access strategies by frequency band priority sorting, switching conditions and access constraints; Frequency band priority sorting is configured based on network signal strength, communication load, terminal energy consumption characteristics, and pre-defined priorities of available frequency bands in the package; switching conditions include dynamic determination based on signal quality degradation, load saturation, terminal status changes, and frequency band availability; access constraints include limiting the number of access handoffs, access rate, and latency thresholds for terminals between different frequency bands; The frequency band handover scheduling mechanism comprehensively evaluates the signal strength, frequency band load rate, and data volume demand of the terminal's current access frequency band in real time to obtain a comprehensive frequency band access score. Based on the comprehensive frequency band access score, a predictive handover algorithm predicts possible signal attenuation or load overload in advance, triggering a handover preparation state. According to the multi-level threshold mapping structure, frequency band switching is performed after the trigger condition is met, and switching synchronization control and buffering strategies are adopted to avoid data transmission interruption; a first scoring threshold for frequency band switching and a second scoring threshold for frequency band switching are preset. When the comprehensive score of frequency band access reaches the preset first scoring threshold for frequency band switching, the switching recommendation state is entered; when the comprehensive score of frequency band access reaches the preset second scoring threshold for frequency band switching, the switching is forced; The package execution deviation compensation logic is used to calculate the deviation between the actual fee quota and the preset fee quota by collecting the terminal's data usage, the number of frequency band switches, the communication duration, and the unit fee of the current frequency band after the frequency band switch. When it is detected that the deviation value exceeds the preset deviation value threshold, the compensation logic is activated. By adjusting the unused traffic between terminals in the package sharing group across terminals, the corresponding unused traffic quota is allocated from the preset adjustable resource pool, and the unused traffic quota is allocated to the terminal according to the preset compensation ratio; the overall tariff quota deviation within the sharing group is balanced, and dynamic adjustment and compensation are performed.

[0015] Preferably, the method for identifying the configuration strategy that affects the actual tariff amount deviation includes: Collect user feedback data, including actual user traffic usage, communication signal quality indicators, data transmission rate, communication stability evaluation, package price perception and satisfaction, and terminal abnormality logs; Establish a multi-dimensional correlation model between user feedback data and tariff deviations. Dynamically calculate and adjust the weight coefficients of each feedback item in the user feedback data through weighted fusion. The weight coefficients reflect the relative impact of each feedback item on the tariff deviation. The weighted user feedback data is integrated with the terminal's operating status data and package execution data to form a comprehensive data set. Based on the comprehensive data set, correlation coefficient analysis is used to preliminarily screen configuration policy parameters and user behavior indicators that are correlated with actual fee deviations. Granger causality analysis is then used to identify configuration strategies that lead to actual fee deviations.

[0016] Preferably, the method for replacing, freezing or rolling back the configuration policy that causes the actual tariff amount deviation includes: Identify and remove configuration policies that affect actual fee quota deviations, archive and manage the removed configuration policies and their version information, and automatically mark the corresponding configuration policy version as abnormal if it is detected that the actual fee quota deviation caused by any configuration policy version exceeds the preset fee quota deviation threshold. Replace the abnormal version with a new version configuration policy generated based on preset optimization rules, and verify the validity of the new version configuration policy in the terminal cluster. Before the replacement, freeze the configuration policy corresponding to the abnormal version, suspend its application on the newly added terminals, and maintain the operation status of the existing terminals. When the configuration policy corresponding to the new version fails verification, the configuration policy can be rolled back to the previous version. Combined with continuously collected user feedback data, terminal operation status data, and package execution data, the configuration policy effect is dynamically evaluated and the iterative update of the configuration policy is automatically triggered.

[0017] The technical effects and advantages of the method for dynamically configuring Internet of Things card packages based on intelligent policy regulation of the present invention are as follows: This invention extracts terminal service feature vectors in real time and combines terminal policy configuration results with service feature similarity to automatically identify service-related terminal clusters based on a clustering algorithm. This avoids the limitations of manual intervention and static configuration, and enables intelligent and dynamic package configuration that efficiently adapts to complex service scenarios. A shared policy triggering decision function is designed that comprehensively considers group size and communication capacity requirements. By dynamically determining whether a group meets the shared policy generation conditions, the accuracy and rationality of shared policy triggering are ensured, avoiding policy misjudgments.

[0018] By calculating the shared allocation coefficient, the shared package quota is dynamically allocated within the terminal cluster based on package traffic demand, making resource allocation more fair and reasonable while also improving network resource utilization efficiency. By introducing a sliding window mechanism, the group status is dynamically monitored to ensure that the group continues to meet the triggering conditions while the sharing policy is in effect, avoiding the failure of the sharing policy due to terminal changes, and enhancing the continuity and applicability of group management. Through the optimization of sharing strategies and resource allocation, not only is the network service quality improved, but operating costs are also significantly reduced, improving economic benefits. By introducing a comprehensive access scoring function, combined with factors such as terminal preference weight, real-time signal strength, and current load, terminal frequency band selection is made intelligent and dynamic, improving spectrum resource utilization efficiency and overall network performance.

