Remote configuration management system and method based on embedded sim specification

Abstract

This application provides a remote configuration management system and method based on the embedded SIM specification, relating to the field of mobile communication technology. The method acquires regional change information and network access state sequences of the target terminal; generates candidate handover prompts based on the regional change information and preset configuration rules; extracts local state fragments corresponding to candidate segments from the network access state sequences to determine the handover stability of each candidate configuration file; determines the active candidate configuration file based on the handover stability and updates the migration trigger conditions; and, in response to the current active configuration file meeting the updated trigger conditions, controls the target terminal to migrate to the active candidate configuration file. This application introduces network access state into configuration migration decisions and dynamically updates trigger conditions based on handover stability, reducing false triggering and repeated migrations caused by short-term fluctuations in network access state, and improving the accuracy of migration timing determination.

Description

Technical Field

[0001] This application relates to the field of mobile communication technology, and more specifically, to a remote configuration management system and method based on the embedded SIM specification. Background Technology

[0002] With the development of embedded SIM technology, terminal devices can download operator configuration files, activate corresponding communication services, and manage different configuration files remotely without replacing the physical SIM card. Existing standards have established a corresponding architectural foundation and configuration file management mechanism for remote configuration and management of embedded SIMs, enabling terminals to perform subscription changes and configuration adjustments even after deployment.

[0003] In existing remote configuration management solutions, candidate configuration files are typically determined based on the terminal's location, preset carrier rules, or preset handover conditions. Download, activation, or handover operations are then performed when these conditions are met. However, during continuous terminal movement, especially with frequent regional changes, network access status is easily affected by localized coverage overlap, short-term attachment fluctuations, or instantaneous access changes. If configuration migration is still performed directly based solely on relatively static regional rules or preset trigger conditions, situations such as false configuration migration triggers, short-term re-switching after migration, or repeated migrations between multiple candidate configuration files can easily occur during regional changes, thus affecting the stability of remote configuration management and the accuracy of configuration migration timing determination. Therefore, there is an urgent need for a remote configuration management method that can reduce the incidence of false configuration migration triggers and repeated migrations caused by short-term fluctuations in network access status during terminal regional changes, and improve the accuracy of configuration migration timing determination. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a remote configuration management system and method based on the embedded SIM specification.

[0005] Firstly, this application provides a remote configuration management method based on the embedded SIM specification, including:

[0006] Acquire regional change information and network access status sequence of the target terminal within a preset time period;

[0007] Based on the regional change information and preset configuration rules, at least one candidate switching prompt is generated. The candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file.

[0008] Based on the candidate handover prompts, local state segments corresponding to each candidate segment are extracted from the network access state sequence, and the handover stability corresponding to each candidate configuration file is determined based on the local state segments.

[0009] Based on the switching stability described above, an active candidate configuration file is determined, and the triggering conditions for configuration state migration are updated based on the active candidate configuration file.

[0010] In response to the current active profile meeting the updated triggering conditions, the target terminal is controlled to migrate from the current active profile to the active candidate profile.

[0011] Optionally, generating at least one candidate switching suggestion includes:

[0012] Based on the regional change information, determine the regional change trajectory and extract the adjacent segments of the regional change trajectory.

[0013] Based on the temporal position of each of the attribution change adjacent segments in the regional change trajectory and the preset configuration rules, candidate configuration files associated with each of the attribution change adjacent segments are determined, and candidate switching prompts are generated for each of the candidate configuration files.

[0014] Optionally, determining the switching stability corresponding to each candidate configuration file includes:

[0015] Taking the timing position corresponding to each candidate switching prompt as the center, extract the preceding state segment and the following state segment from the network access state sequence;

[0016] Determine the continuity of the access identifier and the consistency of the attachment result between the preceding state segment and the following state segment;

[0017] Based on the continuity of the access identifier and the consistency of the attachment result, the handover stability of the corresponding candidate configuration file is determined.

[0018] Optionally, determining the active candidate profile includes:

[0019] Determine the switching stability of each candidate configuration file at at least two different time scales;

[0020] Active candidate configuration files are determined based on the cross-scale consistency of switching stability across different time scales.

[0021] Optionally, the migration to the active candidate profile includes:

[0022] In response to the fact that the active candidate configuration file is not stored on the target terminal, the target terminal is controlled to download the active candidate configuration file;

[0023] After the active candidate configuration file is downloaded, the activation of the active candidate configuration file is delayed, and the network access status sequence corresponding to the target terminal is obtained again.

[0024] The handover stability corresponding to the active candidate configuration file is updated based on the continuously acquired network access state sequence;

[0025] In response to the updated switching stability continuously meeting the updated triggering conditions, the target terminal is controlled to activate the active candidate configuration file.

[0026] Optionally, the migration to the active candidate profile includes:

[0027] After the active candidate profile is activated, the currently active profile is kept in an inactive reserved state.

[0028] Optionally, the triggering conditions for the updated configuration state migration include:

[0029] The switching trigger timing and the original configuration retention duration in the triggering conditions are adjusted according to the cross-scale consistency.

[0030] Optionally, the method further includes:

[0031] Obtain the network access state sequence corresponding to the active candidate configuration file and extract it as a subsequent state segment;

[0032] Determine the consistency of the attachment results and the continuity of the access identifiers corresponding to the subsequent state segments;

[0033] In response to the continued satisfaction of the preset rollback release conditions for the consistency of the attachment results and the continuity of the access identifier, the current active configuration file is adjusted from the inactive retained state to the non-rollback state.

[0034] Optional, also includes:

[0035] Based on the regional change information, determine whether there are queuing passage sections in the regional change trajectories corresponding to the preceding state segment and / or the following state segment where continuous dwell state and short-distance advance state alternate;

[0036] In response to the existence of the queuing passage section, short-period repetitive attachment records in the preceding state segment and the following state segment, where the position change between adjacent successful attachment events is less than a preset advancement threshold and the duration of successful attachment is less than a preset stable duration, are identified and marked as pseudo-stable records.

[0037] Optionally, determining the switching stability of the corresponding candidate configuration file includes:

[0038] The contribution weight of the state sample corresponding to the pseudo-stable record in calculating the continuity of the access identifier and the consistency of the attachment result is reduced to obtain the corrected continuity of the access identifier and the corrected consistency of the attachment result.

[0039] Based on the continuity of the corrected access identifier and the consistency of the corrected attachment result, the handover stability of the corresponding candidate configuration file is determined.

[0040] Secondly, this application provides a remote configuration management system based on the embedded SIM specification, including:

[0041] The acquisition module is used to obtain information on regional changes and network access status sequences of the target terminal within a preset time period;

[0042] The generation module is used to generate at least one candidate switching prompt based on the regional change information and preset configuration rules. The candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file.

