APN delivery efficiency optimization method for eSIM dynamic switching

Abstract

The present invention relates to the field of wireless communication network technology, and discloses a method for optimizing APN delivery efficiency for dynamic eSIM switching, comprising: an eSIM terminal obtaining capability labels of each APN on the network side, analyzing upper-layer application traffic characteristics to extract a profile of the current application's demand characteristics, and then autonomously selecting and connecting to a suitable APN based on the matching result between the profile and the label. Through terminal application perception and APN capability identification capabilities, the present invention enables the APN selection process to dynamically respond to the immediate needs of applications, ensuring a high degree of match between network access points and application characteristics from the initial connection, laying the foundation for subsequent refined adjustments based on actual network status and energy efficiency, and significantly optimizing the accuracy of wireless resource allocation and terminal experience.

Description

Technical Field

[0001] The present invention relates to an APN delivery efficiency optimization method for dynamic switching of eSIMs, and belongs to the technical field of wireless communication networks. Background Art

[0002] Against the backdrop of the rapid evolution of mobile communication networks and the large-scale deployment of eSIMs (embedded subscriber identity modules), terminal devices can dynamically switch operators without having to replace physical cards, which has become a key development direction of the next-generation communication architecture. However, the flexibility of eSIM dynamic switching also places higher demands on the network access process, especially in the configuration and delivery of APNs (Access Point Names). How to achieve efficient and accurate matching and selection under changing network environments and application requirements has become a common technical challenge facing the industry.

[0003] In the traditional mechanism, after completing the attachment or registration process, the eSIM terminal usually obtains an APN from the network side for subsequent data connection based on a static preset or default policy. This APN is often bound to the user's package or operator policy and lacks the ability to perceive the current application scenario. This single-designated APN selection method is inadequate in the increasingly complex application environment, especially in scenarios where a large amount of diverse application traffic (such as high-definition video, low-latency interaction, IoT small data synchronization, etc.) is running concurrently. The traditional method is difficult to achieve the optimal match of resource allocation, which in turn affects the terminal's experience stability and the overall utilization efficiency of network resources.

[0004] Further observation reveals that in typical multi-service concurrent scenarios, traditional technical solutions suffer from the following deep-seated technical bottlenecks: 1. A lack of a real-time analysis mechanism based on the traffic characteristics of the terminal's current applications, resulting in an inability to dynamically adapt APN selection to specific service needs; 2. The representation of APN service capabilities is opaque, and the network side fails to provide standardized capability labels for the terminal to reference, limiting the terminal's ability to make refined judgments; 3. Even if there are multiple potential APN options, it is difficult for the terminal to accurately assess their response performance before establishing a connection, making it impossible to effectively avoid access points with high latency or congested paths, resulting in uncertainty in the initial connection experience. Although some operators have attempted to improve this through policy delivery or pre-set APN classification methods, these methods generally fail to achieve coordinated updates with the terminal's actual application status and fail to fully perceive the real-time status of the network path before establishing a connection, making it difficult to fundamentally solve the above problems. Therefore, how to implement a mechanism that supports eSIM terminals in making dynamic judgments based on application profiles and network capability information, and efficiently selects the best APN candidate path based on real-time response status to optimize APN delivery efficiency and connection experience, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] The present invention provides an APN issuance efficiency optimization method for dynamic switching of eSIMs. Its main purpose is to solve the problem that eSIM terminals lack a dynamic matching mechanism based on application requirements and network capabilities in a multi-service environment, making it difficult to achieve accurate and efficient APN selection.

[0006] To achieve the above objectives, the present invention provides a method for optimizing APN delivery efficiency for dynamic eSIM switching, characterized in that the method comprises the following steps:

[0007] Step 1: The eSIM terminal obtains the APN capability tag provided by the network side. The APN capability tag describes the service capability characteristics of each available access point name (APN).

[0008] Step 2: The eSIM terminal analyzes the traffic statistics generated by the upper-layer application within the specified monitoring time window to extract a demand feature profile of the currently running application. The traffic statistics include one or more of the following: average packet size, uplink / downlink traffic ratio, and packet transmission frequency per unit time.

[0009] In step 3, the eSIM terminal performs a matching calculation based on the extracted demand feature profile and the obtained APN capability tag to select one or more candidate APNs that match the demand feature profile; and the eSIM terminal establishes a data connection through a final APN determined from one or more candidate APNs.

[0010] Preferably, the APN capability tag is defined by the network operator and configured on the network side, and is sent to the eSIM terminal through the cellular network system information broadcast, or is sent to the eSIM terminal through the extended information carried when the eSIM terminal performs the attachment process, tracking area update TAU process or routing area update RAU process, or is sent together when the eSIM configuration file is downloaded and updated.

[0011] Preferably, the eSIM terminal extracts the demand feature profile of the currently running application by analyzing the traffic statistical characteristics generated by the upper-layer application within a determined monitoring time window, specifically: without performing deep packet inspection, the application type sniffing and demand abstraction module in the eSIM terminal monitors and counts the total uplink and downlink traffic, the total number of data packets, the average data packet size of uplink data transmission, and the average data packet size of downlink data transmission within the determined monitoring time window; calculates the uplink and downlink traffic ratio and the average packet sending frequency per unit time based on the statistical results; and compares the calculated uplink and downlink traffic ratio, the average packet sending frequency per unit time, and the average data packet size of uplink and downlink data transmission with the characteristic parameter thresholds corresponding to different application demand feature profile templates stored in the eSIM terminal to determine the demand feature profile that matches the currently running application.

[0012] Preferably, after the eSIM terminal selects one or more candidate APNs that match the demand profile, and before establishing a data connection through the final APN, the following steps are also included: Step 1, if it is determined that there are at least two candidate APNs, the eSIM terminal initiates an initial access signaling interaction for each of the at least two candidate APNs, and the initial access signaling interaction is intended to obtain the instantaneous response status of its corresponding network path; Step 2, the eSIM terminal captures and compares the micro-timing delay of each initial access signaling interaction from the time it is issued to the time it receives the first clear response signaling from the network side ; and the eSIM terminal selects micro timing delay The candidate APN with the smallest value is taken as the final APN, where the micro-timing delay Indicates the response agility of the corresponding candidate APN network path.

