Intelligent optimization method for eSIM service cost based on information sources of different operators

Abstract

The invention discloses an eSIM service cost intelligent optimization method based on different operator information sources, and particularly relates to the technical field of operator network management. The method comprises the following steps: collecting eSIM tariff rules of a plurality of operators and real-time charging data of eSIM terminals, extracting tariff terms and conditional expense trigger mechanisms hidden in charging behaviors of different operators by using an association rule mining algorithm, and constructing a multi-level cost structure model; on the basis of a network registration behavior log of the eSIM terminal, predicting the probability that each operator signal source has an abnormal registration event, and evaluating the risk that each operator signal source triggers a hidden cost trap in combination with a multi-level cost structure model; and according to the real-time network signal parameters of the eSIM terminal, a signal source recommendation strategy is generated in combination with the risk assessment result, and the eSIM terminal is driven to execute a signal source switching operation, so that real-time adjustment and cost optimization control of communication network selection are realized, and the service stability of the eSIM in a cross-network use scene is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of operator network management, and more specifically, to a method for intelligently optimizing eSIM service costs based on different operator sources. Background Art

[0002] When optimizing eSIM costs across operators, existing technologies mainly rely on analyzing and making decisions based on the operators' public and basic tariff information. They lack the ability to predict the true and complete cost structure of each operator, which is complex, hidden, dynamically changing, and comes with conditional clauses. As a result, in actual applications, optimization decisions cannot effectively foresee and avoid potential cost traps caused by non-transparent billing rules, unconventional usage restrictions, or sudden policy adjustments, making it difficult to achieve cost optimization and even risking cost out-of-control.

[0003] In order to solve the above problems, a technical solution is now provided. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method and system for intelligent optimization of eSIM service costs based on different operator sources to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: The intelligent optimization method for eSIM service costs based on different operators' signal sources includes the following steps: S1: Obtain eSIM tariff rules and real-time billing data of eSIM terminals from multiple operators, and use association rule mining algorithms to extract hidden tariff terms and conditional fee triggering mechanisms in the billing behaviors of different operators; S2: Build a multi-level cost structure model based on the hidden tariff terms and conditional fee trigger mechanisms of different operators; S3: Based on the real-time network registration behavior logs of eSIM terminals, predict the probability of abnormal network registration events occurring at each operator's signal source; S4: Based on a multi-level cost structure model and the probability of abnormal network registration events, the risk of triggering hidden cost traps for each operator's signal source is assessed; S5: Obtain the real-time network signal parameters of the eSIM terminal and generate a real-time recommendation strategy for the operator's signal source based on the risk assessment results; S6: Trigger the operator signal source switching operation of the eSIM terminal according to the operator signal source real-time recommendation strategy to complete the eSIM service cost optimization.

[0006] In a preferred embodiment, S1 is specifically: Collect eSIM tariff rules from multiple operators and build a structured tariff rule dataset; Collect real-time billing data generated by eSIM terminals under multiple operator signal sources and build a sample dataset of eSIM terminal billing behavior; Based on the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset, an association rule mining algorithm is used to analyze the billing characteristics of each operator under different usage conditions, and to extract the hidden tariff terms and conditional fee trigger mechanisms in the billing behaviors of different operators.

[0007] In a preferred embodiment, S2 is specifically: Build a multi-level cost structure model for different operators based on the hidden tariff terms and conditional fee trigger mechanisms in the billing behavior of different operators; The multi-tiered cost structure model includes a basic tariff layer, a conditional additional fee layer, and a cross-network operation cost structure.

[0008] In a preferred embodiment, S3 is specifically: Collect real-time network registration behavior logs of eSIM terminals during operation; A network behavior sequence dataset is constructed based on real-time network registration behavior logs, and registration behaviors under the same operator signal source are sequentially labeled in chronological order. The time series analysis method is used to calculate the probability of abnormal network registration events occurring at each operator's signal source in the current geographical location and network environment, and generate the predicted probability value of abnormal network registration events for each operator's signal source.