[0043] The processing module is configured to extract local state segments corresponding to each candidate segment from the network access state sequence according to the candidate handover prompts, and determine the handover stability corresponding to each candidate configuration file based on the local state segments; determine the active candidate configuration file based on each handover stability, and update the triggering conditions for configuration state migration based on the active candidate configuration file;

[0044] The control module is used to control the target terminal to migrate from the current active configuration file to the active candidate configuration file in response to the updated triggering condition of the current active configuration file.

[0045] Compared with existing technologies, this application, based on obtaining the target terminal's regional change information, further obtains the network access state sequence and generates candidate handover prompts and determines handover stability based on the correlation analysis between the two, thereby incorporating the actual performance of the network access state into the decision-making basis for configuration migration. By filtering active candidate configuration files based on handover stability and dynamically updating the migration triggering conditions accordingly, the triggering of configuration migration no longer depends solely on crossing regional boundaries, but comprehensively considers whether the network access conditions experienced by the terminal within the candidate handover segment meet the conditions for stable migration. Therefore, when the terminal is in a short-term fluctuation phase of network access state, its corresponding handover stability is unlikely to meet the updated triggering conditions, and the configuration migration operation will not be executed, thus reducing the possibility of false triggering. Simultaneously, since the triggering conditions are dynamically adjusted with changes in handover stability, when the terminal repeatedly experiences short-term changes in regional affiliation in the boundary area, the triggering conditions can adaptively tighten, suppressing repeated migrations.

[0046] Overall, this application improves the accuracy of determining the timing of configuration file migration in cross-regional scenarios by introducing a stability assessment based on network access status and a dynamic update mechanism for triggering conditions. This helps reduce the risk of communication interruption caused by improper migration timing and improves the service continuity of terminals during regional changes. Attached Figure Description

[0047] Figure 1 A flowchart illustrating a remote configuration management method based on the embedded SIM specification provided in this application embodiment;

[0048] Figure 2 A flowchart of a method for generating candidate switching prompts provided in an embodiment of this application;

[0049] Figure 3 A flowchart illustrating the method for migrating to active candidate profiles provided in this application embodiment;

[0050] Figure 4 This is a schematic diagram of a remote configuration management system based on the embedded SIM specification provided in an embodiment of this application. Detailed Implementation

[0051] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0052] See Figure 1 The diagram shows a flowchart of a remote configuration management method based on the embedded SIM specification provided in an embodiment of this application, including steps S101 to S105, wherein:

[0053] S101: Obtain the regional change information and network access status sequence of the target terminal within a preset time period;

[0054] S102: Based on the regional change information and preset configuration rules, generate at least one candidate switching prompt, wherein the candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file;

[0055] S103: Based on the candidate handover prompt, extract the local state fragments corresponding to each candidate segment from the network access state sequence, and determine the handover stability corresponding to each candidate configuration file based on the local state fragments;

[0056] S104: Based on the switching stability described above, determine the active candidate configuration file, and update the triggering conditions for configuration state migration based on the active candidate configuration file;

[0057] S105: In response to the current active profile meeting the updated triggering condition, control the target terminal to migrate from the current active profile to the active candidate profile.

[0058] Regarding the above S101:

[0059] The target terminal can be a communication terminal equipped with an embedded SIM, such as an in-vehicle terminal, a T-Box, an industrial gateway, or other terminals with cellular network access capabilities. Area change information refers to information characterizing the changes in the spatial location or home area of ​​the target terminal over a preset time period; network access status sequence refers to a sequence of status data recorded in chronological order, characterizing the changes in the network access process of the target terminal. The downloading, enabling, disabling, and switching of the embedded SIM configuration file, as well as its underlying remote configuration interaction process, can be implemented using conventional remote configuration management techniques in this field, and are not elaborated upon here.

[0060] In one embodiment, regional change information can be obtained through a single positioning source or a multi-source fusion method. For example, the latitude and longitude coordinates of the target terminal at different sampling times can be obtained based on global satellite positioning information; or the regional affiliation change information corresponding to the target terminal can be obtained based on cell identifiers, location area identifiers, tracking area identifiers, and base station measurement results; furthermore, the location information or regional affiliation information can be corrected by combining vehicle inertial navigation information, odometer information, or electronic map matching results. In other words, regional change information is not limited to precise coordinates; any information that reflects the spatial advancement process, regional entry process, or regional affiliation change process of the target terminal within a preset time period can be used as regional change information in this application.

[0061] In one embodiment, the network access state sequence can be periodically collected and formed by the terminal communication module, baseband chip, operating system network management module, or access log module of remote management platform. For example, the state items in the network access state sequence may include one or more of the following: attachment result, registration status, access identifier, access network type, reselection result, session establishment status, and data connection maintenance status.

[0062] It should be noted that this application does not require all of the above-mentioned status items to exist simultaneously. As long as the collected status items can reflect the evolution of the target terminal's network access behavior within a preset time period, a network access status sequence required for subsequent analysis can be formed.

[0063] In one embodiment, the preset time period can be set according to the target terminal's moving speed, the density of the area boundary distribution, the degree of network coverage overlap, and the need to counteract short-term fluctuations in subsequent migration determination. If the target terminal moves at a high speed and the area crossing process is short, the preset time period can be relatively short to improve the tracking sensitivity to recent area changes; if the target terminal moves at a low speed, is in a slow-moving or stationary scenario, or there is significant overlapping coverage near the area boundary, the preset time period can be appropriately extended to retain a more complete state evolution context.

[0064] For example, the preset time period can be set from 10 seconds to 300 seconds. For instance, in scenarios with continuous driving and clear area switching, 20 seconds can be used; in scenarios with slow-moving traffic, passing near boundaries, or queuing, 90 seconds or 120 seconds can be used. The above example values ​​correspond to different setting scenarios under different movement rhythms, different coverage complexities, and different anti-fluctuation requirements.

[0065] In one embodiment, when acquiring regional change information and network access state sequences, a state sampling period can be preset. The state sampling period is used to determine the time interval between two adjacent sampling moments within a preset time period. A state sampling period that is too short will significantly increase the amount of sampled data and may introduce a large number of transient jitter states that are of little significance for subsequent migration determination; a state sampling period that is too long may miss key access state inflection points in the process of local regional changes. Therefore, the state sampling period can be set according to the terminal's moving speed, the communication module's reporting capability, and the time resolution required for subsequent stability analysis.

[0066] For example, the state sampling period can be 0.5 seconds to 10 seconds. For instance, when moving at high speed or needing to capture finer-grained access changes, 1 second can be used; when moving at low speed and the state fluctuations are relatively slow, 3 seconds or 5 seconds can be used. Further, in one embodiment, the preset time period and the state sampling period can satisfy a preset sample quantity relationship, for example, so that a preset time period contains 10 to 200 consecutive state samples, in order to balance the ability to characterize state evolution and computational complexity.