[0013] Preferably, the initial access signaling interaction is a PDU session establishment request NAS message that does not contain complete user plane parameters, or the initial signaling of the DNN selection process for different core network data network names when the RRC layer establishes a connection.

[0014] Preferably, after the eSIM terminal establishes a data connection through the final APN, the following steps are also included: Step 1, the application type sniffing and demand abstraction module in the eSIM terminal continuously monitors the application service quality QoS indicator through the current data connection, the QoS indicator includes the throughput perceived by the eSIM terminal or the round-trip time RTT estimated through application layer feedback information; Step 2, if the monitored QoS indicator is lower than the QoS expectation baseline associated with the current demand feature profile by a predetermined deviation threshold, and there are other alternative APNs whose APN capability tags match the demand feature profile in the local cache, the eSIM terminal actively attempts to switch to one of the alternative APNs.

[0015] Preferably, after the eSIM terminal establishes a data connection through the final APN, the following steps are also included: Step 1, the eSIM terminal profiles the dual context information based on the APN capability tag of the final APN and the previously extracted demand characteristics of the currently running application; Step 2, from the energy-saving strategy knowledge base preset in the eSIM terminal, query and determine a set of energy-saving communication parameter configurations that match the current dual context information combination, the energy-saving strategy knowledge base defines energy-saving communication parameter adjustment rules corresponding to different context combinations; Step 3, the communication module of the eSIM terminal adjusts at least one of its current radio resource control RRC parameters, discontinuous reception DRX cycle parameters or power saving mode PSM configuration parameters based on the determined energy-saving communication parameter configuration to optimize energy consumption when conducting data communication through the final APN.

[0016] Preferably, the energy-saving communication parameter adjustment rules defined in the energy-saving strategy knowledge base include: if the APN capability label is energy-saving connection and the application requirement characteristic profile is small data intermittent type, then configure the communication module to adopt a longer DRX cycle and enable PSM; or if the APN capability label is low latency optimization and the application requirement characteristic profile is changed to background data synchronization, then configure the communication module to extend the DRX cycle.

[0017] Preferably, the APN capability tag is a set of standardized descriptors used to characterize the core service characteristics of the APN, and the core service characteristics are selected from one or more of: low latency optimization, uplink bandwidth priority, downlink stable transmission, energy-saving connection and basic Best Effort service.

[0018] Compared with the background technology problems, the beneficial effects of the present invention are:

[0019] 1. This method enables the eSIM terminal to form a dynamic application demand feature profile based on real-time analysis of the actual traffic characteristics of the currently running application, and actively obtain the standardized capability labels provided by the network for each APN. By matching these two - that is, the real-time application requirements on the terminal side and the APN service characteristics declared by the network side - the terminal can autonomously select one or more candidate APNs that best match the current application requirements. This mechanism ensures that APN selection is no longer statically preset or blindly assigned, but is application-oriented from the beginning of connection establishment, laying a characteristically appropriate access foundation for upper-layer applications, thereby improving the matching degree between network resources and application requirements.

[0020] 2. After initially screening candidate APNs based on application requirements and APN capability tags, this method further introduces a refined assessment mechanism for the instantaneous state of network paths. Specifically, the eSIM terminal initiates a minimalist initial access signaling interaction for each candidate APN and captures the micro-timing delay caused by the network's first clear response to this signaling. By comparing these delays, the terminal can identify the most responsive network path and prioritize the APN corresponding to this path to establish the final data connection. This step goes beyond considering the static service capabilities of the APN and incorporates the immediate activity of the network path into the decision-making process, ensuring that the APN selection not only matches the functional requirements of the application but also adapts to the actual network load conditions at the moment.

[0021] 3. After the data connection is established, this method can use the capability tag of the selected APN and the previously extracted current application demand feature portrait as dual context information. Based on this combined information, the eSIM terminal can query and determine a set of matching energy-saving communication parameter configurations from its internal preset energy-saving strategy knowledge base, and adjust its own wireless resource control parameters, discontinuous reception cycle or power saving mode configuration accordingly; this design makes the terminal's energy-saving behavior no longer isolated and universal, but can be refined and dynamically adapted according to the current application scenario and the characteristics of the connected APN, thereby effectively optimizing the terminal's energy consumption while ensuring the application experience.

[0022] 4. This method builds a closed-loop system with continuous optimization. After establishing a data connection through the final APN, the eSIM terminal will continue to monitor the application service quality indicators through the current data connection. If the monitored service quality indicators are lower than the expected baseline associated with the current demand feature profile, and there are other alternative APNs matching the demand feature profile in the local cache, the terminal will actively try to switch to one of the alternative APNs. This mechanism combines intelligent selection during the initial connection with continuous quality perception and adaptive adjustment during the connection process, enabling the eSIM terminal to dynamically maintain and improve service quality in the face of network fluctuations or changes in application requirements, thereby improving the resilience of the connection and the stability of the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a performance comparison chart of the present invention and traditional static selection in video streaming and IoT application scenarios;

[0024] Figure 2 This is a flowchart of access optimization for the eSIM terminal of the present invention based on matching application requirement feature profiles with APN capability tags;

[0025] Figure 3This is a flowchart of the intelligent matching and path evaluation optimization of the eSIM terminal based on capability tags and feature profiles of the present invention.