[0019] Utilizing terminal compensation weights and package unit price coefficients, we achieve real-time correction of package execution deviations, balancing user experience and operator revenue, and improving the rationality and fairness of tariffs. By averaging historical deviations to balance short-term fluctuations, we achieve smooth dynamic adjustment of compensation weights, enhancing the system's adaptability and robustness, and improving the stability of long-term tariff management. By aggregating and comprehensively regulating the dynamic actual tariff amounts of all terminals, we balance priority access demands with total tariff constraints, promoting balanced allocation of system resources and overall improved access performance. Based on real-time data and a dynamic adjustment mechanism, we achieve automatic optimization of frequency band access and tariff management. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flowchart of a method for dynamically configuring IoT card packages based on intelligent policy control; Figure 2 This is a schematic diagram of the structure of a dynamic configuration system for IoT card packages based on intelligent policy regulation; Figure 3 The present invention provides a flow chart of a method for identifying configuration strategies that affect actual tariff amount deviations. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] Example 1 See also Figure 1 and Figure 3As shown, embodiment 1 further illustrates a method for dynamically configuring an IoT card package based on intelligent policy regulation proposed by the present invention, including: The rapid development of IoT technology has driven the demand for access to a massive number of terminal devices. As a crucial communication medium for these devices, the configuration and management of IoT card packages (hereafter referred to as "IoT Cards") directly impact network resource utilization and operating costs. Existing methods for configuring IoT Card packages primarily rely on static grouping strategies or manual rule-based configuration, lacking dynamic awareness and adaptive adjustment to the actual service needs of terminal devices.

[0023] Specifically, the existing technology has the following deficiencies: Existing sharing policy configurations typically rely on static grouping or manually defined rules, failing to dynamically identify service relevance between terminals based on their service characteristics and policy configuration. This results in inefficient sharing policy configuration and significant waste of network resources. Sharing policies often use simple traffic sums or single-point capacity as trigger conditions, failing to fully consider both group size and overall capacity requirements. This can easily lead to delayed or false triggering of sharing policies, and even inappropriate configurations due to extreme terminal traffic.

[0024] Traditional IoT card package allocation methods often rely on static proportional allocations, failing to dynamically adjust based on actual terminal needs. This can lead to resource shortages on some terminals and idle resources in others, reducing overall resource utilization. Furthermore, existing terminal access frequency band selection is often based on fixed configurations or simple priority strategies, lacking awareness and analysis of the real-time network environment. This results in inefficient access resource utilization, irrational frequency band selection, and a limited user experience.

[0025] Existing technologies often focus solely on static calculations of package charges, ignoring the discrepancy between actual terminal communication behavior and expected package charges, as well as the additional energy costs associated with frequency band switching. This can lead to inaccurate rate allocation, impacting operator revenue and user fairness. While frequency band switching helps achieve load balancing and signal optimization, the energy consumption, latency, and resource overhead associated with the switching process are often not fully quantified and considered, resulting in a lack of comprehensive management and compensation mechanisms for frequency band switching costs.

[0026] Most existing technologies focus on single-terminal access management and rate scheduling, failing to coordinate system resources and control costs in a multi-terminal environment. Compensation strategies typically employ static configurations, making them difficult to adapt to changes in terminal behavior and the dynamic operating environment. These strategies lack adaptive adjustment capabilities, hindering fair and efficient rate management and optimization.

[0027] In order to effectively solve the above problems, the present invention proposes a method for dynamically configuring IoT card packages based on intelligent policy regulation, comprising: S1. Collect multi-source operating data of IoT terminals, identify terminal operating status by building a segmented threshold mapping structure, and generate status labels with unified coding; S2. Based on status tags, the system dynamically generates policy configuration results, including recommended network card package parameters, adjusted elasticity boundaries, and confidence weights, through a rule-matching process and trend evolution mechanism. S3. Based on the policy configuration results, identify terminal clusters with service-related business relevance, generate group network card package sharing configuration policies through the device commonality screening structure, sharing policy trigger mechanism and resource allocation control process, and synchronously manage the network card package execution status of terminals in the group; S4. Coordinate the access scheduling of terminals between different communication standards according to the communication frequency band access rules preset in the package sharing configuration strategy, and dynamically adjust the actual tariff amount based on the frequency band switching scheduling mechanism and package execution deviation compensation logic; S5. Collect user feedback data and, in combination with the weight self-adjustment mechanism, identify configuration policies that affect actual fee amount deviations. Through the policy version management process, replace, freeze, or roll back configuration policies that cause actual fee amount deviations, thereby achieving continuous evolution and adaptive adjustment of package configuration logic.

[0028] Multi-source operation data includes equipment behavior data, communication data, business data and historical tariff data.

[0029] Device behavior data includes terminal CPU usage, memory occupancy, remaining storage space, operating time, idle / active state switching frequency and energy consumption; communication data includes data transmission volume, uplink and downlink rates, signal strength, connection duration and communication frequency; business data includes the type of business run by the terminal and the type of communication interface called; historical tariff data includes tariff change records within the package usage period, overage occurrence points and arrears suspension information.