[0067] In one specific embodiment, the target terminal can be a T-Box installed in a vehicle traveling across regions. The remote configuration management platform can continuously acquire the latitude and longitude trajectory, cell identifier sequence, attachment result, and data connection status of the T-Box within the last 120 seconds at a sampling period of 1 second, and form regional change information and network access status sequence according to a unified timestamp. When the vehicle is in a continuous high-speed driving state, the preset time period can be shortened to 30 seconds to more sensitively reflect recent regional changes; when the vehicle is in a slow-moving or stopping state or passing through the boundary of a nearby area, the 120-second window can be kept unchanged to retain more sufficient local state context.

[0068] Regarding S102 above:

[0069] The reason for executing this step first, rather than triggering the configuration file switch directly after detecting a region change, is that in embedded SIM remote configuration scenarios, a region change does not equate to the configuration migration conditions being ripe. Especially near region boundaries, in areas with overlapping coverage, or in low-speed, slow-moving traffic scenarios, although the target terminal may show a region advancement trend, the network access status may still be affected by short-term coverage fluctuations, local reselection, or instantaneous attachment changes. If the target configuration file is selected and migration is executed directly based solely on a single region entry event or a single region affiliation change, it is prone to false triggering or repeated migrations.

[0070] Therefore, in step S102, this application first generates candidate handover prompts based on regional change information and preset configuration rules, converging the complete regional change process into one or more candidate analysis objects corresponding to specific candidate configuration files. This allows for more granular judgment of migration timing for these candidate analysis objects in conjunction with the network access state sequence. The GSMA's eSIM / RSP (Remote SIM Provisioning) framework and the ETSI's eUICC (Embedded Universal Integrated Circuit Card) standards have provided the infrastructure for remote downloading, enabling, disabling, and managing configuration files. The improvement in this application focuses on the organization of migration timing determination, rather than rewriting the underlying remote configuration protocol.

[0071] In one embodiment, the preset configuration rule refers to a pre-established set of rules used to associate the state of a target terminal during regional changes with one or more candidate configuration files. The candidate switching prompt refers to a prompt message generated for a specific candidate configuration file, indicating that subsequent local network state analysis around a specific candidate segment is required. The candidate switching prompt may include a candidate configuration file identifier, the prompt generation time, and the candidate segment range; in another embodiment, it may further include a prompt validity period, a candidate segment identifier, or a rule source identifier.

[0072] For example, a candidate switching prompt can be represented by a single prompt record containing the fields "profile_id", "t_hint", "t_start", "t_end", and "valid_duration". Here, "profile_id" indicates the corresponding candidate profile, "t_hint" indicates the prompt generation time, "t_start" and "t_end" indicate the start and end times of the candidate segment, and "valid_duration" indicates the effective duration of the prompt in subsequent analysis. Of course, candidate switching prompts are not limited to the above field combinations and can also use other data structures that can characterize the correspondence between candidate profiles and candidate segments.

[0073] In one embodiment, the preset configuration rules can be established in the form of a rule table. For example, a candidate handover rule table can be pre-established, which may include at least a region change condition field, a current active configuration file condition field, a candidate configuration file set field, and a prompt parameter field. The region change condition field describes under what region advancement or region affiliation change scenarios require the generation of candidate handover prompts; the current active configuration file condition field limits the applicable current active configuration file status; the candidate configuration file set field provides one or more candidate configuration files corresponding to the rule; and the prompt parameter field provides parameters such as the candidate segment range and the effective duration of the prompt. For example, the following rule can be set: when the target terminal continuously advances from the first operating region to the second operating region within the most recent preset time period, and the current active configuration file is the preferred configuration file P1 for the first operating region, and the terminal simultaneously stores a configuration file P2 applicable to the second operating region and a configuration file P3 applicable to the boundary coverage overlap region, then the candidate configuration file set {P2, P3} is output, and corresponding candidate handover prompts H2 and H3 are generated for P2 and P3 respectively. This rule indicates that step S102 does not immediately and uniquely determine a target configuration file after the regional change, but rather first clarifies the set of candidate configuration files that need to be analyzed in the future.

[0074] In another embodiment, the preset configuration rule can also be expressed in the form of a conditional expression. For example, it can be expressed as: "If the regional advancement condition A, the current active configuration file condition B, and the candidate configuration file meets the availability condition C, then output the candidate configuration file set D, and generate a candidate switching prompt according to the parameter set E." Here, the regional advancement condition A can be "In the last 120 seconds, the cumulative advancement amount of the target terminal location along the preset boundary direction is greater than the first advancement threshold"; the current active configuration file condition B can be "The current active configuration file belongs to the preferred configuration file of the first operating area"; the availability condition C can be "The candidate configuration file is in an enabled and allowed state and has not reached the retry limit"; the parameter set E can include the candidate segment forward expansion duration, backward expansion duration, and prompt validity duration. In yet another embodiment, the preset configuration rule can also be implemented using a strategy mapping method from regional status to the candidate configuration file set, for example, by pre-establishing a mapping relationship of "regional status identifier—candidate configuration file set—prompt parameters." The above rule table, conditional expression, and strategy mapping are merely examples and are not limited thereto.

[0075] In one embodiment, the system can first identify whether the target terminal is in a configuration migration sensitive stage based on the regional change information. If the identification result is yes, a candidate switching prompt is generated according to the preset configuration rules. The configuration migration sensitive stage refers to a stage where, during the regional advancement process, the necessity of configuration file migration for the target terminal significantly increases, but a migration command cannot be directly issued based solely on a single location event. For example, when the target terminal exhibits at least one of the following conditions within a preset time period, it can be considered to have entered the configuration migration sensitive stage: the target terminal continuously advances towards a target area different from the applicable area of ​​the current active configuration file; the target terminal has entered a preset boundary transition area; the target terminal has consistently moved away from the preferred service area of ​​the current active configuration file in recent sampling periods; or the target terminal repeatedly meets the regional advancement conditions in the same direction within a preset time period. The emphasis here is on "continuous advancement" or "transitional trend," rather than a single location point falling into a certain area. The principle behind this is that by first identifying the configuration migration sensitive stage, the originally continuous, complex, and noisy regional change process can be compressed into a small number of more analytically valuable candidate stages, thereby reducing subsequent invalid analysis.

[0076] In one embodiment, the candidate segment range can be set based on the transition span during the regional change process, the average advancement speed of the target terminal, and the expected length of the retained state context. If the candidate segment range is too narrow, there will be too few local network state samples that can be extracted subsequently, making it difficult to fully reflect the evolution of the access state before and after the configuration migration; if the candidate segment range is too wide, it is easy to introduce background states that are not directly related to the current migration candidate, reducing the relevance of subsequent analysis.