[0009] In a preferred embodiment, S4 is specifically: Obtaining rule feature vectors of the basic tariff layer, conditional additional fee layer, and cross-network operation cost layer in the multi-level cost structure model; The rule feature vector is weighted with the predicted probability value of abnormal network registration events of each operator's signal source to obtain the expected cost risk value of each structural layer; The expected cost risk values of each structural layer are accumulated to generate dynamic cost risk values corresponding to the hidden cost traps triggered by each operator's signal source.

[0010] In a preferred embodiment, S5 is specifically: Collect real-time network signal parameters of the eSIM terminal in the current geographical location and network environment; The real-time network signal parameters and the dynamic cost risk value of each operator's signal source are jointly input to construct a joint feature vector; Based on the joint feature vector, a multi-objective optimization algorithm is used to score and rank the currently available multiple operator signal sources; Generate real-time recommendation strategies for operator signal sources based on the scoring and ranking results.

[0011] In a preferred embodiment, S6 is specifically: Determine the target operator signal source based on the operator signal source's real-time recommendation strategy; Call the signal source switching control interface of the eSIM terminal to perform the operator signal source switching operation, switching the current operator signal source to the target operator signal source; After the switching operation is completed, the operator signal source configuration of the eSIM terminal is updated to complete the eSIM service cost optimization.

[0012] The technical effects and advantages of the present invention based on the intelligent optimization method of eSIM service costs of different operators are as follows: By mining association rules between multiple operators' eSIM tariff rules and real-time billing data of eSIM terminals, extracting hidden tariff terms and conditional triggering mechanisms, building a multi-level cost structure model, and combining the probability of abnormal network registration events to assess the risk of each operator's signal source triggering hidden cost traps, collecting real-time network signal parameters of eSIM terminals and generating a signal source optimization strategy based on the risk assessment results, driving the eSIM terminal to complete signal source switching, it can fully model the operator's complex billing system and predict the cost deviation caused by non-transparent billing rules, achieving optimal business costs in cross-operator scenarios, and improving cost control accuracy, usage stability and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 Schematic diagram of the intelligent optimization method for eSIM service costs based on different operators' signal sources according to the present invention. DETAILED DESCRIPTION

[0014] The following will provide a clear and complete description of 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0015] Example 1

[0016] Figure 1 The present invention provides an intelligent optimization method for eSIM service costs based on different operator sources, which includes the following steps: S1: Obtain eSIM tariff rules and real-time billing data of eSIM terminals from multiple operators, and use association rule mining algorithms to extract hidden tariff terms and conditional fee triggering mechanisms in the billing behaviors of different operators; S2: Build a multi-level cost structure model based on the hidden tariff terms and conditional fee trigger mechanisms of different operators; S3: Based on the real-time network registration behavior logs of eSIM terminals, predict the probability of abnormal network registration events occurring at each operator's signal source; S4: Based on a multi-level cost structure model and the probability of abnormal network registration events, the risk of triggering hidden cost traps for each operator's signal source is assessed; S5: Obtain the real-time network signal parameters of the eSIM terminal and generate a real-time recommendation strategy for the operator's signal source based on the risk assessment results; S6: Trigger the operator signal source switching operation of the eSIM terminal according to the operator signal source real-time recommendation strategy to complete the eSIM service cost optimization.

[0017] S1: Obtain eSIM tariff rules and real-time billing data from multiple operators and eSIM terminals. Use association rule mining algorithms to extract hidden tariff terms and conditional fee triggering mechanisms from different operators' billing behaviors, including: Collect eSIM tariff rules from multiple operators and build a structured tariff rule dataset; Specifically, a crawler program is used to retrieve tariff rule text data from each operator's official website or application interface. This data includes information such as basic tariffs, value-added service charges, package overlay rules, network access fees, roaming billing methods, and network switching additional billing rules. This data is then fed into a pre-defined tariff rule text parsing model, a text structuring model based on natural language processing. The model comprises a text segmentation module, a semantic feature extraction module, and a rule structure mapping module. The text segmentation module performs Chinese word segmentation and syntactic analysis on the tariff rule text data to obtain refined word units. The semantic feature extraction module classifies and identifies the semantic categories corresponding to each word, such as tariff category, applicable conditions, billing unit, and triggering rule. The rule structure mapping module maps the semantic categories according to a pre-defined tariff rule structure template, creating a clearly structured multi-field record to produce a structured tariff rule dataset. The eSIM, also known as an embedded subscriber identity module (eSIM), is a secure storage unit integrated into a device as a soft entity in accordance with GSMA specifications.