[0030] Methods for generating status labels with unified encoding include: The collected multi-source operation data is normalized by standard deviation and mapped into potential variables to simulate the membrane potential characteristics of biological neurons. Based on the preset resting potential threshold and action potential threshold, the potential state intervals corresponding to biological neurons are set, including the resting state interval, the threshold readiness state interval, and the action state interval, which correspond to the terminal's idle state, warning state, and abnormally active state respectively. Each interval corresponds to a different terminal operation state. All acquired potential variables are accumulated to obtain the total potential value at the current moment to support global state judgment. To enhance the ability to respond to sudden state changes, a potential mutation detection logic similar to the depolarization-repolarization process of neurons is introduced. Based on the existing potential variable construction and potential state interval division, this potential mutation detection logic mechanism further introduces the calculation and judgment logic of the potential mutation rate to characterize the rapid transition phenomenon of the operating state. The total potential change rate within a continuous time period is defined as: ;in, Indicates the current time point The total potential value; Indicates the previous time point The total potential value; Indicates a time interval; Indicates the index of the time point; presets the total potential change rate threshold, combines the total potential value and the potential state interval, and further marks the terminal operation state; the marked terminal operation state includes depolarization state, repolarization state, hyperpolarization state and threshold preparation state; The definition of depolarization state means that if the total potential value is greater than or equal to the preset action potential threshold, and the total potential change rate exceeds the preset total potential change rate threshold, it is determined that a potential mutation has occurred and the terminal enters the action state interval; for example, network surge, abnormal high-frequency response of the sensor; The repolarization state means that if the total potential value is less than or equal to the resting potential threshold, and the total potential change rate is lower than the negative preset total potential change rate threshold, then it is determined that the terminal has returned to the resting state range; for example, normal standby or low-load operation; The hyperpolarized state means that if the total potential value decreases and remains below the resting potential threshold, the terminal is judged to be in an abnormally low activity state range; for example, communication link is disconnected or sleep failure; The threshold readiness state indicates that if the total potential value is greater than the resting potential threshold and less than the action potential threshold, the terminal is judged to be in the threshold readiness state interval; for example, steady-state operation and low-risk tasks; based on the currently identified terminal operation state, a uniformly coded state label is generated.

[0031] To support subsequent policy reasoning and configuration management: the status label includes a status category field, a status subdivision field, timestamp information, and a terminal identification field, for example: [terminal ID]-[status category]-[status subcategory]-[timestamp]. Whenever the identified terminal operating status changes, a new status label is immediately generated to replace the old status label. At the same time, the historical status labels are retained to construct the terminal operating status evolution trajectory for subsequent decision analysis.

[0032] The methods for generating policy configuration results include: The preset rule base matches the corresponding policy rules based on the status tag. The matching process includes matching based on the current status tag, fuzzy matching, and default matching to obtain preliminary recommended network card package parameters, elastic boundary range, and confidence weight. Compare the change trends of the current state label with the terminal's historical state label, extract the label migration trajectory and evolution trend, and predict the state evolution direction and speed within the next n periods of time. Combined with the terminal's state label change trend, identify evolution characteristics and behavior patterns to assist in correcting the current terminal's prediction results. Based on the prediction results, the parameters of the preliminary recommended package are adjusted, and the elastic boundary range is expanded or contracted. Based on the changing trend of the terminal's historical status label, the number and average interval of the terminal's label switching within a given m time period are counted to obtain the trend confidence, and the confidence weight is increased or decreased based on the trend confidence.

[0033] A dynamic weighted fusion mechanism is adopted to fuse the current state label matching results with the current terminal's prediction results, and output dynamic policy configuration results. The dynamic policy configuration results include recommended package parameters, adjusted elastic boundary range and revised confidence weight.

[0034] The method for generating a group network card package sharing configuration policy includes: For each terminal, a set of corresponding service feature vectors is extracted based on its policy configuration results, and the service similarity between any two terminals is calculated using cosine similarity. Based on the service feature vectors, the DBSCAN clustering algorithm is used to cluster different terminals according to their service similarity to form different preliminary terminal clusters. Screen the preliminary terminal clusters and retain the terminal clusters that meet the preset condition that the number of preliminary terminal clusters is greater than or equal to the preset minimum number threshold for the formation of preliminary terminal clusters; adopt a sharing policy trigger mechanism and define a policy trigger judgment function to determine whether to generate a group network card package sharing configuration policy; The policy triggering judgment function is: ;in, Indicates the terminal clusters; Indicates the minimum number threshold for forming a preset preliminary terminal cluster; Indicates the The package traffic requirements of a terminal cluster; Indicates the package traffic demand threshold for the sharing policy trigger mechanism; and Indicates the number of the terminal cluster; if , then the policy triggers the judgment function to determine and generate the group network card package sharing configuration strategy; if , then the policy triggering judgment function determines that the group network card package sharing configuration policy is not generated; Share resources among the terminal clusters that generate the group network card package sharing configuration strategy, and calculate the shared sharing coefficient for each terminal cluster; Shared sharing coefficient ;in, Indicates the The package traffic requirements of a terminal cluster; Indicates the number of the terminal cluster; the shared package quota allocated to each terminal group is calculated based on the shared allocation coefficient; ;in, Indicates allocation to terminal groups The shared package quota; Indicates allocation to terminal groups The total shared package quota is calculated; and finally a group network card package sharing configuration strategy is generated.

[0035] For example, large-scale logistics companies operating nationwide (such as SF Express, JD Logistics, and Deppon) have fleets comprised of tens of thousands of transport vehicles, each equipped with an IoT card for real-time vehicle location upload, cargo status monitoring, onboard device communication, and fleet dispatch instruction transmission. Existing technologies often configure fixed packages for different vehicles in batches. This results in: the packages fail to dynamically adjust to the actual business needs of the vehicles, resulting in some vehicles having excessive package quotas (causing waste), and others incurring excessive charges due to excessive demand. Furthermore, the lack of intelligent identification and automatic adjustment mechanisms complicates operational management and delays timely response.