[0077] For example, a preset duration can be extended forward and backward around the prompt generation time to form candidate segments. The forward extension duration and the backward extension duration can be the same or different. For example, the candidate segment range can be set to 3 to 30 seconds forward and 3 to 30 seconds backward. For instance, in scenarios where the area boundary is relatively clear and the terminal passes through continuously at high speed, 5 seconds before and after can be used; in scenarios where the boundary coverage transition is relatively slow and the terminal moves at low speed or alternates between stopping and moving, 15 or 20 seconds before and after can be used. The above example values ​​are not isolated values, but correspond to different settings under different boundary clarity, different moving speeds, and different local context preservation requirements.

[0078] In one embodiment, the effective duration of the prompt is used to limit the time range within which the candidate handover prompt participates in subsequent local state analysis. The logic behind this setting is that if the target terminal has clearly left the corresponding area transition phase after the prompt is generated, continuing to use the prompt in subsequent analysis will reduce timeliness and may even introduce delayed judgment. The effective duration of the prompt can be set according to the area boundary span, the average speed of the target terminal, and the coverage transition length. For example, the effective duration of the prompt can be set to 5 to 60 seconds. For instance, in scenarios with a short area boundary span and continuous terminal passage, 10 seconds can be used; in scenarios with a long area boundary span or low-speed terminal advancement, 30 or 40 seconds can be used. In other words, the effective duration of the prompt is not an arbitrarily given isolated value, but a control parameter that matches the area transition length and movement rhythm.

[0079] For example, the target terminal can be a T-Box installed in a vehicle traveling across regions. The remote configuration management platform, sampling at a 1-second interval, obtains the T-Box's regional change information and network access status sequence within the last 120 seconds. It then first determines whether the vehicle has entered a configuration migration sensitive phase. Assume the currently active configuration file is the preferred configuration file P1 for the first operating area, and the terminal also stores the preferred configuration file P2 for the second operating area and the applicable configuration file P3 for the boundary overlapping coverage area. If the platform identifies, based on the regional change information, that the vehicle has been continuously moving towards the second operating area within the last 120 seconds, and that the angle between the direction of position change and the boundary normal has been continuously less than a preset angle threshold within the last 20 seconds, then it can be determined that the vehicle has entered the configuration migration sensitive phase. At this time, according to preset configuration rules, P2 and P3 are jointly determined as candidate configuration files, and candidate switching prompts H2 and H3 are generated respectively. Further, a candidate segment range of 15 seconds before and after each prompt generation time can be set, and the effective duration of the prompts is set to 30 seconds. Thus, the output of step S102 is no longer "Switch to P2 or P3 immediately," but rather two candidate switching prompts corresponding to P2 and P3 respectively. Subsequent steps then extract local state fragments around the candidate segments corresponding to H2 and H3 and make further judgments. This process avoids directly migrating the configuration file based on a single region entry event, thereby laying the foundation for more accurately reducing false triggers and repeated migrations in the future.

[0080] Regarding the above S103:

[0081] This application does not directly determine whether a candidate configuration file is suitable for migration based on the complete network access state sequence. This is because the complete network access state sequence typically includes the background state before the candidate migration stage, the key state near the candidate migration stage, and the disconnection state after the candidate migration stage. Using these states as a whole for judgment can weaken the representation effect of local access behavior changes near the candidate segment, thereby affecting the accuracy of the configuration migration timing determination. Therefore, this application first extracts local state fragments around the candidate segment represented by the candidate handover prompt, and then determines the handover stability based on these local state fragments, so that subsequent migration determination focuses more on the local network behavior of the candidate migration stage itself.

[0082] In one embodiment, a local state segment refers to a continuous state sample extracted from the network access state sequence and temporally associated with a candidate segment corresponding to a certain candidate profile; handover stability refers to the determination result used to characterize whether the corresponding candidate profile has stable migration conditions within the current candidate segment. The handover stability reflects not a single successful attachment or a single access change, but the overall stability of network access behavior within a certain duration interval in the local state segment.

[0083] In one embodiment, the candidate segment range can be determined first based on the candidate handover prompt, and then continuous state samples within the candidate segment range can be extracted from the network access state sequence as local state fragments of the corresponding candidate configuration file. For example, if the candidate handover prompt includes the start and end times of the candidate segment, state samples within that start and end time range can be directly extracted; if the candidate handover prompt includes the prompt generation time and segment expansion parameters, the corresponding state samples can be extracted after extending the prompt generation time forward and backward by a preset duration. The length of the local state fragment can be set according to the state sampling period, the duration of the area transition, and the length of the local context to be retained. For example, the local state fragment can correspond to continuous state samples within the range of 4 to 60 seconds; for example, in scenarios where the area boundary is relatively clear and terminals pass through continuously, 8 seconds or 10 seconds can be used; in scenarios where the boundary coverage transition is relatively slow or terminals advance at low speeds, 15 seconds, 20 seconds, or 30 seconds can be used. The above values ​​are not isolated values, but rather settings corresponding to different area transition spans and different state fluctuation intensities.

[0084] In one embodiment, the state samples in a local state segment may include at least one or more of the following: attach result, access identifier, and connection hold status. The attach result reflects whether the target terminal can continuously attach to the target network within the candidate segment; the access identifier reflects whether the target terminal maintains relatively continuous network access within the candidate segment; and the connection hold status reflects whether the target terminal can continuously maintain a data connection or service session after successful attach. Based on these state samples, it can be further determined whether the access behavior of the corresponding candidate profile within the candidate segment tends to be stable or exhibits strong fluctuations, and the handover stability corresponding to the candidate profile can be determined accordingly.

[0085] In one implementation, a tiered determination method can be used to determine handover stability. For example, the continuous attach duration, access identifier change count, and connection retention ratio within a local state segment can be statistically analyzed. The continuous attach duration refers to the longest duration during which the device remains in a successful attach state within the local state segment; the access identifier change count refers to the number of times the access identifier changes within the local state segment; and the connection retention ratio refers to the proportion of samples in the connection retention state within the local state segment to the total number of samples in that local state segment. If the continuous attach duration is greater than or equal to a first determination threshold, the access identifier change count is less than or equal to a second determination threshold, and the connection retention ratio is greater than or equal to a third determination threshold, then the handover stability of the corresponding candidate profile can be determined as high. If two of the above three conditions are met, the handover stability can be determined as medium. If only one condition is met or none are met, the handover stability can be determined as low. In this way, a directly executable stability determination result can be provided without relying on complex mathematical models.

[0086] In one embodiment, the first determination threshold can be set based on the total duration of the local state segment and the length of the transient attachment fluctuations to be excluded. If the first determination threshold is set too low, a small number of successful transient attachments may be misjudged as relatively stable states; if the first determination threshold is set too high, candidate configuration files that actually meet the migration conditions may be misjudged as unstable. For example, the first determination threshold can be set to 40% to 80% of the total duration of the local state segment. For instance, when the total duration of the local state segment is 20 seconds, the first determination threshold can be 8 to 16 seconds; in a normal cross-regional continuous passage scenario, it can be 10 or 12 seconds; in a scenario with obvious boundary coverage overlap and many short-term attachment fluctuations, it can be 14 or 15 seconds. The above example values ​​correspond to different setting scenarios under different fluctuation suppression requirements.