[0018] Collect real-time billing data generated by eSIM terminals under multiple operator signal sources and build a sample dataset of eSIM terminal billing behavior; Specifically, a real-time data collection program is deployed on the eSIM terminal. This program uses one minute as a collection cycle and, by calling the terminal's underlying communication interface, records the terminal's traffic usage, billing start and end times, actual billing amount, network identity information, user behavior tags, and billing trigger events in real time. Traffic usage records the data traffic generated during each data transmission process of the terminal; billing start and end times record the start and end times of each billing; the actual billing amount records the actual costs incurred for each billing; network identity information records the signal source identifier of the operator currently registered on the network; user behavior tags record the classification of usage behavior, such as video playback, web browsing, or file downloads; and billing trigger events record the specific reasons for the charge trigger, such as network switching, package overage, or roaming. The real-time data collection program integrates the collected information into complete data records and saves them in a structured form to form a sample dataset of eSIM terminal billing behavior.

[0019] The data collection program deployed on the eSIM terminal only collects technical parameters directly related to the tariff calculation, and explicitly does not involve sensitive data such as user identity information, communication content, location trajectory, account authentication information, etc.; among them, user behavior labels are limited to anonymizing and categorizing traffic usage behaviors (such as video, web page, download), and do not record specific visited URLs or application names. Local feature extraction and desensitizing encoding methods are used to ensure that behavior labels are only used for billing feature analysis to prevent privacy leaks.

[0020] Based on a dataset of structured tariff rules and a sample dataset of eSIM terminal billing behavior, we used an association rule mining algorithm to analyze the billing characteristics of each operator under different usage conditions and extract the hidden tariff terms and conditional fee triggering mechanisms in the billing behaviors of different operators. Specifically, data preprocessing is performed on the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset. Data preprocessing includes standardizing the data records in the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset, removing abnormal data, and completing missing data. Standardization involves unifying the data format and units, such as unifying the data unit to megabytes for traffic and RMB for fees. Abnormal data removal involves removing records with obvious errors, such as records with obviously abnormal billing amounts. Missing data completion uses historical averages or regression prediction methods to ensure dataset quality.

[0021] After data preprocessing, the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset are imported into the association rule mining algorithm. The association rule mining algorithm specifically uses a priori algorithms, which include a frequent item set generation step and a rule mining step. In the frequent item set generation step, the co-occurrence frequency of records is analyzed based on the usage condition characteristics and the actual billing event characteristics; the rule mining step extracts hidden tariff terms and conditional fee trigger mechanisms that meet the support threshold and confidence threshold in the frequent item set. For example, hidden tariff terms include rules that are not explicitly disclosed but will trigger repeated charges, network re-access charges, or skipped billing after package overage when actually implemented; conditional fee trigger mechanisms include additional tariff change rules caused by changes in specific usage time periods, network categories, roaming status, registration frequency, and cumulative daily usage.

[0022] By setting different support and confidence thresholds, you can optimize the screening of hidden tariff terms and conditional fee triggering rules. The support threshold indicates the proportion of a specific billing feature combination in all records; the confidence threshold indicates the probability of a billing event occurring under certain conditions. Choosing appropriate support and confidence thresholds can effectively reduce the risk of misjudgment and improve analysis accuracy.