[0036] The company has deployed a dynamic configuration method for IoT card packages based on intelligent policy control. The specific process is as follows: For each logistics vehicle terminal, the system analyzes recent communication logs to extract a service feature vector: [average daily traffic volume, communication frequency, service type label, activity area]. It then uses cosine similarity to compare service similarities between terminals and, using the DBSCAN clustering algorithm, automatically identifies clusters of vehicles with similar service characteristics. For example, approximately 5,000 logistics vehicles were initially formed into 120 clusters, ranging in size from 10 to 100 members.

[0037] Set the minimum cluster size threshold to 15 and the capacity requirement threshold to 800MB; the system uses the policy triggering judgment function to judge each terminal cluster, and the cluster that meets the conditions starts the package sharing configuration policy. For the sharing group, there are 3 terminal clusters participating in the sharing configuration, numbered as follows: 、 and The corresponding traffic requirements of each terminal cluster are: is 150000MB, is 250000MB, 10,000MB; the total package quota is dynamically calculated based on the total demand of the fleet and allocated according to the cluster capacity; the total shared package quota is 1,000,000MB, and the total traffic requirement is ; Calculate the shared allocation coefficient for each terminal cluster separately: ; ; ; Then the shared package quotas for each terminal cluster are: ; ; ; This solves the following problems with existing technologies: Existing technologies typically configure packages based on static grouping or manual rules, making it difficult to dynamically identify business relevance between terminals based on the actual business needs and policy configurations of terminal devices. This results in inefficient sharing policy configuration and severe resource waste. Existing sharing policies typically use simple traffic sums or single-point capacity as trigger conditions, lacking a dual assessment of group size and overall capacity requirements. This can lead to delayed or false triggering of sharing policies, or irrational configurations due to extreme terminal traffic. Traditional IoT package allocations typically use static ratios and are not dynamically adjusted based on actual terminal demand. This results in resource constraints on some terminals and idle resources on others, reducing resource utilization.

[0038] Advantages over existing technologies: By extracting terminal service feature vectors in real time and combining terminal policy configuration results with service feature similarity, the system automatically identifies service-related terminal clusters based on a clustering algorithm, avoiding the limitations of manual intervention and static configuration, and achieving intelligent, dynamic package configuration and the ability to efficiently adapt to complex service scenarios. A shared policy triggering judgment function is designed, taking into account the group size and communication capacity requirements. By dynamically determining whether the group meets the shared policy generation conditions, the accuracy and rationality of shared policy triggering are ensured, and policy misjudgments are avoided. By calculating the shared allocation coefficient, shared package quotas are dynamically allocated within the terminal cluster based on package traffic demand, making resource allocation more equitable and reasonable while also improving network resource utilization efficiency. A sliding window mechanism is introduced to dynamically monitor group status, ensuring that the group continues to meet trigger conditions while the sharing policy is in effect. This prevents sharing policy failures due to terminal changes and enhances the continuity and applicability of group management. By optimizing sharing policies and resource allocation, network service quality is improved while operating costs are significantly reduced, resulting in greater economic benefits.

[0039] The network card package execution status includes package activation status, traffic usage status, access frequency band status, shared resource occupancy status, traffic deviation status, abnormal event status and policy version status.

[0040] The package activation status describes whether the terminal is currently using a package sharing feature. Values ​​include: Activated, Inactive, and Suspended. Purpose: Determines whether a shared package has been applied to the terminal and its current activation status. The traffic usage status describes the ratio of the terminal's used traffic to the total allocated quota in the shared package; it includes: current usage, remaining quota, and usage rate. Purpose: Dynamically monitors the terminal's package consumption progress and provides a basis for adjusting sharing policies.

[0041] The access band status describes the communication band (such as 4G, 5G, NB-IoT) and standard currently accessed by the terminal. Purpose: Determines the effectiveness of terminal access scheduling based on the frequency band access rules preset in the package sharing policy. The shared resource occupancy status describes the terminal's resource usage within the shared package, including channel, bandwidth, and QoS level. Purpose: Ensures fair resource distribution within the group and prevents excessive occupancy by individual terminals.

[0042] The traffic deviation status describes the deviation between the actual usage of a terminal's current package and the expected value. Purpose: Deviation analysis supports dynamic adjustment of package allocations and confidence weights. The abnormal event status describes abnormal events that occur during package execution (such as access failure, package quota overage, signal loss, etc.). Purpose: Detects and handles anomalies to improve the stability and security of shared packages. The policy version status describes the shared policy version number currently being executed by the terminal. Purpose: Supports policy version rollback, freezing, or upgrading to ensure consistency between policy and terminal status.

[0043] The method for coordinating access scheduling of terminals between different communication standards includes: The preset communication frequency band access rules include the definition of the accessible frequency band set, frequency band access priority, switching scheduling conditions, and the correspondence between the frequency band charges within the package. Based on the communication frequency band access rules preset by the package sharing configuration strategy, the accessible frequency band set is defined for each terminal, and the access preference weight of each terminal in different frequency bands, the network signal strength of the frequency band, and the communication load of the frequency band are collected; The access priority score of each terminal for each frequency band is calculated through the access scoring function, and the frequency band with the highest access priority score is selected to determine the frequency band as the terminal's priority access frequency band; during the frequency band access process, the terminal's cumulative communication behavior is continuously tracked, and the accumulated communication behavior includes the actual package charges , expected package charges , frequency band switching times and the additional energy consumption generated by a single switch ; Calculate terminal package execution deviation based on cumulative communication behavior ; ;in, Indicates that the terminal is Additional energy consumption during the secondary switching; Indicates the sequence number of the terminal when performing frequency band switching; Based on the execution deviation of the terminal package, the terminal compensation weight and package unit price coefficient are introduced to calculate the dynamic actual fee amount of the terminal ; ;in, Indicates the unit price coefficient of the package; Indicates the terminal compensation weight; based on expert experience, the terminal compensation weight and package unit price coefficient range from 0 to 1; For terminal clusters, the dynamic actual tariff amounts of each terminal in the terminal group are counted and summed to determine the total actual tariff amount of the terminal; based on each terminal's access priority score for each frequency band and the total actual tariff amount of the terminal, the access scheduling of the terminal between different communication standards is coordinated.