[0087] In one embodiment, the second determination threshold can be set based on the length of the local state segment, the state sampling period, and the allowable access fluctuation level. If the second determination threshold is set too low, a small number of normal access handovers may be misjudged as unstable; if the second determination threshold is set too high, it will be difficult to effectively suppress the back-and-forth switching of the access identifier in the local state segment. For example, in a local state segment, the second determination threshold can be set to 0 to 3 times. For instance, when the area boundary is relatively clear and more stringent suppression of access identifier fluctuations is required, it can be set to 1 time; in scenarios where the coverage transition is relatively slow and a small number of normal reselections are allowed, it can be set to 2 times. The above values ​​can be adjusted according to the degree of boundary coverage overlap and the terminal movement rhythm.

[0088] In one embodiment, the third determination threshold can be set according to business continuity requirements and the length of local state segments. If the third determination threshold is set too low, candidate configuration files with poor connection maintenance may also be judged as having migration conditions; if the third determination threshold is set too high, it may be overly sensitive to normal short-term connection fluctuations. For example, the third determination threshold can be set to 50% to 90%. For instance, in general data acquisition business scenarios, 60% or 70% can be used; in remote control or real-time backhaul scenarios with high business continuity requirements, 80% or 85% can be used. The above values ​​correspond to the business type and connection continuity requirements.

[0089] In one specific embodiment, the target terminal can be a T-Box installed in a vehicle traveling across regions. Assuming that a corresponding candidate handover prompt H2 has been generated for candidate profile P2 in step S102, and the candidate segment corresponding to H2 is determined to be 15 seconds before and after the prompt generation time, the platform can extract continuous state samples within this 30-second range from the network access state sequence as a local state segment corresponding to P2. If the longest continuous attachment duration within this local state segment is 18 seconds, the access identifier changes once, the connection retention rate is 80%, and the first judgment threshold set in the current scenario is 15 seconds, the second judgment threshold is 2 times, and the third judgment threshold is 70%, then all three conditions are met, and the handover stability of P2 can be determined as high. Conversely, if the longest continuous attachment duration within the local state segment corresponding to another candidate profile is only 6 seconds, the access identifier changes four times, and the connection retention rate is 40%, then its handover stability can be determined as low.

[0090] This application does not directly trigger configuration migration based on a single area entry event or a single successful attachment event. Instead, it combines the evolution of local access behavior within the candidate segment to determine whether the candidate configuration file is suitable for migration, thereby helping to reduce the occurrence of false triggers and repeated migrations.

[0091] In another implementation, handover stability is not limited to the above-mentioned high, medium, and low three-level classification. For example, handover stability can also be divided into two or four levels, or other equivalent comprehensive judgment methods can be adopted under the premise of meeting the three judgment criteria of "continuous attachment duration, number of access identifier changes, and connection retention ratio".

[0092] Regarding S104 and S105 above:

[0093] After obtaining the switching stability of each candidate configuration file, this application first determines the active candidate configuration file and updates the triggering conditions, rather than directly selecting the candidate configuration file with the highest stability for immediate migration. This is because the stability differences between candidate configuration files are not absolutely stable in many scenarios. Especially near region boundaries, in overlapping coverage areas, or in low-speed advancement scenarios, multiple candidate configuration files may exhibit high stability in turn for a short period of time. If migration is performed immediately based on only one calculation result, the configuration migration decision may become overly sensitive to short-term fluctuations, resulting in premature or repeated migrations. Therefore, in step S104, this application first further converges from multiple candidate configuration files to select the truly suitable active candidate configuration file to participate in the current migration decision, and dynamically updates the migration triggering conditions based on this active candidate configuration file, thereby establishing the migration action in step S105 on a more stable judgment basis.

[0094] In this context, the active candidate profile refers to the candidate profile that, compared to other candidate profiles, is more suitable as the actual migration target during the current candidate migration phase. The triggering condition refers to the criteria that allow the target terminal to migrate from the current active profile to the active candidate profile. The triggering condition is not fixed but can be adjusted based on the stability level of the current active candidate profile. In other words, the migration in this application is not directly triggered by static rules, but rather, based on the processing chain of candidate switching prompts, local state fragments, switching stability, and active candidate profiles, the migration triggering condition is dynamically modified in conjunction with the active candidate profile before execution.

[0095] In one implementation, the switching stability of multiple candidate profiles can be compared first, and the active candidate profile can be determined accordingly. For example, when only one candidate profile exists, if the switching stability of that candidate profile reaches a preset active threshold, it can be determined as an active candidate profile. When multiple candidate profiles exist, the candidate profile with the highest switching stability can be selected as the first preferred option, and then it can be determined whether the stability difference between the first preferred option and other candidate profiles meets a preset distinction condition. If it does, the first preferred option can be determined as an active candidate profile; if it does not, it indicates that there is still strong competition among the multiple candidate profiles. In this case, the active candidate profile can be temporarily determined, or the current active profile can be maintained and the determination can be re-evaluated after the subsequent state is further clarified. This approach avoids premature convergence to a single migration object when the stability of multiple candidate profiles is close.

[0096] In one embodiment, to ensure the direct implementation of determining the active candidate profile, the following hierarchical determination method can be adopted. If the handover stability of each candidate profile has been determined to be high, medium, or low in step S103, it can be stipulated that: only when the handover stability of a candidate profile is high does it have the basic qualification to become an active candidate profile; when multiple candidate profiles are all high, the longest continuous attachment holding time and the number of access identifier changes in their corresponding local state segments can be further compared, and the candidate profile with a longer continuous attachment holding time and fewer access identifier changes can be preferentially selected as the active candidate profile; when no candidate profile reaches high stability, the active candidate profile can be temporarily determined, and the current active profile can continue to operate.

[0097] In this way, the active candidate profile is not simply selected based on the highest score, but rather requires that a basic threshold for stable migration be reached first, and then further convergence is performed among multiple candidate profiles that have reached the basic threshold.

[0098] In one implementation, when updating the triggering conditions for configuration state migration based on active candidate profiles, at least one of the following parameters can be updated: migration trigger retention duration and original configuration retention duration. The migration trigger retention duration limits the minimum duration an active candidate profile must remain active after meeting the migration conditions; the original configuration retention duration limits the minimum duration the currently active profile should remain unchanged before triggering migration. If the switching stability of the active candidate profile is high, the migration trigger retention duration and / or the original configuration retention duration can be appropriately shortened to improve the configuration migration response speed. If a candidate profile has been selected as an active candidate profile, but its stability difference with other candidate profiles is small, or the current scenario has significant overlap and many short-term fluctuations, the migration trigger retention duration and / or the original configuration retention duration can be appropriately extended to suppress erroneous migrations. In other words, even when meeting the switching trend, the required persistence conditions for actually triggering migration can differ under different levels of stability and different scenarios.