[0023] S2: Based on the hidden tariff terms and conditional fee trigger mechanisms of different operators, a multi-level cost structure model is constructed, including: Build a multi-level cost structure model for different operators based on the hidden tariff terms and conditional fee trigger mechanisms in the billing behavior of different operators; The multi-tiered cost structure model includes a base tariff layer, a conditional additional fee layer, and a cross-network operating cost structure; Specifically, based on the extracted hidden tariff terms and conditional fee trigger mechanisms, a multi-level cost structure model for different telecom operators was constructed. A three-layer data model was designed, consisting of a basic tariff structure layer, a conditional additional fee structure layer, and a cross-network operation cost structure layer. The three-layer data model was chosen because different operators' billing behaviors involve three dimensions: basic charges, conditional trigger charges, and network switching charges. The three-layer data model can clearly express the triggering logic of different types of billing, facilitating cost analysis and optimized decision-making.

[0024] Construct the basic tariff structure layer in the three-layer data model. The basic tariff structure layer represents the tariff rule information of different communication operators under normal conditions, such as the unit price of data traffic, the billing rules for voice calls, and the calculation method of SMS charges. The basic tariff data in the structured tariff rule data set is classified into tariff rule entries. The classification method includes dividing the basic tariff data into traffic tariff data, voice tariff data, and SMS tariff data according to different communication methods. Among them, traffic tariff data is specifically described in terms of unit data volume price, such as the fee corresponding to each megabyte of data traffic; voice tariff data specifically describes how call duration is billed, such as the fee per minute of call; SMS tariff data specifically describes the fixed charge standard for sending a single SMS.

[0025] Data related to general usage conditions in the structured tariff rule dataset is identified and extracted as basic tariff data. General usage conditions refer to usage scenarios that do not include time restrictions, geographic location restrictions, usage threshold triggers, network switching, or package status conditions. For example, these include voice, SMS, and data services generated by users accessing the local network normally.

[0026] Construct a conditional additional fee structure layer in a three-layer data model. The conditional additional fee structure layer is used to represent tariff rule information triggered by specific usage conditions, such as additional fees during special usage time periods, specific data usage thresholds, user package status, or special geographical locations. Record the conditional fee triggering mechanisms as separate tariff rule entries. Each tariff rule entry records the triggering conditions, specific fees, and the specific implementation methods of the triggering rules. For example, one of the tariff rule entries is described as: "When a user exceeds the set data usage threshold during a non-standard time period (such as nighttime), the resulting traffic charges are calculated at a specific rate higher than the basic rate"; another tariff rule entry is described as: "When the user's cumulative daily registration times exceed the threshold set by the operator, an additional registration fee will be incurred." Each of the above tariff trigger rule entries independently stores and marks the corresponding triggering conditions and tariff standards.

[0027] Construct the cross-network operation cost structure layer within the three-layer data model. The cross-network operation cost structure layer represents the tariff rule information for additional network fees and network access penalty fees triggered by special operations such as network switching, cross-border network registration, and non-local network access. Based on the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset, extract and record specific cross-network operation tariff rule entries. Cross-network operation tariff rule entries describe the tariff changes that occur when registering or switching between different telecom operators' networks. For example, one cross-network operation tariff rule entry might read: "When a terminal switches from a domestic network to an international roaming network, specific roaming registration fees and network switching surcharges will be incurred." Another cross-network operation tariff rule entry might read: "When a terminal frequently switches networks, each time a predetermined number of times is exceeded, an additional penalty network access fee will be triggered." Each of these cross-network operation tariff rule entries is recorded separately, and the corresponding triggering scenarios and fee standards are individually identified.

[0028] All tariff rule entries in the basic tariff structure layer, conditional additional fee structure layer, and cross-network operation cost structure layer are combined and associated to form a complete multi-level cost structure model. During implementation, a tariff rule association mapping relationship is established. This relationship describes the relationship and influence mechanism between each structure layer. For example, the basic tariff structure layer provides basic billing information, while the conditional additional fee structure layer and cross-network operation cost structure layer provide scenarios and tariff change methods for incurring additional fees beyond the basic tariff.