[0044] This technology addresses the following issues with existing technologies: Traditional package configuration methods rely on fixed configurations or simple priorities for terminal access frequency bands, making it difficult to dynamically select the optimal frequency band based on the real-time network environment. This results in poor resource utilization and a limited user experience. Existing technologies typically focus solely on static rate calculations, ignoring the discrepancy between actual terminal communication behavior and expected package rates, as well as the additional energy costs associated with frequency band switching. This leads to inaccurate rate allocation, impacting operator revenue and user fairness.

[0045] While frequency band switching facilitates load balancing and signal optimization, the energy consumption, latency, and resource overhead incurred during the switching process are not fully quantified and considered, leading to a lack of comprehensive management and compensation for frequency band switching costs. Access and tariff management for a single terminal cannot meet the requirements for system resource coordination and cost control in a multi-terminal coexistence environment, lacking a dynamic optimization scheduling strategy based on a global perspective. Traditional compensation strategies are static and difficult to respond to changes in terminal behavior and the operating environment. They lack adaptive adjustment capabilities, making it difficult to achieve fair and efficient tariff management.

[0046] Compared with existing technologies, this approach offers several advantages: By introducing a comprehensive access scoring function that combines factors such as terminal preference weight, real-time signal strength, and current load, it enables intelligent and dynamic terminal frequency band selection, improving spectrum resource utilization efficiency and overall network performance. It also explicitly incorporates the additional energy consumption and latency generated during frequency band switching into the cumulative communication behavior, improving rate calculations and achieving more accurate cost accounting, thereby promoting the rationality and balance of frequency band switching behavior.

[0047] Utilizing terminal compensation weights and package unit price coefficients, we achieve real-time correction of package execution deviations, balancing user experience and operator revenue, and improving the rationality and fairness of tariffs. By averaging historical deviations to balance short-term fluctuations, we achieve smooth dynamic adjustment of compensation weights, enhancing the system's adaptability and robustness, and improving the stability of long-term tariff management. By aggregating and comprehensively regulating the dynamic actual tariff amounts of all terminals, we balance priority access demands with total tariff constraints, promoting balanced allocation of system resources and overall improved access performance. Based on real-time data and a dynamic adjustment mechanism, we achieve automatic optimization of frequency band access and tariff management.

[0048] Methods for dynamically adjusting the actual fee amount include: The communication frequency band access rules preset in the package sharing configuration strategy include dynamic monitoring and evaluation of the signal quality, communication load, data transmission requirements and frequency band capacity in the package sharing resource pool under different operating states of the terminal, and the establishment of different levels of frequency band access strategies by frequency band priority sorting, switching conditions and access constraints; Frequency band priority sorting is configured based on network signal strength, communication load, terminal energy consumption characteristics, and pre-defined priorities of available frequency bands in the package; switching conditions include dynamic determination based on signal quality degradation, load saturation, terminal status changes, and frequency band availability; access constraints include limiting the number of access handoffs, access rate, and latency thresholds for terminals between different frequency bands; The frequency band handover scheduling mechanism comprehensively evaluates the signal strength, frequency band load rate, and data volume demand of the terminal's current access frequency band in real time to obtain a comprehensive frequency band access score. Based on the comprehensive frequency band access score, a predictive handover algorithm predicts possible signal attenuation or load overload in advance, triggering a handover preparation state. According to the multi-level threshold mapping structure, frequency band switching is performed after the trigger condition is met, and switching synchronization control and buffering strategies are adopted to avoid data transmission interruption; a first scoring threshold for frequency band switching and a second scoring threshold for frequency band switching are preset. When the comprehensive score of frequency band access reaches the preset first scoring threshold for frequency band switching, the switching recommendation state is entered; when the comprehensive score of frequency band access reaches the preset second scoring threshold for frequency band switching, the switching is forced; The package execution deviation compensation logic is used to calculate the deviation between the actual fee quota and the preset fee quota by collecting the terminal's data usage, the number of frequency band switches, the communication duration, and the unit fee of the current frequency band after the frequency band switch. When it is detected that the deviation value exceeds the preset deviation value threshold, the compensation logic is activated. By adjusting the unused traffic between terminals in the package sharing group across terminals, the corresponding unused traffic quota is allocated from the preset adjustable resource pool, and the unused traffic quota is allocated to the terminal according to the preset compensation ratio; the overall tariff quota deviation within the sharing group is balanced, and dynamic adjustment and compensation are performed.

[0049] Methods for identifying configuration policies that affect actual tariff deviations include: Collect user feedback data, including actual user traffic usage, communication signal quality indicators, data transmission rate, communication stability evaluation, package price perception and satisfaction, and terminal abnormality logs; Establish a multi-dimensional correlation model between user feedback data and rate deviations. Dynamically calculate and adjust the weight coefficients of each feedback item in the user feedback data through weighted fusion. The weight coefficients reflect the relative impact of each feedback item on rate deviations, ensuring that key influencing factors are given higher priority during the identification process. The weighted user feedback data is integrated with the terminal's operating status data and package execution data to form a comprehensive data set. Based on the comprehensive data set, correlation coefficient analysis is used to preliminarily screen configuration policy parameters and user behavior indicators that are correlated with actual fee deviations. Granger causality analysis is then used to identify configuration strategies that lead to actual fee deviations.