[0026] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0027] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0028] The present invention proposes a method for optimizing APN issuance efficiency for dynamic switching of eSIM. Its core goal is to realize an intelligent access point selection mechanism between the terminal and the network side in a changing application environment, thereby improving access accuracy, shortening connection delay, and enhancing resource matching and service continuity after connection. This technical solution uses terminal application traffic perception + network capability label + multi-APN path evaluation as a three-in-one collaborative mechanism to construct a complete APN selection closed-loop path with a high degree of engineering feasibility. Specifically, the eSIM terminal first obtains the APN capability label provided by the network side. The label describes the service capability characteristics of each APN in a standardized form, and constructs a feature profile of the current application requirements locally by analyzing the application traffic characteristics. Then, the terminal selects one or more candidate APNs based on the matching result of the portrait and the APN capability label. On this basis, the instant responsiveness of each candidate path is further evaluated, and the final access APN is determined with the optimal delay path, and a data connection is established accordingly. This method defines and uses a structured description of APN capability tags as a standard interface for exposing network-side capabilities to terminals. These tags, developed by operators, typically include information on dimensions such as whether low latency is optimized, whether uplink bandwidth priority is supported, whether downlink stable transmission is suitable, and whether energy-saving connections are suitable. Each characteristic can be encoded using Boolean or hierarchical flags and delivered to terminals via multiple channels, such as cellular system broadcasts, attachment signaling, and eSIM configuration updates, to achieve broad compatibility and dynamic updates. The corresponding terminal-side processing mechanism is implemented by an application type sniffing and demand abstraction module embedded in the eSIM terminal system. This module does not require deep packet inspection (DPI) but instead models application behavior based on structural statistical indicators of data packets, including total uplink and downlink traffic per unit time, average packet size, and packet transmission frequency. Key parameters are calculated through sliding statistics within a time window and compared with multiple preset application profile templates to identify the characteristic profiles corresponding to currently active applications. Energy-saving connections refer to the APN's ability to support terminals invoking energy-saving parameters such as long DRX cycles or power saving mode (PSM).

[0029] During the matching stage, the terminal compares and calculates the identified demand feature profile with the received APN capability label, which can be specifically performed through a threshold scoring mechanism: for example, if the demand feature profile reflects high-frequency small data packets and a high uplink ratio, it is more inclined to select the APN marked as uplink bandwidth priority in the label. This matching mechanism is essentially a multi-dimensional feature fitting process, which supports screening out several candidate paths with the best functions from several APNs; further, in order to avoid path unavailability or congestion problems caused by relying solely on static matching, the present invention proposes a micro-timing delay evaluation mechanism, that is, in multiple candidate APNs, the eSIM terminal can initiate minimalist access request signaling in parallel, such as sending a PDU session that does not contain complete user plane information to establish an initial NAS message or RRC connection request, and then capture the time delay from sending to the network side response. ,by The APN corresponding to the smallest value is used as the final connection target. This step effectively eliminates potential high-latency paths and achieves perceptual optimization of the network's real-time status. If the terminal detects a decrease in QoS (Quality of Service) after successful access, and there are other APN capability tags matching the current demand profile in the local cache, the fast switching process can be executed again. QoS indicators may include throughput, RTT, etc. The judgment is based on whether the current observed value deviates from the QoS expectation baseline set by the profile by a certain threshold. This mechanism ensures dynamic stability during the connection phase. In addition, an energy-saving communication strategy mechanism has been introduced: after the terminal establishes a link, based on the combination of the capability tag of the final selected APN and the current application profile, it accesses the preset energy-saving strategy knowledge base, selects the most matching communication parameter configuration, including RRC parameters, DRX cycle or PSM (power saving mode), and adjusts the communication module operating parameters accordingly. This mechanism achieves a real-time compromise between communication efficiency and energy efficiency, reflecting engineering considerations for IoT and small data sensing scenarios; in terms of parameters, for key indicators such as uplink and downlink traffic ratio, average packet sending frequency, and data packet size, the following calculation formula is used: Uplink and downlink ratio = , packet sending frequency = , average packet length = ,in, 、 are the total number of uplink and downlink bytes in the time window, is the number of packets, The duration of the monitoring window; the QoS expectation baseline is derived from the preset profile parameter library in the terminal, and is summarized and set for typical service quality expectations corresponding to different types of applications. For example, for video applications, the QoS expectation baseline can be set to a throughput rate not lower than the specified threshold, while IoT applications mainly focus on the stability of the RTT value. This baseline serves as a reference standard for the terminal to determine quality fluctuations during operation, and supports intelligent adjustment decisions during dynamic connection maintenance; and in the process of extracting application demand feature profiles, modeling is only based on structural statistical features of the network layer / transport layer, such as upstream and downstream traffic, average packet length, and packet sending frequency. It does not involve deep packet inspection DPI, nor does it parse or record any application content data or user identity information, fundamentally avoiding access to user privacy content. In addition, all data statistical processing performed by the application type sniffing and demand abstraction module on the terminal side is completed locally, and does not involve transmitting raw traffic data to the network side or third-party servers. In order to ensure data security, the local module adopts the following strategies in its implementation, such as data minimization: only the necessary statistics within the current time window are retained, and historical detailed data is not stored; local encapsulation processing: all calculations are completed locally on the terminal device to avoid the risk of data leakage; sandbox isolation mechanism: the application identification module and the user business application process run in an isolated environment to prevent unauthorized access; regular data clearing: statistical data is only used for real-time feature matching purposes and is cleared after use; optional user authorization mechanism, if required for product implementation: users can view and decide whether to enable the application feature recognition function in the settings, all of which are extended implementation methods known to ordinary technicians in this field.

[0030] Example 1: This example discloses an access point selection method suitable for an eSIM dynamic switching scenario. The core of the method is to realize a complete APN selection process covering application identification, network capability identification, path optimization and energy efficiency strategy adjustment through the application perception and APN capability matching engine integrated in the eSIM terminal, the coordinated response timing streamlining detection mechanism and the context-driven energy-saving control module. The typical application scenario for this process is: the eSIM terminal runs multiple types of applications concurrently in the mobile communication network, and the network side supports multiple functionally heterogeneous APN configurations at the same time. The terminal needs to establish a data connection quickly, accurately and with sustainable optimization capabilities based on the current actual usage status.