[0029] S3: Based on the real-time network registration behavior logs of eSIM terminals, the probability of abnormal network registration events occurring at each operator's signal source is predicted, including: Collect real-time network registration behavior logs of eSIM terminals during operation; Specifically, a network registration behavior log collection program is configured. The network registration behavior log collection program collects the terminal network registration status and records network registration behavior events at a fixed sampling interval. The collected network registration behavior logs include but are not limited to the following fields: network registration success events, network registration failure events, network switching records, registration duration, switching failure reason code, communication network identity information, and registration request timestamp. The network registration behavior log collection program uniformly encodes the above fields and writes the encoded network registration behavior logs to the local cache file in real time. The network registration behavior log collection program supports geographic tagging based on the terminal's geographic location information to ensure that network registration behaviors in different geographical environments can be accurately distinguished.

[0030] A network behavior sequence dataset is constructed based on real-time network registration behavior logs, and registration behaviors under the same operator signal source are sequentially labeled in chronological order. Specifically, using communication network identity information as the primary key, registration behavior records belonging to the same communication operator in the network registration behavior log are classified and aggregated. After classification, the registration behavior events corresponding to each operator signal source are chronologically sorted according to the registration request timestamp, forming a time-continuous behavior sequence. A behavior sequence is a time-continuous record of all registration behavior events from the same operator signal source, indexed by the registration request timestamp, arranged from earliest to latest. Each sequence corresponds to a single operator signal source. A behavior sequence is the basic unit of a network behavior sequence dataset, which consists of multiple behavior sequences. Each registration behavior event is assigned a status label in the behavior sequence. Status labels include registration success, registration failure, short-term deregistration, network handover interruption, and connection stability. Status labels are assigned based on the specific event type and registration duration parameters. For example, if a registration behavior event is actively released within 50 seconds after connection completion and a new registration request is initiated, the registration behavior event is marked as short-term deregistration.

[0031] Using time series analysis methods, we calculate the probability of abnormal network registration events occurring at each operator's signal source in the current geographic location and network environment, and generate a predicted probability value for abnormal network registration events at each operator's signal source. Specifically, the network behavior sequence dataset extracts behavioral subsequences for each operator's signal source in its current geographic location and network environment. The behavioral subsequence is an ordered data stream with time as the primary axis and state labels as the content. The time series analysis method employed is a sliding window-based probability prediction model for abnormal network registration events. This model analyzes the transition patterns of state labels within each sliding window period and predicts the probability of abnormal network registration events occurring within future sliding window periods. Abnormal network registration events include high-frequency handovers, failed registration retries, deregistration and reregistration within a short period of time, and unstable connections after handover. The abnormal network registration event probability prediction model constructs a transition probability matrix by statistically analyzing the transition frequency and average duration of abnormal network registration events within several previous sliding windows. It then outputs the predicted probability values for each type of abnormal network registration event occurring within future sliding windows.

[0032] Taking high-frequency switching events as an example, if the same communication operator signal source has three registration failures and two short-term deregistration events within the first three sliding window periods, and the event interval is less than the preset maximum interval time threshold, then the abnormal network registration event probability prediction model calculates the predicted probability of high-frequency switching of the communication operator signal source in the next sliding window period as the value of the transition from the normal state to the high-frequency switching state in the transition probability matrix.