[0050] Methods for replacing, freezing, or rolling back configuration policies that cause actual fee deviations include: Identify and remove configuration policies that affect actual fee quota deviations, archive and manage the removed configuration policies and their version information, and automatically mark the corresponding configuration policy version as abnormal if it is detected that the actual fee quota deviation caused by any configuration policy version exceeds the preset fee quota deviation threshold. Replace the abnormal version with a new version configuration policy generated based on preset optimization rules, and verify the validity of the new version configuration policy in the terminal cluster. Before the replacement, freeze the configuration policy corresponding to the abnormal version, suspend its application on the newly added terminals, and maintain the operation status of the existing terminals. When the configuration policy corresponding to the new version fails verification, the configuration policy can be rolled back to the previous version. Combined with continuously collected user feedback data, terminal operation status data, and package execution data, the configuration policy effect is dynamically evaluated and the iterative update of the configuration policy is automatically triggered.

[0051] The resting potential threshold is set by the staff. By collecting different resting potentials, the average value of multiple resting potentials is taken as the resting potential threshold; similarly, the action potential threshold, the total potential change rate threshold, the preset action potential threshold, the resting potential threshold, the preset total potential change rate threshold, the minimum number threshold for the formation of the initial terminal cluster, the delay threshold, the first scoring threshold for frequency band switching, the second scoring threshold for frequency band switching, the preset deviation value threshold and the preset tariff amount deviation threshold are set.

[0052] This embodiment extracts terminal service feature vectors in real time and combines terminal policy configuration results with service feature similarity to automatically identify service-related terminal clusters based on a clustering algorithm. This avoids the limitations of manual intervention and static configuration, and enables intelligent, dynamic package configuration that efficiently adapts to complex service scenarios. A shared policy triggering determination function is designed, taking into account group size and communication capacity requirements. By dynamically determining whether a group meets the shared policy generation conditions, the accuracy and rationality of shared policy triggering are ensured, avoiding policy misjudgments. By calculating the shared allocation coefficient, the shared package quota is dynamically allocated within the terminal cluster based on package traffic demand, making resource allocation more fair and reasonable while also improving network resource utilization efficiency. By introducing a sliding window mechanism, the group status is dynamically monitored to ensure that the group continues to meet the triggering conditions while the sharing policy is in effect, avoiding the failure of the sharing policy due to terminal changes, and enhancing the continuity and applicability of group management. Through the optimization of sharing strategies and resource allocation, not only is the network service quality improved, but operating costs are also significantly reduced, improving economic benefits. By introducing a comprehensive access scoring function, combined with factors such as terminal preference weight, real-time signal strength, and current load, terminal frequency band selection is made intelligent and dynamic, improving spectrum resource utilization efficiency and overall network performance.

[0053] Utilizing terminal compensation weights and package unit price coefficients, we achieve real-time correction of package execution deviations, balancing user experience and operator revenue, and improving the rationality and fairness of tariffs. By averaging historical deviations to balance short-term fluctuations, we achieve smooth dynamic adjustment of compensation weights, enhancing the system's adaptability and robustness, and improving the stability of long-term tariff management. By aggregating and comprehensively regulating the dynamic actual tariff amounts of all terminals, we balance priority access demands with total tariff constraints, promoting balanced allocation of system resources and overall improved access performance. Based on real-time data and a dynamic adjustment mechanism, we achieve automatic optimization of frequency band access and tariff management.

[0054] Example 2 See also Figure 2 As shown, for the parts not described in detail in this embodiment, please refer to the description of Example 1. A dynamic configuration system for Internet of Things card packages based on intelligent policy regulation is provided, including: The multi-dimensional perception state mapping unit collects multi-source operating data of IoT terminals, identifies the terminal operating status by building a segmented threshold mapping structure, and generates a state label with a unified code; The package policy generation and evolution unit, based on status tags, combines rule matching processes with trend evolution mechanisms to dynamically generate policy configuration results that include recommended network card package parameters, adjusted elasticity boundaries, and confidence weights. The group symbiotic resource allocation unit identifies terminal clusters with service-related business relevance based on policy configuration results. Through the device commonality screening structure, sharing policy trigger mechanism and resource allocation control process, it generates group network card package sharing configuration policies and synchronously manages the network card package execution status of terminals in the group. The collaborative access compensation execution unit coordinates the access scheduling of terminals between different communication standards according to the communication frequency band access rules preset in the package sharing configuration strategy, and dynamically adjusts the actual tariff amount based on the frequency band switching scheduling mechanism and package execution deviation compensation logic; The feedback optimization unit collects user feedback data and, combined with the weight self-adjustment mechanism, automatically identifies configuration strategies that affect deviations in actual tariff amounts. Through the policy version management process, configuration strategies that cause deviations in actual tariff amounts are replaced, frozen, or rolled back, enabling continuous evolution and adaptive adjustment of package configuration logic.