[0031] In this embodiment, the eSIM terminal first activates its internal application type sniffing and demand abstraction module. This module sets a fixed monitoring time window (e.g., ten seconds) and continuously collects IP layer packet feature information within this window, including the total amount of uplink and downlink data transmitted, the total number of packets, the average packet transmission frequency per unit time, and the average packet size in both the uplink and downlink directions. This module does not rely on deep packet inspection, but instead models data flow behavior based on structural statistical features that are independent of the transmission protocol. After completing data collection, the system processes these features in real time to form a set of three characteristic indicators: packet transmission frequency per unit time, uplink and downlink data transmission ratio, and average uplink and downlink packet length. These three indicators constitute a traffic characteristic profile of the currently active application, which is then compared with a typical application profile template preset in the terminal's internal database. For example, if the profile of a video playback application set in the template is characterized by a high proportion of downlink data, a large average packet length, and a medium packet transmission frequency, then if all three indicators fall within the parameter tolerance range set by the template, the system determines that the profile matches and further identifies the type of the currently running application. At the same time, the terminal obtains multiple APN capability tags defined on the network side through any one of three methods, including: 1. Receiving the APN capability tag from the system broadcast information during the attachment process or registration process; 2. Parsing the capability tag content of each APN from the carried extended information field in the tracking area update or routing area update process; 3. During the eSIM profile update, the APN tag update data is obtained together with the profile; the APN capability tag is formulated by the network operator and uniformly adopts a structured description format to characterize whether the access point has one or more service capabilities such as low latency optimization, uplink bandwidth priority, downlink stable transmission, and energy-saving connection.

[0032] Next, the eSIM terminal activates the APN capability adaptation module, which fits and prioritizes the multiple APN capability tags obtained one by one based on the previously extracted application demand feature portrait. For example, if the current application demand feature portrait shows a high-frequency uplink small data packet type, the APN with the uplink bandwidth priority capability identifier will get a higher score; if the current portrait is background intermittent data synchronization, the system will give priority to APNs with energy-saving connection capabilities. After scoring and sorting by the module, the top two APNs are selected as candidate access points. In order to further avoid the problem of poor response performance of the selected candidate APN path, the system then starts the response agility evaluation sub-process; in this process, the terminal does not establish a complete PDU session, but initiates an initial access request signaling to each candidate APN separately, for example, for a network that does not carry a complete PDU session, the system will send an initial access request signaling to the network. The terminal receives a connection request message with user-side parameters, and then measures the time difference between the issuance of the signaling and the receipt of the response from the network side, that is, the micro-timing delay value, and determines the candidate APN with the smallest delay as the final access object, and then establishes a formal data connection through the APN; after completing the data connection, the terminal enters the running state service quality perception and connection stability maintenance mechanism. Under this mechanism, the application type sniffing module continuously monitors the service quality indicators of the application carried by the connection, including throughput and round-trip time. The system compares and judges the above monitoring indicators with the service quality expectation baseline bound to the application demand feature portrait; if the detection indicator is lower than the baseline and exceeds the set deviation threshold, and there are other optional APN tags matching the portrait in the terminal's local cache, the fast switching process is immediately started to reselect and connect to another APN to maintain the stability of the overall service quality.

[0033] In addition, in order to optimize the energy efficiency during the terminal communication process, the system further introduces an energy-saving strategy knowledge base and context joint configuration module. This module takes the currently connected APN capability tag and the currently identified application requirement feature profile as joint context input, and retrieves the most suitable communication parameter configuration scheme from the energy-saving strategy knowledge base preset in the terminal; if the currently connected APN has energy-saving connection capability and the application requirement feature profile is background synchronization type, the system will find the recommended configuration to extend the discontinuous reception period and enable the power saving mode. The terminal communication module will adjust the wireless resource control parameters, the discontinuous reception period duration or the activation conditions of the power saving mode accordingly to achieve energy consumption control optimization under the premise of service quality assurance, which is especially suitable for low-frequency communication, small data sensing and other scenarios.

[0034] Example 2: With the increasing complexity of current mobile communication networks and the widespread deployment of eSIM terminals, terminal devices face the challenge of efficiently and accurately selecting access point names (APNs) in diverse application scenarios. Traditional APN selection mechanisms are often based on static presets or operator policies and lack the ability to perceive the immediate needs of upper-layer terminal applications and the instantaneous state of the network. This leads to insufficient matching between APN selection and application characteristics in scenarios with multiple services running concurrently, which in turn affects user experience and network resource utilization efficiency. For example, high-definition video streaming has high bandwidth and latency requirements, while Internet of Things (IoT) devices typically require low power consumption and intermittent connections. This experiment aims to build a test platform that simulates real-world application scenarios and focus on verifying the proposed APN dynamic switching optimization method. Specifically, by extracting application demand feature profiles on the eSIM terminal side, obtaining and intelligently matching APN capability tags on the network side, and implementing an optimal path selection mechanism based on micro-timing delay evaluation, the proposed method significantly improves APN delivery efficiency, optimizes connection agility, and ultimately improves terminal user experience and energy efficiency.

[0035] In this experiment, we built a comprehensive test platform consisting of an eSIM terminal simulator, a core network function simulator (including AMF, SMF, UPF and some UDM functions) and multiple simulated base stations. The platform can simulate a network environment with multiple APN configurations and support eSIM terminals to obtain dynamic capability tags and interact with signaling. First, we used an eSIM terminal simulator deeply customized based on a mainstream mobile operating system. The simulator has built-in application type sniffing and demand abstraction modules, APN capability adaptation modules, response agility evaluation modules, and energy-saving strategy knowledge base and context joint configuration modules. The application type sniffing and demand abstraction module sets a configurable monitoring time window with a typical value of 5 seconds. Without deep packet inspection (DPI), it constructs an application demand feature portrait by real-time statistics of the structural indicators of the transport layer data packets. These indicators include: total uplink and downlink traffic ( , ) The total number of bytes transmitted in the upstream and downstream directions within the monitoring window; the total number of data packets ( ) The total number of data packets transmitted within the monitoring window; the packet transmission frequency: / (Monitoring window duration); average packet length / Through these statistical features, the simulator can dynamically identify and match three typical application demand feature profile templates: video flow type is typically characterized by a high proportion of downlink traffic ( Much greater than ), the average packet length is large, and the packet sending frequency is medium to high; the IoT intermittent type is typically characterized by small data packets, low or intermittent packet sending, and small upstream and downstream traffic; the interactive type (such as online games / instant messaging) is typically characterized by high-frequency small data packets, relatively balanced upstream and downstream traffic, and sensitive to latency; the APN capability adaptation module is responsible for receiving and parsing the APN capability label sent by the network side, and based on the current application demand feature profile, prioritizes the available APNs based on the preset matching algorithm (using a weighted scoring mechanism); when there are multiple candidate APNs, the response agility evaluation module measures the micro-timing delay (from the time of sending to the time of receiving the first clear response from the network side) by initiating a simplified initial access signaling interaction (a PDU session establishment request NAS message that does not contain complete user plane parameters). ), and selects the APN with the smallest delay; the energy-saving strategy knowledge base and context joint configuration module stores energy-saving communication parameter configuration rules under different combinations of APN capability tags and application requirement feature profiles, which are used to dynamically adjust the terminal's radio resource control (RRC) parameters, discontinuous reception (DRX) cycle parameters or power saving mode (PSM) configuration.