[0099] In one embodiment, the migration trigger hold duration can be set based on the handover stability level, the area boundary transition length, and the intensity of network state fluctuations. For example, the migration trigger hold duration can be set to 3 to 60 seconds. For instance, when the handover stability of the active candidate profile is high and the current area boundary is relatively clear, 5 or 8 seconds can be used; when the active candidate profile has reached high stability, but the current terminal is in a scenario with a slow boundary transition or significant coverage overlap, 15 or 20 seconds can be used. The original configuration hold duration can also be set based on the remaining stability of the current active profile in the current area and the need to avoid premature exit. For example, the original configuration hold duration can be set to 5 to 90 seconds.

[0100] For example, a shorter hold duration can be used when the current active profile has clearly lost its preferred access conditions; a longer hold duration can be used when the current active profile still has a certain degree of stable access capability and migration caution needs to be increased. The example values ​​above are not isolated values, but control parameters that match the boundary transition length, fluctuation intensity, and migration caution level.

[0101] In one embodiment, in step S105, in response to the current active profile meeting the updated triggering condition, the target terminal can be controlled to migrate from the current active profile to the active candidate profile. Meeting the updated triggering condition means that the state relationship between the current active profile and the active candidate profile has continuously met the migration requirements dynamically updated in step S104.

[0102] For example, once an active candidate profile is determined, if the active candidate profile maintains high stability throughout the updated migration trigger retention period, and the current active profile has met the updated original configuration retention period requirement, then migration can be triggered. "Migration" may include enabling the active candidate profile, switching it to the new active profile, and changing the current active profile from an active state to an inactive state. Actions such as enabling / disabling profiles, network access reconstruction, and terminal-side state synchronization can be implemented using conventional embedded SIM remote configuration management methods, and are not elaborated upon here.

[0103] For example, the target terminal can be a T-Box installed in a vehicle traveling across regions. In step S103, the platform has obtained the switching stability of candidate profiles P2 and P3, where the switching stability of P2 is high and the switching stability of P3 is medium. Therefore, the platform can determine P2 as the active candidate profile. Further, assuming the current vehicle is in a scenario with clear boundaries and continuous passage, the platform can update the migration trigger hold duration to 8 seconds and the original configuration hold duration to 10 seconds. If the current vehicle is in a scenario with slow speed or significant boundary overlap, the migration trigger hold duration can be updated to 20 seconds and the original configuration hold duration to 25 seconds. Subsequently, the platform continues to monitor the status of the current active profile and the active candidate profile. When P2 maintains high stability within the updated migration trigger hold duration, and the current active profile has met the updated original configuration hold duration requirement, the platform controls the target terminal to migrate from the current active profile to P2. Through this process, this application does not migrate immediately after obtaining the switching stability. Instead, it first converges active candidate configuration files, then dynamically adjusts the migration triggering conditions based on these active candidate configuration files, and finally executes the migration only when the current active configuration file meets the updated conditions. This helps to further reduce the incidence of false triggering and repeated migration caused by short-term network fluctuations during regional changes.

[0104] In another embodiment, the method for determining the active candidate configuration file is not limited to the hierarchical convergence method described above. For example, when the number of candidate configuration files is small, a unique active candidate configuration file can be determined directly based on stability ranking, or when the stability of multiple candidate configuration files is close, the active candidate configuration file can be temporarily not output and the migration decision can be postponed. Similarly, the method for updating the triggering conditions is not limited to adjusting the migration trigger retention time and the original configuration retention time; adjustments can also be made in conjunction with other control parameters related to migration persistence, without deviating from the core idea of ​​this application.

[0105] Optional, see Figure 2 The flowchart below shows a method for generating candidate switching prompts provided in this application embodiment, including steps S201 to S202, wherein:

[0106] S201: Determine the regional change trajectory based on the regional change information, and extract the adjacent segments of the regional change trajectory;

[0107] S202: Based on the temporal position of each of the attribution change adjacent segments in the regional change trajectory and the preset configuration rules, determine the candidate configuration files associated with each of the attribution change adjacent segments, and generate candidate switching prompts for each of the candidate configuration files.

[0108] In one implementation, simply knowing that the target terminal has undergone a regional change is insufficient to support subsequent migration determination, because a real regional change process often simultaneously includes stable segments, transition segments, and departure segments. If the local segments truly related to the change of ownership cannot be extracted from the complete regional change process, the generation of subsequent candidate handover prompts will still be too coarse, easily incorporating irrelevant states into the analysis. To address this, this application further first determines the regional change trajectory, and then extracts the "adjacent segments of the ownership change" from the regional change trajectory, so that the candidate handover prompts are no longer broad prompts targeting the entire regional change process, but rather fine prompts targeting the local transition segments most relevant to the ownership change.

[0109] In one embodiment, the regional change trajectory refers to the trajectory representation result obtained by organizing regional change information acquired within a preset time period in chronological order. For example, when the regional change information is represented using a latitude and longitude coordinate sequence, the coordinate points at each sampling time can be connected in chronological order, and combined with an electronic map or preset regional boundary information, each sampling point can be mapped to a corresponding regional label, thereby forming a regional label trajectory with chronological order. When the regional change information is represented using cellular area affiliation information, the cell affiliation label, location area label, or tracking area label corresponding to each sampling time can be organized in chronological order and then processed to form the regional change trajectory. The jitter removal process refers to preventing occasional positioning offsets or instantaneous affiliation drifts of a single sampling point from interfering with the regional trajectory. It requires the same candidate regional label to appear consecutively a preset number of times before the regional label is confirmed as a valid trajectory state. For example, when the state sampling period is 1 second, the consecutive occurrence count can be 3 to 5 times. For instance, 3 times can be used in normal continuous passage scenarios; 5 times can be used in scenarios with overlapping border coverage or significant positioning jitter. The example values ​​above correspond to different settings for varying positioning fluctuation intensities and different trajectory smoothing requirements.

[0110] In one implementation, the attribution change adjacent segment refers to the local segment in the region change trajectory that is closest to the transition of region attribution from the first attribution region to the second attribution region. In other words, the attribution change adjacent segment is not the entire trajectory during the transition period, but rather a local segment surrounding the attribution change event that best represents the state before and after the attribution switch. The principle behind this is that the generation of candidate profiles should focus as much as possible on the local stage where the attribution change actually occurs, and should not include stable region states long before the attribution change or disengagement states long after the attribution change in the same candidate generation logic.