[0055] Since the electronic device introduced in this embodiment is an electronic device used to implement a method for dynamic configuration of an Internet of Things card package based on intelligent policy regulation in the embodiment of this application, based on the method for dynamic configuration of an Internet of Things card package based on intelligent policy regulation introduced in the embodiment of this application, those skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as those skilled in the art implement the electronic device used by a method for dynamic configuration of an Internet of Things card package based on intelligent policy regulation in the embodiment of this application, they all fall within the scope of protection of this application.

[0056] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0057] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for users of ordinary skill in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamically configuring Internet of Things card packages based on intelligent policy regulation, characterized in that: include: S1. Collect multi-source operating data of IoT terminals, identify terminal operating status by building a segmented threshold mapping structure, and generate status labels with unified coding; S2. Based on status tags, the system dynamically generates policy configuration results, including recommended network card package parameters, adjusted elasticity boundaries, and confidence weights, through a rule-matching process and trend evolution mechanism. S3. Based on the policy configuration results, identify terminal clusters with service-related business relevance, generate group network card package sharing configuration policies through the device commonality screening structure, sharing policy trigger mechanism and resource allocation control process, and synchronously manage the network card package execution status of terminals in the group; S4. Coordinate the access scheduling of terminals between different communication standards according to the communication frequency band access rules preset in the package sharing configuration strategy, and dynamically adjust the actual tariff amount based on the frequency band switching scheduling mechanism and package execution deviation compensation logic; S5. Collect user feedback data and, in combination with the weight self-adjustment mechanism, identify configuration policies that affect actual fee amount deviations. Through the policy version management process, replace, freeze, or roll back configuration policies that cause actual fee amount deviations, thereby achieving continuous evolution and adaptive adjustment of package configuration logic.

2. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 1 is characterized in that: The multi-source operation data includes device behavior data, communication data, service data and historical tariff data.

3. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 2 is characterized in that: The method for generating a status label with a unified code includes: The collected multi-source operation data is normalized by standard deviation and mapped into potential variables. Based on the preset resting potential threshold and action potential threshold, the potential state intervals corresponding to biological neurons are set, including the resting state interval, the threshold readiness state interval, and the action state interval. Each interval corresponds to a different terminal operation state. All acquired potential variables are accumulated to obtain the total potential value at the current moment; potential mutation detection logic similar to the neuronal depolarization-repolarization process is introduced; the total potential change rate within a continuous time period is defined, and a total potential change rate threshold is preset. The terminal operation state is further annotated by combining the total potential value and potential state interval; the annotated terminal operation state includes depolarization state, repolarization state, hyperpolarization state, and threshold readiness state; The definition of depolarization state means that if the total potential value is greater than or equal to the preset action potential threshold, and the total potential change rate exceeds the preset total potential change rate threshold, it is determined that a potential mutation occurs and the terminal enters the action state interval. The repolarization state means that if the total potential value is less than or equal to the resting potential threshold, and the total potential change rate is lower than the negative preset total potential change rate threshold, the terminal is judged to have returned to the resting state range; the hyperpolarization state means that if the total potential value decreases and remains, and continues to be lower than the resting potential threshold, the terminal is judged to be in the abnormally low activity state range; The threshold readiness state indicates that if the total potential value is greater than the resting potential threshold and less than the action potential threshold, the terminal is determined to be in the threshold readiness state interval; a uniformly coded state label is generated based on the currently identified terminal operation state.

4. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 3 is characterized in that: The method for generating the policy configuration result includes: The preset rule base matches the corresponding policy rules based on the status tag. The matching process includes matching based on the current status tag, fuzzy matching, and default matching to obtain preliminary recommended network card package parameters, elastic boundary range, and confidence weight. Compare the change trends of the current state label with the terminal's historical state label, extract the label migration trajectory and evolution trend, and predict the state evolution direction and speed within the next n periods of time. Combined with the terminal's state label change trend, identify evolution characteristics and behavior patterns to assist in correcting the current terminal's prediction results. Based on the prediction results, the parameters of the preliminary recommended package are adjusted, and the elastic boundary range is expanded or contracted. Based on the changing trend of the terminal's historical status label, the number and average interval of the terminal's label switching within a given m time period are counted to obtain the trend confidence, and the confidence weight is increased or decreased based on the trend confidence. A dynamic weighted fusion mechanism is adopted to fuse the current state label matching results with the current terminal's prediction results, and output dynamic policy configuration results. The dynamic policy configuration results include recommended package parameters, adjusted elastic boundary range and revised confidence weight.

5. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 4 is characterized in that: The method for generating the group network card package sharing configuration strategy includes: For each terminal, a set of corresponding service feature vectors is extracted based on its policy configuration results, and the service similarity between any two terminals is calculated using cosine similarity. Based on the service feature vectors, the DBSCAN clustering algorithm is used to cluster different terminals according to their service similarity to form different preliminary terminal clusters. Screen the preliminary terminal clusters and retain the terminal clusters that meet the preset condition that the number of preliminary terminal clusters is greater than or equal to the preset minimum number threshold for the formation of preliminary terminal clusters; adopt a sharing policy trigger mechanism and define a policy trigger judgment function to determine whether to generate a group network card package sharing configuration policy; Perform resource allocation on the terminal cluster that generates the group network card package sharing configuration strategy, and calculate the sharing allocation coefficient of each terminal cluster; calculate the sharing package quota allocated to each terminal group based on the sharing allocation coefficient, and finally generate the group network card package sharing configuration strategy.

6. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 5 is characterized in that: The network card package execution status includes package activation status, traffic usage status, access frequency band status, shared resource occupancy status, traffic deviation status, abnormal event status and policy version status.

7. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 6 is characterized in that: The method for coordinating access scheduling of terminals between different communication standards includes: The preset communication frequency band access rules include the definition of the accessible frequency band set, frequency band access priority, switching scheduling conditions, and the correspondence between the frequency band charges within the package. Based on the communication frequency band access rules preset by the package sharing configuration strategy, the accessible frequency band set is defined for each terminal, and the access preference weight of each terminal in different frequency bands, the network signal strength of the frequency band, and the communication load of the frequency band are collected; The access priority score of each terminal for each frequency band is calculated through the access scoring function, and the frequency band with the highest access priority score is selected to determine the frequency band as the terminal's priority access frequency band; during the frequency band access process, the terminal's cumulative communication behavior is continuously tracked, and the accumulated communication behavior includes the actual package charges , expected package charges , frequency band switching times and the additional energy consumption generated by a single switch ; Calculate terminal package execution deviation based on cumulative communication behavior ; Based on the execution deviation of the terminal package, the terminal compensation weight and package unit price coefficient are introduced to calculate the dynamic actual fee amount of the terminal ; For terminal clusters, the dynamic actual tariff amount of each terminal in the terminal group is counted and summed to determine the total actual tariff amount of the terminal; based on the access priority score of each terminal for each frequency band and the total actual tariff amount of the terminal, the access scheduling of the terminal between different communication standards is coordinated.

8. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 7 is characterized in that: The method for dynamically adjusting the actual fee amount includes: The communication frequency band access rules preset in the package sharing configuration strategy include dynamic monitoring and evaluation of the signal quality, communication load, data transmission requirements and frequency band capacity in the package sharing resource pool under different operating states of the terminal, and the establishment of different levels of frequency band access strategies by frequency band priority sorting, switching conditions and access constraints; Frequency band priority sorting is configured based on network signal strength, communication load, terminal energy consumption characteristics, and pre-defined priorities of available frequency bands in the package; switching conditions include dynamic determination based on signal quality degradation, load saturation, terminal status changes, and frequency band availability; access constraints include limiting the number of access handoffs, access rate, and latency thresholds for terminals between different frequency bands; The frequency band handover scheduling mechanism comprehensively evaluates the signal strength, frequency band load rate, and data volume demand of the terminal's current access frequency band in real time to obtain a comprehensive frequency band access score. Based on the comprehensive frequency band access score, a predictive handover algorithm predicts possible signal attenuation or load overload in advance, triggering a handover preparation state. According to the multi-level threshold mapping structure, frequency band switching is performed after the trigger condition is met, and switching synchronization control and buffering strategies are adopted to avoid data transmission interruption; a first scoring threshold for frequency band switching and a second scoring threshold for frequency band switching are preset. When the comprehensive score of frequency band access reaches the preset first scoring threshold for frequency band switching, the switching recommendation state is entered; when the comprehensive score of frequency band access reaches the preset second scoring threshold for frequency band switching, the switching is forced; The package execution deviation compensation logic is used to calculate the deviation between the actual fee quota and the preset fee quota by collecting the terminal's data usage, the number of frequency band switches, the communication duration, and the unit fee of the current frequency band after the frequency band switch. When it is detected that the deviation value exceeds the preset deviation value threshold, the compensation logic is activated. By adjusting the unused traffic between terminals in the package sharing group across terminals, the corresponding unused traffic quota is allocated from the preset adjustable resource pool, and the unused traffic quota is allocated to the terminal according to the preset compensation ratio; the overall tariff quota deviation within the sharing group is balanced, and dynamic adjustment and compensation are performed.

9. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 8, characterized in that: The method for identifying a configuration strategy that affects the deviation of the actual tariff amount includes: Collect user feedback data, including actual user traffic usage, communication signal quality indicators, data transmission rate, communication stability evaluation, package price perception and satisfaction, and terminal abnormality logs; Establish a multi-dimensional correlation model between user feedback data and tariff deviations. Dynamically calculate and adjust the weight coefficients of each feedback item in the user feedback data through weighted fusion. The weight coefficients reflect the relative impact of each feedback item on the tariff deviation. The weighted user feedback data is integrated with the terminal's operating status data and package execution data to form a comprehensive data set. Based on the comprehensive data set, correlation coefficient analysis is used to preliminarily screen configuration policy parameters and user behavior indicators that are correlated with actual fee deviations. Granger causality analysis is then used to identify configuration strategies that lead to actual fee deviations.

10. The method for dynamically configuring Internet of Things card packages based on intelligent policy control according to claim 9, characterized in that: The method for replacing, freezing, or rolling back the configuration policy that causes the actual fee amount deviation includes: Identify and remove configuration policies that affect actual fee quota deviations, archive and manage the removed configuration policies and their version information, and automatically mark the corresponding configuration policy version as abnormal if it is detected that the actual fee quota deviation caused by any configuration policy version exceeds the preset fee quota deviation threshold. Replace the abnormal version with a new version configuration policy generated based on preset optimization rules, and verify the validity of the new version configuration policy in the terminal cluster. Before the replacement, freeze the configuration policy corresponding to the abnormal version, suspend its application on the newly added terminals, and maintain the operation status of the existing terminals. When the configuration policy corresponding to the new version fails verification, the configuration policy can be rolled back to the previous version. Combined with continuously collected user feedback data, terminal operation status data, and package execution data, the configuration policy effect is dynamically evaluated and the iterative update of the configuration policy is automatically triggered.

Images