[0036] We configured three virtual APNs in the core network function simulator and defined standardized capability tags for them. These tags were sent to the eSIM terminal simulator through the simulated cellular network system information broadcast mechanism: the APN-A capability tag is optimized for low latency and prioritizes uplink bandwidth; the APN-B capability tag is for energy-saving connection and stable downlink transmission; and the APN-C capability tag is for basic Best Effort service. This experiment mainly focuses on the following performance indicators: APN selection accuracy is to evaluate the degree of match between the APN selected by the terminal and the actual application demand profile; connection establishment delay is the end-to-end delay from the terminal initiating the APN connection request to the successful establishment of the data path, focusing on micro-timing delay ( ) optimization effect; energy efficiency is manifested as the average power consumption of the terminal in different application scenarios; Quality of Service (QoS) maintenance capability: after the connection is established, the terminal continuously monitors the throughput and round-trip time (RTT) and has adaptive switching capabilities. All delay measurements are synchronized with high-precision timestamps and averaged over multiple repeated measurements to eliminate accidental errors. Power consumption measurements are sampled in real time by a high-precision current sensor integrated in the terminal simulator.

[0037] This experiment designed three groups of typical scenarios. Each group of scenarios was repeated 20 times in a controlled environment, and the average values ​​of various key indicators were recorded. Video streaming application connection: The application background is to simulate users watching high-definition online videos on eSIM terminals. This application has high requirements for downlink bandwidth and latency; experimental steps: the eSIM terminal simulator is started, and the video streaming application is started; the application type sniffing and demand abstraction module monitors the traffic characteristics of the video stream and extracts the application demand feature portrait; the APN capability adaptation module obtains the APN capability label sent by the network side, and performs matching and priority sorting; the response agility evaluation module performs micro-timing delay tests on the top two candidate APNs; the terminal selects the optimal APN and establishes a data connection to start video playback; continuously monitors QoS indicators (throughput, RTT); compares the performance differences between traditional static APN selection (default is APN-C) and the method of the present invention. Observation and data recording: After the video streaming application is started, the application type sniffing and demand abstraction module identifies the following traffic characteristics: the total downlink traffic is significantly higher than the uplink (average / The ratio is about 9.5:1), the average packet length is about 1450 bytes, and the packet sending frequency is about 80pps (packets per second). These characteristics are highly consistent with the preset video stream profile. In the matching stage, the APN capability adaptation module prioritizes APN-B (downlink stable transmission) as the highest, followed by APN-A (low latency optimization), and APN-C as the lowest. In the micro-timing delay evaluation, the average of APN-B is The average latency of APN-A is 28.1ms. The time taken is 32.5ms. As APN-B performs well in both capability matching and response agility, the terminal finally selects APN-B to establish a data connection. The performance comparison data between the method of the present invention and the traditional static APN selection in the video streaming application connection scenario are shown in Table 1.

[0038] Table 1: Comparison of connection establishment performance in scenario 1.

[0039]

[0040] The method of the present invention reduces the connection establishment delay by about 50%, improves the initial throughput by 126%, and reduces the RTT by 54% by intelligently matching the application profile with the APN capability tag and evaluating the micro-timing delay. This shows that the dynamic selection mechanism can significantly optimize the connection performance in the video streaming scenario.

[0041] IoT intermittent application connection: The application background is to simulate the periodic uploading of small data packets by smart city sensors (such as environmental monitoring nodes). This application is sensitive to power consumption and has relatively loose requirements on latency, but it is necessary to ensure reliable data transmission; Experimental steps: The eSIM terminal simulator is started, and the IoT data reporting application is started (reporting once every 1 minute, with a data volume of about 100 bytes each time); the application type sniffing and demand abstraction module monitors traffic characteristics and extracts application demand feature portraits; the APN capability adaptation module performs matching and priority sorting; the terminal selects the optimal APN and establishes a data connection; the terminal power consumption is continuously monitored; the energy efficiency performance of the traditional static APN selection and the method of the present invention are compared; Observation and data recording: After the IoT application is started, the application type sniffing and demand abstraction module The following traffic characteristics were identified: small data packets (average packet length approximately 100 bytes), extremely low and intermittent packet transmission frequency, and very small total uplink and downlink traffic. These characteristics highly match the preset intermittent IoT profile. During the matching phase, the APN capability adaptation module, based on the matching algorithm, prioritizes APN-B (energy-saving connection) the highest. The terminal ultimately selects APN-B to establish a data connection. After the connection is established, the energy-saving policy knowledge base and context joint configuration module adjusts the communication module's DRX cycle to a longer value (for example, from the default 320ms to 2560ms) based on APN-B's energy-saving connection capability tag and the IoT intermittent application profile, and enables PSM. The energy efficiency performance comparison results for IoT intermittent application connections are shown in Table 2.

[0042] Table 2: Comparison of energy efficiency performance in scenario 2.

[0043]

[0044] In IoT scenarios, the method of the present invention reduces the average power consumption of terminals by approximately 75% through intelligent matching and energy-saving strategy adjustment, while maintaining the reliability of data transmission. This fully verifies the significant effect of the method in optimizing terminal energy consumption, which is crucial for battery-powered IoT devices. The connection establishment delay is also synchronously optimized.