[0033] S4: Based on a multi-level cost structure model and the probability of abnormal network registration events, assess the risk of each operator's signal source triggering hidden cost traps, including: Obtaining rule feature vectors of the basic tariff layer, conditional additional fee layer, and cross-network operation cost layer in the multi-level cost structure model; Specifically, according to the tariff rule information of the basic tariff layer, the conditional additional fee layer and the cross-network operation cost layer in the multi-level cost structure model, the tariff rule information of each structural layer is converted into the form of rule feature vectors, specifically: the tariff rule information of the basic tariff layer is field standardized, including unifying the dimensions of traffic tariff data, voice tariff data and SMS tariff data and setting them as basic traffic tariff unit price, basic voice communication tariff unit price and SMS tariff unit price respectively; for the tariff rule information defined in the conditional additional fee layer, a mapping relationship is established according to the trigger condition type, and the tariff increase coefficient corresponding to each trigger condition is extracted; for the rule items such as cross-network switching fee, roaming registration fee, frequent switching penalty fee involved in the cross-network operation cost layer, the cross-network switching single cost, roaming registration fee coefficient, and frequent switching penalty fee coefficient are assigned respectively; all components in each structural layer are arranged in sequence to form a rule feature vector for each structural layer. A rule feature vector is a set of rule attribute features represented by numerical values. For example, at the basic tariff level, the numerical attributes of the rule feature vector include the basic traffic tariff unit price, the basic voice communication tariff unit price, and the SMS tariff unit price. At the conditional additional fee level, the numerical attributes of the rule feature vector include the tariff increase coefficients under different triggering conditions, such as the tariff increase coefficients triggered by specific time periods, traffic thresholds, daily registration frequencies, and geographical regions. At the cross-network operation cost level, the numerical attributes of the rule feature vector include the single cost of cross-network switching, the roaming registration fee coefficient, and the frequent switching penalty fee coefficient. The rule feature vector is extracted using numerical normalization, and all feature values are uniformly adjusted to a numerical range between zero and one to ensure comparability between rule feature vectors and eliminate the impact of unit and scale differences between different tariff types.

[0034] The rule feature vector is weighted with the predicted probability value of abnormal network registration events of each operator's signal source to obtain the expected cost risk value of each structural layer; Specifically, the obtained rule feature vector is weighted and calculated with the predicted probability value of abnormal network registration events of the corresponding communication operator signal source to obtain the expected cost risk value of each structural layer. The probability of each type of abnormal network registration event occurring for each operator signal source is determined based on the predicted probability value of abnormal network registration events of each operator signal source, and is applied as a risk weight to the rule feature vector. The specific calculation method is as follows: the expected cost risk value of the basic tariff layer is obtained by multiplying the basic traffic tariff unit price, basic voice communication tariff unit price, and SMS tariff unit price by the predicted probability value of each type of abnormal network registration event, and then summing them up; the expected cost risk value of the conditional additional fee layer is obtained by multiplying the tariff increase coefficient under different trigger conditions by the predicted probability value of each type of abnormal network registration event, and then summing them up; the expected cost risk value of the cross-network operation cost layer is obtained by multiplying the single cost of cross-network switching, the roaming registration fee coefficient, and the frequent switching penalty fee coefficient by the predicted probability value of each type of abnormal network registration event, and then summing them up.

[0035] Accumulate the expected cost risk values of each structural layer to generate dynamic cost risk values corresponding to the hidden cost traps triggered by each operator's signal source; Specifically, hidden cost traps refer to additional charges or rate fluctuations that are not directly listed in the telecom operator's public tariff documents but arise under specific usage conditions. These hidden cost traps are caused by the combined effects of hidden tariff clauses and conditional fee triggering mechanisms.

[0036] S5: Obtain the real-time network signal parameters of the eSIM terminal and, based on the risk assessment results, generate a real-time recommendation strategy for the operator's signal source, including: Collect real-time network signal parameters of the eSIM terminal in the current geographical location and network environment; Specifically, the terminal's built-in wireless signal quality monitoring module collects parameters at a fixed interval. The collected real-time network signal parameters include network signal strength, network signal quality, base station coverage, base station load level, and network latency. Network signal strength is measured in units of wireless signal power; network signal quality is expressed as the signal-to-interference ratio; base station coverage is expressed as the wireless propagation distance between the terminal and the base station; base station load level is expressed as the ratio between the number of currently connected users and the theoretical maximum user capacity of the base station; and network latency is measured by sending a network probe packet from the terminal to the base station and recording the time difference between sending and receiving the packet.

[0037] The real-time network signal parameters and the dynamic cost risk value of each operator's signal source are jointly input to construct a joint feature vector; Specifically, the real-time network signal parameters corresponding to each carrier's signal source are aligned with the dynamic cost risk value to the same scale range. A normalization method is then used to map all values to a standard scale between zero and one. The normalized network signal strength, network signal quality, base station load level, network latency, and dynamic cost risk value are then arranged in an ordered sequence to form a joint feature vector corresponding to each carrier's signal source.