[0111] For example, the "last stable moment belonging to the first home region" and the "first stable moment belonging to the second home region" can be identified first in the regional change trajectory, and then the transition interval between these two moments can be used as the basic adjacency segment. Furthermore, a preset buffer duration can be extended forward and backward respectively to form the home change adjacency segment. The buffer duration can be set according to the average moving speed of the target terminal and the width of the regional boundary transition. For example, the buffer duration can be 3 to 20 seconds. For instance, in a high-speed continuous cross-region scenario, 5 seconds can be used; in a low-speed advancement or a scenario with a relatively gentle boundary coverage transition, 10 or 15 seconds can be used.

[0112] In one embodiment, the "temporal position" of an adjacent segment with a change in ownership can be represented by the center time of the adjacent segment or by the start time of entering the adjacent segment. To facilitate subsequent extraction of local state segments before and after a given time, in one specific embodiment, the center time of the adjacent segment can be used as its temporal position. Specifically, the midpoint between the start and end timestamps of the adjacent segment with the change in ownership can be used as the temporal position corresponding to that adjacent segment. The advantage of this approach is that subsequent steps extracting state segments forward and backward around this temporal position can more evenly cover the local access evolution process before and after the ownership change.

[0113] In one implementation, after extracting the adjacent segments of the attribution change and their temporal positions, candidate configuration files associated with them can be determined according to preset configuration rules, and corresponding candidate switching prompts can be generated. Here is a rule example that can be directly implemented. Assume the target terminal is a T-Box installed in a cross-regional vehicle, the currently active configuration file is the preferred configuration file P1 for the first operating area, and the terminal also stores the preferred configuration file P2 for the second operating area and the applicable configuration file P3 for the boundary overlapping coverage area. The remote configuration management platform first forms a regional change trajectory based on the regional change information within the last 120 seconds, and identifies the adjacent segments of the attribution change transitioning from the first operating area to the second operating area within this trajectory. If the temporal position of this adjacent segment is denoted as t0, the platform can generate the following candidate switching prompts according to preset configuration rules: candidate switching prompt H2 corresponding to P2, and candidate switching prompt H3 corresponding to P3. Each candidate switching prompt may include at least the corresponding configuration file identifier, the temporal position of the adjacent segment, and the range of candidate segments constructed around that temporal position. For example, by extending the timeframe forward and backward by 15 seconds around t0, candidate segments corresponding to H2 and H3 can be generated respectively. Therefore, subsequent steps do not require indiscriminate analysis of the entire region's change process; instead, they can focus on determining the local state of the candidate configuration files around the local stages truly relevant to the change in ownership.

[0114] This approach first generates a regional change trajectory from the regional change information, then extracts the adjacent segments of the attribution change and their temporal positions from the trajectory, and finally generates candidate switching prompts corresponding to the candidate configuration files based on these temporal positions and preset configuration rules. This ensures that the generation of candidate switching prompts is closely aligned with the actual attribution change process, thereby improving the relevance of subsequent local state analysis.

[0115] In one embodiment, even if the candidate analysis scope has been narrowed down to the vicinity of the adjacent segment of the attribution change, if only the state in a single direction within that segment is observed, such as only looking at the state after the candidate's temporal position, it is still difficult to determine whether the local access behavior corresponding to a certain candidate profile is a stable migration condition formed by continuous transition or an occasional attachment result caused by instantaneous fluctuations. To this end, this application further extracts the preceding state segment and the following state segment simultaneously, centered on the temporal position corresponding to the candidate switching prompt, and compares the access behavior continuity relationship and the attachment result retention relationship between the preceding and following segments to determine whether the candidate profile truly possesses stable migration conditions.

[0116] In one embodiment, a preceding state segment refers to a continuous state sample that is temporally adjacent to the corresponding time position of the candidate switching prompt; a following state segment refers to a continuous state sample that is temporally adjacent to the corresponding time position. For example, if the time position corresponding to the candidate switching prompt is t0, then continuous state samples within the range [t0-T1, t0) can be used as the preceding state segment, and continuous state samples within the range (t0, t0+T2) can be used as the following state segment, where T1 and T2 represent the duration of the preceding segment and the duration of the following segment, respectively.

[0117] For example, T1 and T2 can be set identically, with the preceding and following state segments each taking 5 to 30 seconds. For instance, in scenarios where the region boundary is clear and the terminal passes through continuously, 8 or 10 seconds can be taken before and after; in scenarios where the boundary coverage transition is gradual, the terminal advances slowly, or alternates between stopping and moving, 15 or 20 seconds can be taken before and after. These example values ​​are not isolated values, but rather parameter settings that match the region transition length, state sampling period, and the accuracy requirements for subsequent continuity judgment.

[0118] In one embodiment, access identifier continuity refers to the determination result used to characterize whether the evolution of access identifiers between preceding and subsequent state segments exhibits a continuous, unidirectional migration process, rather than a repeated back-and-forth process. For example, the access identifier can be defined as identification information that characterizes the current access affiliation of the target terminal, such as an operator network identifier, a regional access identifier, or a preset composite access identifier. Further, access identifier continuity can be determined as follows: first, statistically analyze the changes in access identifiers in several samples at the end of the preceding state segment and several samples at the beginning of the subsequent state segment; then, determine whether the changes represent a continuous transition towards the same target affiliation direction.

[0119] For example, if the access identifier changes in the same direction only once between the end of the preceding state segment and the beginning of the following state segment, and does not return to the previous access identifier after the change, the access identifier continuity can be determined as high; if there is a back-and-forth bounce, the access identifier continuity can be determined as medium; if there are two or more repeated switching near the boundary between the preceding and following segments, the access identifier continuity can be determined as low. The "high, medium, and low" given here is a judgment embodiment that can be directly implemented, and its essence is to distinguish between "continuous migration" and "boundary jitter".

[0120] In one embodiment, the consistency of the attachment result refers to the determination result used to characterize whether the attachment result between the preceding state segment and the following state segment exhibits a stable transition relationship in the same direction. For ease of implementation, the determination can be based on the distribution of successful attachment samples in the preceding and following state segments. For example, if the successful attachment state is maintained continuously in the following state segment, and there are no consecutive attachment failure samples exceeding a preset number between the end of the preceding state segment and the beginning of the following state segment, the consistency of the attachment result can be determined as high; if the following state segment is dominated by successful attachment states, with a small number of short-term attachment failures near the boundary, the consistency of the attachment result can be determined as medium; if the successful attachment state cannot be maintained continuously in the following state segment, or if long failure segments appear consecutively near the boundary, the consistency of the attachment result can be determined as low. The preset number can be set according to the state sampling period and the length of the instantaneous attachment jitter to be excluded. For example, when the state sampling period is 1 second, the number of consecutive failed samples allowed can be set to 1 to 3 times; for example, 1 time can be taken in a normal continuous passage scenario, and 2 or 3 times can be taken in a scenario where the boundary coverage fluctuation is obvious.