[0045] Multi-service concurrency and QoS adaptive maintenance: The application background is to simulate a terminal running an online game (interactive) and a background data synchronization application (intermittent IoT) simultaneously. Online games have extremely high requirements for latency and response agility, while background data synchronization requires stable transmission and energy efficiency. This scenario focuses on verifying the initial selection of APNs under multiple services and the adaptive switching capabilities when QoS degrades. The test steps are: the eSIM terminal simulator simultaneously starts the online game and the background data synchronization application; the application type sniffing and demand abstraction module identifies the current main active traffic characteristics (mainly dominated by online game traffic at this time). Extract the demand characteristics of interactive applications; the APN capability adaptation module and the response agility assessment module collaborate to select the optimal APN and establish the initial data connection; simulate sudden congestion on the network side (for example, by introducing an additional end-to-end delay of 100ms), resulting in QoS degradation of the current connection; the application type sniffing and demand abstraction module continuously monitors QoS indicators. If it is lower than the expected baseline and reaches the predetermined deviation threshold (this experiment is set to RTT deviation of more than 20ms), and there is a matching alternative APN in the local cache, the active switching process is triggered; record QoS changes and the switching process; observation and data recording: In the early stage of multi-service concurrency, online gaming traffic characteristics (high frequency of small data packets, balanced uplink and downlink traffic, and sensitivity to latency) were identified as the main application profile. After matching and evaluation, APN-A (low latency optimization, uplink bandwidth priority) was selected as the best, with a micro-timing delay of 26.7ms. The terminal successfully connected to APN-A with an initial RTT average of 35.1ms, providing a smooth online gaming experience. At the 5th minute of the experiment, we simulated the network side introducing additional delay to the APN-A path, causing its RTT to rise rapidly to an average of 75.3ms. The QoS monitoring module of the eSIM terminal identified R The RTT deviated from the expected QoS baseline (average RTT no higher than 45ms) set for the interactive application profile and exceeded the 20ms deviation threshold. At this point, an alternative APN (APN-B) existed in the terminal's local cache. Although its capability tag favored energy conservation, its low-latency performance was suboptimal and unaffected by the current congestion. The terminal proactively initiated a switch, completing the switch to APN-B within approximately 2 seconds. After the switch, the online game RTT returned to an average of 40.5ms, effectively maintaining service smoothness. Table 3 details the performance data for QoS adaptive maintenance in a multi-service concurrent scenario.

[0046] Table 3: QoS adaptive maintenance performance in scenario three.

[0047]

[0048] This scenario verifies the robustness of the method of the present invention in complex multi-service scenarios. It can not only accurately select the initial APN based on the dominant application type, but also quickly identify and actively switch to an alternative APN when the network status deteriorates and causes QoS degradation, thereby effectively maintaining the continuity and stability of the user experience.

[0049] This series of experimental data fully and powerfully demonstrates the practical effectiveness and significant advantages of the APN delivery efficiency optimization method for eSIM dynamic switching proposed in the present invention. The accuracy of APN selection is greatly improved: through the deep perception of upper-layer application traffic characteristics and the extraction of demand portraits on the eSIM terminal side, combined with the standardized description of the APN capability label on the network side, the terminal can achieve refined matching of application requirements and APN service capabilities, which is significantly different from the traditional static or blind allocation method, ensuring that the network access point is highly consistent with the application characteristics from the beginning of the connection establishment; the agility of connection establishment is significantly optimized: the introduction of the micro-timing delay evaluation mechanism enables the terminal to further evaluate and select the network path with the fastest instantaneous response among multiple matching candidate APNs, which effectively avoids potential high-latency or congested paths, greatly shortens the connection establishment delay, and improves the user experience of the first access; the energy efficiency of the terminal is effectively improved: based on the dual context information of the APN capability label and the application demand feature portrait, the terminal can intelligently query and determine the most matching energy-saving communication parameters. Number configuration (such as DRX cycle and PSM), which enables the terminal's energy-saving behavior to be refined and dynamically adapted according to the current application scenario and the characteristics of the connected APN, which is especially suitable for IoT applications that are sensitive to power consumption, and significantly extends the terminal's battery life; Enhanced service quality maintenance capability: This method constructs a continuously optimized closed-loop system. After the data connection is established, the eSIM terminal will continuously monitor the application service quality indicators. If the monitored QoS indicators are lower than the expected baseline and reach the preset threshold, the terminal can actively try to switch to other alternative APNs that match the profile, effectively responding to network fluctuations or changes in application requirements, and improving the connection resilience and the stability of user experience; In summary, the technical solution of the present invention constructs an efficient and adaptive APN dynamic selection and optimization system through the organic integration of terminal application perception, APN capability identification, path agility evaluation and intelligent energy-saving strategies.

[0050] Example 3: This example combines Figures 1 to 3 , a method for optimizing the efficiency of APN delivery for dynamic eSIM switching is described. Figure 1As shown, the left side of the figure is scenario 1: comparison of video streaming application connection performance. The horizontal axis is marked with performance indicators, including three indicators: initial average RTT (ms), initial average throughput (Mbps) and average connection establishment delay (ms). The vertical axis is the performance indicator value. The light gray column in the figure represents the traditional static selection method, and the dark gray column represents the method of the present invention. In the initial average RTT (ms) indicator, the traditional static method is about 65ms, while the method of the present invention is reduced to about 30ms; in the initial average throughput (Mbps) indicator, the traditional static method is about 12.5Mbps, while the method of the present invention is increased to about 28.3Mbps; in the average connection establishment delay In terms of time (ms), the traditional static method is about 85ms, while the method of the present invention is shortened to about 42ms; the right side of the figure is scenario 2: IoT intermittent application energy efficiency comparison, the horizontal axis is performance indicators, including average power consumption per hour (mW), and the vertical axis is power consumption value (mW). The power consumption value of the traditional static method is about 185mW, while the method of the present invention is reduced to about 45mW. Through the joint matching of the energy-saving connection characteristics in the APN capability tag and the intermittent small data packet characteristics of the application portrait, the terminal adjusts the communication parameters according to the policy knowledge base, such as extending the DRX cycle and enabling PSM, which significantly reduces the average power consumption (mW) per hour, thereby enhancing the energy efficiency and battery life of the terminal device.