[0038] Based on the joint feature vector, a multi-objective optimization algorithm is used to score and rank the currently available multiple operator signal sources; Specifically, the multi-objective optimization algorithm selected uses a multi-objective evolutionary optimization algorithm based on non-dominated sorting. This algorithm takes the joint feature vector of each carrier's signal source as input and optimizes the scoring and prioritization of carrier signal sources through non-dominated sorting and congestion calculation. The specific implementation method for non-dominated sorting is to use network quality and cost risk as optimization objectives. The network quality objective is to maximize network signal strength and quality while minimizing network latency, and the cost risk objective is to minimize the dynamic cost risk value. Compare the joint feature vectors of all optional communication operator signal sources one by one. For any two communication operator signal sources, if one of the signal sources has better network quality and lower cost risk, it is marked as a non-dominated solution. For signal sources where the non-dominated relationship cannot be clearly determined, sort them by congestion calculation. Specifically, for any communication operator signal source, sort all communication operator signal sources in each feature dimension of the joint feature vector from small to large. For each signal source, calculate the numerical difference between the two adjacent signal sources in each dimension, and then divide the difference between the previous signal source and the next signal source by two. Sum the differences in all the above dimensions as the congestion of the signal source. The smaller the congestion, the more stable the signal source characteristics, and the higher the priority recommendation. Through the above sorting process, a score ranking list of all communication operator signal sources is finally obtained.

[0039] Generate real-time recommendation strategies for operator signal sources based on the rating ranking results; Specifically, a recommendation strategy generation rule is set, which selects the carrier signal source with the highest ranking position and the highest non-dominated level in the score ranking list as the preferred network for the current terminal in the next time period. If multiple carrier signal sources have the same non-dominated level, the target carrier signal source with the highest priority is determined based on the lowest congestion value, allowing the terminal to perform network switching operations. The benefit of generating a real-time recommendation strategy is that the terminal can dynamically adapt to changes in the current network environment and price risks, avoiding hidden cost traps while ensuring network service quality, effectively reducing business cost risks.

[0040] S6: Triggering the operator signal source switching operation of the eSIM terminal based on the operator signal source real-time recommendation strategy to optimize the eSIM service cost, including: Determine the target operator signal source based on the operator signal source's real-time recommendation strategy; Specifically, the real-time recommendation strategy includes the target operator's signal source with the highest recommendation priority, including the target operator's public land mobile network identification code, operator's internal network code, network registration type identifier, and base station identification information. The terminal interprets the real-time recommendation strategy and selects the target operator's signal source with the highest recommendation priority as the communication connection target for the next time period.

[0041] Call the signal source switching control interface of the eSIM terminal to perform the operator signal source switching operation, switching the current operator signal source to the target operator signal source; Specifically, the terminal calls the preset network access control module and passes the network identification parameters of the target communication operator's signal source to it. The network access control module initiates a registration request to the target communication network and enters the switching processing state. The registration request process includes the following sub-operations: releasing the current communication network connection resources, initializing the target network access protocol stack, completing the network identity authentication and authority confirmation process, and establishing a connection maintenance channel. If the registration request returns a confirmation response and the signal quality parameters meet the set threshold conditions, the target signal source registration is completed and the main communication channel is established. If the registration fails or the signal quality does not meet the standard, the switching failure event is recorded, and the current recommended strategy result is marked as a low availability state, and the next priority target operator signal source is re-evaluated for replacement attempt.

[0042] After the switching operation is completed, the operator signal source configuration of the eSIM terminal is updated to complete the eSIM service cost optimization; Specifically, after the signal source is successfully switched, the terminal calls the network status synchronization module to update the identification information and supporting network parameters of the currently connected communication operator's signal source. The updated content includes the network identity information of the target operator's signal source, the allocated communication resource number, the network registration timestamp, the access network type and other data.