[0051] like Figure 2 As shown, first, the functional module in the eSIM terminal starts application type sniffing and demand abstraction, which is used to obtain and analyze the characteristic information of the currently running application to form an application demand characteristic portrait. The portrait characterizes typical application types such as video streaming, IoT and interactive by analyzing indicators including uplink and downlink traffic ratio, average packet length and packet sending frequency. At the same time, the eSIM terminal receives information from the APN capability label on the network side. The label is obtained through a downward method such as system broadcast attachment signaling or eSIM configuration. Its core features include low latency optimization, uplink bandwidth priority and energy-saving connection. The eSIM terminal matches the application demand characteristic portrait with the network side APN capability label according to the intelligent matching mechanism, and completes the candidate APN selection based on the multi-dimensional feature fitting method in the matching mechanism. After that, the eSIM terminal enters the signaling test phase, performs micro-timing delay evaluation, sends test signaling by simplifying NAS / RRC requests, collects and evaluates indicators, and performs the following operations: Minimization is performed to determine the response agility of candidate paths. Based on the optimal selection, the final APN connection is performed. This step is based on the selection criteria of capability matching + response agility to ensure the optimal data link connection. Subsequently, the dynamic configuration process is entered to adjust the energy-saving strategy. The optimization parameters include DRX cycle / PSM configuration, so that the terminal can achieve energy-saving communication effects while ensuring service quality.

[0052] like Figure 3 As shown, the process starts at the eSIM terminal. On the one hand, the terminal obtains the capability tag and obtains relevant information from the APN capability tag on the network side. The capability tag includes three core features: low latency optimization, uplink bandwidth priority, and energy-saving connection. On the other hand, the terminal analyzes the traffic characteristics of the running applications and extracts the application demand feature profile, which covers three typical features: video streaming characteristics, IoT intermittent characteristics, and interactive characteristics. Based on the above two inputs, the terminal completes the candidate APN selection through the intelligent matching mechanism and enters the micro-timing delay evaluation stage through signaling interaction testing. The system judges the response agility of the candidate path based on the test feedback, and then executes the optimal path and completes the final APN connection. The connection selection is based on the dual standards of capability matching and response agility. After the connection is established, the system starts the energy-saving strategy adjustment module based on the dynamic parameter configuration mechanism to further optimize parameters such as the DRX cycle / PSM configuration.

[0053] Example 4: In a typical mobile communication application scenario with multiple concurrent services, the eSIM terminal is deployed in a city bus and operates as a vehicle communication control unit. The terminal simultaneously runs multiple upper-layer application modules, including: a high-frequency position reporting module for real-time trajectory positioning and scheduling management, an on-board video backhaul module for uploading in-vehicle security monitoring data, and a background maintenance synchronization module for system status updates and log synchronization. The above applications have significant differences in data packet structure, transmission frequency, bandwidth occupancy, and delay tolerance, and therefore put forward higher adaptability requirements for the selection of access point names. In this scenario, the eSIM terminal first activates its internally configured application type sniffing and demand abstraction module. This module uses a fixed five-second monitoring time window to continuously perform structural statistics on IP layer data transmission. During a typical operating cycle, the monitored data characteristics are characterized by a large number of packets sent per unit time, a small average packet size, and a nearly equal total amount of uplink and downlink data. Based on this, the system identifies this feature combination as a high-frequency and bidirectional symmetric interactive service, and generates a corresponding application demand feature profile, marked as low latency, high frequency, and bidirectional. At the same time, the eSIM terminal has obtained three sets of APN capability tag information from the network side in the previously completed tracking area update process, corresponding to three optional access points. These capability tags are configured by the operator in a structured manner and embedded in the extension field of the NAS signaling for distribution. After receiving them, the terminal parses and caches them in key-value pair format. These three sets of capability labels are respectively identified as: 1. Support for low-latency optimization and uplink bandwidth priority; 2. Support for energy-saving connections and stable downlink transmission; 3. Possessing basic Best Effort service capabilities; among them, basic Best Effort service indicates that the APN does not have specific optimization features, follows the general service allocation strategy, and is suitable for non-critical business loads.

[0054] The eSIM terminal then activates its APN capability matching calculation module, performing a dimensional fit between the aforementioned capability tags and the currently identified application requirement profile. The match is assessed based on a locally pre-defined weighted scoring mechanism. In this scenario, the low latency optimization and uplink bandwidth priority characteristics of the first capability tag strongly match the transmission characteristics required for high-frequency location reporting. Therefore, the system determines it as the highest-priority candidate access point at the current moment. The remaining two APNs are selected as alternative paths in order. To verify the network responsiveness of the prioritized candidate APN at the current moment, the terminal initiates a simplified access signaling interaction with it. This is achieved by constructing and sending a PDU Session Establishment Request message containing only the minimum necessary parameters, avoiding the full user plane configuration. This is intended to assess the immediate agility of the network response. After the signaling is sent, the terminal starts a microsecond timer and records the time difference between the signaling and the receipt of the first clear response from the network. The measured micro-sequence delay is 27 milliseconds, which is within the acceptable low latency range set by the local system. Based on this, the terminal determines this APN as the current final access target and immediately establishes a formal data connection through this path.

[0055] After the data connection is established, the energy-saving policy matching module within the eSIM terminal is triggered. It uses the APN capability tag of the currently established connection and the previously extracted application requirement feature profile as context information input, and accesses the terminal's built-in energy-saving communication policy knowledge base. This knowledge base forms a mapping relationship based on context combinations and predefined rules. The rule conditions include the capability tag feature combination, the application profile category, and the operating environment context (such as the current battery level, system load status, etc.). In the scenario of this embodiment, because the connected APN has low latency optimization characteristics and the current application profile is a high-frequency interactive service, the system recommends, after knowledge base matching, that the policy is to disable the power saving mode and adjust the discontinuous reception cycle parameter to a medium range. This configuration takes into account the agility of service transmission and energy efficiency control during terminal communication. The optimization result is achieved by adjusting the radio resource control parameters of the communication module, specifically by extending the cell release waiting time and keeping the RRC connection activated to enhance the processing capability of burst uplink data requests and service continuity. During operation after the formal connection is established, the terminal continuously monitors the service quality indicators of the current business data transmission through the application type sniffing module, including the average round-trip delay and throughput per unit time, and calculates their short-term averages using a sliding window method. If any monitored quality indicator deviates from the performance baseline set by the current application profile by more than a 25% deviation threshold, the system triggers a service quality degradation warning mechanism and automatically queries the local cache for other APN capability tags that closely match the application profile. If so, the path switching process is restarted to maintain stable connection service quality. It should be noted that the setting of each key parameter and its adjustment logic during implementation are based on engineering principles widely verified in actual application environments. For example, the sensitive threshold range for micro-timing delay is set to 20 to 30 milliseconds. This range is derived from statistical analysis of terminal access data in a large number of urban public transportation scenarios and can more accurately distinguish the response performance differences between acceptable paths and congested paths. The medium range setting of the non-continuous reception period parameter refers to the battery decay curve and connection uninterrupted rate evaluation results of the same device in different application modes. These are all extended implementation methods known to ordinary technicians in this field.

[0056] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing APN delivery efficiency for dynamic eSIM switching, characterized in that: The method comprises the following steps: Step 1: The eSIM terminal obtains the APN capability tag provided by the network side. The APN capability tag describes the service capability characteristics of each available access point name (APN). Step 2: The eSIM terminal analyzes the traffic statistics generated by the upper-layer application within the specified monitoring time window to extract a demand feature profile of the currently running application. The traffic statistics include one or more of the following: average packet size, uplink / downlink traffic ratio, and packet transmission frequency per unit time. Step 3: The eSIM terminal performs a matching calculation based on the extracted demand feature profile and the obtained APN capability tag to select one or more candidate APNs that match the demand feature profile; and the eSIM terminal establishes a data connection through a final APN determined from the one or more candidate APNs. And after the eSIM terminal selects one or more candidate APNs that match the demand feature profile, and before establishing a data connection through the final APN, the following steps are also included: Step 1, if it is determined that there are at least two candidate APNs, the eSIM terminal initiates an initial access signaling interaction for each of the at least two candidate APNs, and the initial access signaling interaction is intended to obtain the instantaneous response status of its corresponding network path; Step 2, the eSIM terminal captures and compares the micro-timing delay of each initial access signaling interaction from the time it is issued to the time it receives the first clear response signaling from the network side ; and the eSIM terminal selects micro timing delay The candidate APN with the smallest value is used as the final APN, where the micro-timing delay It indicates the response agility of the corresponding candidate APN network path; and after the eSIM terminal establishes a data connection through the final APN, it also includes the following steps: the application type sniffing and demand abstraction module in the eSIM terminal continuously monitors the application service quality QoS indicators through the current data connection, and the QoS indicators include the throughput perceived by the eSIM terminal or the round-trip time RTT estimated through application layer feedback information; if the monitored QoS indicator is lower than the QoS expectation baseline associated with the current demand feature profile by a predetermined deviation threshold, and there are other alternative APNs whose APN capability tags match the demand feature profile in the local cache, the eSIM terminal actively attempts to switch to one of the alternative APNs.

2. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 1, characterized in that: The APN capability tag is defined and configured on the network side by the network operator, and is sent to the eSIM terminal through the cellular network system information broadcast, or sent to the eSIM terminal through the extended information it carries when the eSIM terminal performs the attachment process, tracking area update TAU process or routing area update RAU process, or sent together when the eSIM configuration file is downloaded and updated.

3. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 2, characterized in that: The eSIM terminal extracts a demand feature profile of the currently running application by analyzing traffic statistical features generated by the upper-layer application within a determined monitoring time window, specifically comprising: without performing deep packet inspection, the application type sniffing and demand abstraction module within the eSIM terminal monitors and counts the total uplink and downlink traffic, the total number of data packets, the average data packet size of uplink data transmission, and the average data packet size of downlink data transmission within the determined monitoring time window; Calculate the uplink and downlink traffic ratio and the average packet sending frequency per unit time based on the statistical results; The calculated uplink and downlink traffic ratio, the average packet sending frequency per unit time, and the average data packet size of uplink and downlink data transmission are compared with the characteristic parameter thresholds corresponding to different application demand feature profile templates stored in the eSIM terminal to determine the demand feature profile that matches the currently running application.

4. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 1, characterized in that: The initial access signaling interaction is a PDU session establishment request NAS message that does not contain complete user plane parameters, or the initial signaling of the DNN selection process for different core network data network names when the RRC layer establishes a connection.

5. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 1, characterized in that: After the eSIM terminal establishes a data connection through the final APN, the following steps are also included: the eSIM terminal profiles the dual context information based on the APN capability tag of the final APN and the previously extracted demand characteristics of the currently running application; queries and determines a set of energy-saving communication parameter configurations that match the current dual context information combination from an energy-saving strategy knowledge base preset in the eSIM terminal, where the energy-saving strategy knowledge base defines energy-saving communication parameter adjustment rules corresponding to different context combinations; the communication module of the eSIM terminal adjusts at least one of its current radio resource control RRC parameters, discontinuous reception DRX cycle parameters, or power saving mode PSM configuration parameters based on the determined energy-saving communication parameter configuration to optimize energy consumption when performing data communication through the final APN.

6. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 5, characterized in that: The energy-saving communication parameter adjustment rules defined in the energy-saving strategy knowledge base include: if the APN capability tag is energy-saving connection and the application demand characteristic profile is small data intermittent type, the communication module is configured to use a DRX cycle of 2560ms and enable PSM; or if the APN capability tag is low latency optimization and the application demand characteristic profile is changed to background data synchronization, the communication module is configured to extend the DRX cycle.

7. The method for optimizing APN delivery efficiency for dynamic eSIM switching according to claim 1, characterized in that: The APN capability tag is a set of standardized descriptors used to characterize the core service features of the APN. The core service features are selected from one or more of the following: low latency optimization, uplink bandwidth priority, downlink stable transmission, energy-saving connection, and basic Best Effort service.

Images