[0043] eSIM service cost optimization refers to the optimization of communication fee expenditure incurred by end users when using cross-operator networks, that is, reducing customers' usage costs and helping users minimize communication fees while ensuring network performance.

[0044] 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.

[0045] The above embodiments can be implemented in whole or in part via software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in the embodiments of this application are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0046] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0047] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0049] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0050] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0051] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0052] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0053] Finally: 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, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent optimization method for eSIM service costs based on different operators' signal sources, characterized by: The steps include: S1: Obtain eSIM tariff rules and real-time billing data of eSIM terminals from multiple operators, and use association rule mining algorithms to extract hidden tariff terms and conditional fee triggering mechanisms in the billing behaviors of different operators; S2: Build a multi-level cost structure model based on the hidden tariff terms and conditional fee trigger mechanisms of different operators; S3: Based on the real-time network registration behavior logs of eSIM terminals, predict the probability of abnormal network registration events occurring at each operator's signal source; S4: Based on a multi-level cost structure model and the probability of abnormal network registration events, the risk of triggering hidden cost traps for each operator's signal source is assessed; S5: Obtain the real-time network signal parameters of the eSIM terminal and generate a real-time recommendation strategy for the operator's signal source based on the risk assessment results; S6: Trigger the operator signal source switching operation of the eSIM terminal according to the operator signal source real-time recommendation strategy to complete the eSIM service cost optimization.

2. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 1 is characterized in that: S1, specifically: Collect eSIM tariff rules from multiple operators and build a structured tariff rule dataset; Collect real-time billing data generated by eSIM terminals under multiple operator signal sources and build a sample dataset of eSIM terminal billing behavior; Based on the structured tariff rule dataset and the eSIM terminal billing behavior sample dataset, an association rule mining algorithm is used to analyze the billing characteristics of each operator under different usage conditions, and to extract the hidden tariff terms and conditional fee trigger mechanisms in the billing behaviors of different operators.

3. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 2, characterized in that: S2, specifically: Build a multi-level cost structure model for different operators based on the hidden tariff terms and conditional fee trigger mechanisms in the billing behavior of different operators; The multi-tiered cost structure model includes a basic tariff layer, a conditional additional fee layer, and a cross-network operation cost structure.

4. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 3 is characterized in that: S3, specifically: Collect real-time network registration behavior logs of eSIM terminals during operation; A network behavior sequence dataset is constructed based on real-time network registration behavior logs, and registration behaviors under the same operator signal source are sequentially labeled in chronological order. The time series analysis method is used to calculate the probability of abnormal network registration events occurring at each operator's signal source in the current geographical location and network environment, and generate the predicted probability value of abnormal network registration events for each operator's signal source.

5. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 4 is characterized in that: S4, specifically: Obtaining rule feature vectors of the basic tariff layer, conditional additional fee layer, and cross-network operation cost layer in the multi-level cost structure model; The rule feature vector is weighted with the predicted probability value of abnormal network registration events of each operator's signal source to obtain the expected cost risk value of each structural layer; The expected cost risk values of each structural layer are accumulated to generate dynamic cost risk values corresponding to the hidden cost traps triggered by each operator's signal source.

6. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 5, characterized in that: S5, specifically: Collect real-time network signal parameters of the eSIM terminal in the current geographical location and network environment; The real-time network signal parameters and the dynamic cost risk value of each operator's signal source are jointly input to construct a joint feature vector; Based on the joint feature vector, a multi-objective optimization algorithm is used to score and rank the currently available multiple operator signal sources; Generate real-time recommendation strategies for operator signal sources based on the scoring and ranking results.

7. The method for intelligent optimization of eSIM service costs based on different operator sources according to claim 6, characterized in that: S6, specifically: Determine the target operator signal source based on the operator signal source's real-time recommendation strategy; Call the signal source switching control interface of the eSIM terminal to perform the operator signal source switching operation, switching the current operator signal source to the target operator signal source; After the switching operation is completed, the operator signal source configuration of the eSIM terminal is updated to complete the eSIM service cost optimization.

Images