[0121] In one implementation, the handover stability of a corresponding candidate profile can be determined based on both access identifier continuity and attachment result consistency. Here is a directly implementable example: If both access identifier continuity and attachment result consistency are high, the handover stability of the corresponding candidate profile can be determined as high; if one is high and the other is medium, or both are medium, the handover stability can be determined as medium; if either access identifier continuity or attachment result consistency is low, the handover stability can be determined as low. The principle behind this is that whether a candidate profile is suitable for migration depends not only on the success of the attachment result itself, but also on the continuity of the access identifier evolution; only when the preceding and following segments exhibit stable transitions in both dimensions can it be considered to truly possess high migration stability.

[0122] For example, if a candidate handover prompt H2 has been generated for candidate configuration file P2, its corresponding time position is t0. The platform can extract 10 seconds of state samples before t0 as the preceding state segment and 10 seconds of state samples after t0 as the following state segment. If the access identifier at the end of the preceding state segment smoothly transitions from the first home access identifier to the second home access identifier, and the following state segment does not return to the first home access identifier after its start, the access identifier continuity can be determined as high. At the same time, if the successful attachment state is maintained continuously within the following state segment, and only one short-term attachment failure occurs near the boundary, the attachment result consistency can be determined as high. Based on these two determinations, the handover stability of P2 can be further determined as high. Through this processing, the handover stability is no longer determined solely by "whether the attachment was successful," but by combining the continuity of access behavior before and after the candidate time position and the consistency of the attachment results, thereby more accurately distinguishing between true stable migration conditions and occasional fluctuations near the boundary.

[0123] In one embodiment, even if it is possible to extract preceding and succeeding state segments around candidate time-series positions and determine handover stability at a certain time scale based on access identifier continuity and attachment result consistency, stability at a single time scale may still be affected by short-term sporadic fluctuations. For example, a candidate profile may appear stable within a short observation window but may not maintain the same stability within a longer observation window. If the active candidate profile is determined solely based on the results at a single time scale, it may still be overly sensitive to short-term pseudo-stability. Therefore, this application further introduces handover stability judgments at at least two different time scales and determines the active candidate profile based on the cross-scale consistency between results at different time scales.

[0124] In one implementation, different time scales can be understood as using different lengths of preceding and following state segments to determine the stability of a transition around the same candidate transition prompt corresponding to a temporal position. For example, a first time scale can be set as a shorter local observation scale to more sensitively capture immediate transition features near the candidate temporal position; a second time scale can be set as a longer local observation scale to determine whether the transition feature can remain consistent over a longer context. For ease of implementation, in one specific implementation, when the preceding and following segment lengths corresponding to the first time scale are 8 seconds or 10 seconds, the second time scale can correspond to preceding and following segment lengths of 20 seconds or 30 seconds. That is, the second time scale does not reselect another candidate temporal position, but rather expands the observation range around the same candidate temporal position.

[0125] In one embodiment, the switching stability of each candidate profile can be determined at a first time scale and a second time scale, respectively, in the manner described above. For example, for candidate profile P2, its first switching stability can be determined based on the preceding / following state segments at the first time scale, and its second switching stability can be determined based on the preceding / following state segments at the second time scale; similarly, for candidate profile P3, its corresponding switching stability at both time scales can also be determined. Through such dual-scale determination, "short-term stability" can be distinguished from "stability over a longer period of time".

[0126] In one implementation, cross-scale consistency refers to whether the switching stability results of the same candidate configuration file obtained at at least two different time scales exhibit a stable and consistent determination relationship.

[0127] For example, the following determination method can be adopted: if the switching stability of a candidate profile is high at both time scales, its cross-scale consistency can be determined as high; if it is high at a shorter time scale and medium at a longer time scale, or medium at both time scales, its cross-scale consistency can be determined as medium; if it is high at a shorter time scale but low at a longer time scale, or the difference between the two time scales is large, its cross-scale consistency can be determined as low. In this way, the object that is truly suitable to become an active candidate profile should not only perform well within a short time window, but should also maintain similar stability judgment results after expanding the observation range.

[0128] In one implementation, active candidate profiles can be determined based on cross-scale consistency among different candidate profiles. A directly implementable determination logic is provided here. If only one candidate profile has high cross-scale consistency, it can be directly identified as an active candidate profile. If multiple candidate profiles have high cross-scale consistency, the candidate profile with higher switching stability over a longer timescale can be preferentially selected. If none of the candidate profiles have high cross-scale consistency, the active candidate profile can be temporarily not determined, and the current active profile can continue operating, awaiting new candidate switching prompts or new state samples before re-determining. In this way, the determination of active candidate profiles depends not only on their instantaneous performance at a single scale but also on whether they can consistently exhibit migration stability across different observation ranges.

[0129] For example, for candidate configuration files P2 and P3, the platform has determined the corresponding handover stability based on shorter time scales (10 seconds before and after) and longer time scales (20 seconds before and after). If the handover stability of P2 is high and high at the two time scales, respectively, while that of P3 is high and low at the two time scales, then the cross-scale consistency of P2 can be determined as high, and the cross-scale consistency of P3 as low, thus identifying P2 as the active candidate configuration file. Alternatively, if P2 is high and medium at the two time scales, and that of P3 is medium and medium at the two time scales, then P2 can be prioritized as the more suitable candidate configuration file. This processing further suppresses the impact of short-term, occasional fluctuations on the selection of migration objects, thereby ensuring that subsequent configuration state migrations are based on more stable candidate objects.

[0130] This elevates the assessment of switching stability at a single timescale to a cross-scale consistency assessment across at least two different timescales. Instead of relying solely on local state performance within a short window to determine active candidate profiles, it requires candidate profiles to exhibit relatively consistent migration stability across different observation ranges, thus better reducing the misjudgment rate caused by short-term fluctuations.

[0131] Optional, see Figure 3 The flowchart below shows a method for migrating to an active candidate profile provided in an embodiment of this application, including steps S301 to S304, wherein:

[0132] S301: In response to the fact that the active candidate configuration file is not stored in the target terminal, control the target terminal to download the active candidate configuration file;

[0133] S302: After the active candidate configuration file is downloaded, the activation of the active candidate configuration file is delayed, and the network access status sequence corresponding to the target terminal is obtained again.

[0134] S303: Update the handover stability corresponding to the active candidate configuration file based on the network access state sequence that is continuously acquired;

[0135] S304: In response to the updated handover stability continuously satisfying the updated triggering conditions, control the target terminal to activate the active candidate configuration file.

[0136] In one implementation, after identifying an active candidate configuration file, activation is not performed immediately. Instead, it is first determined whether the active candidate configuration file is already stored on the target terminal. This determination is made first because, in cross-regional migration scenarios, the target terminal may already have multiple available configuration files pre-stored, or it may not yet have a configuration file corresponding to the current migration target. If the active candidate configuration file is pre-stored on the target terminal, the subsequent activation preparation stage can proceed directly; if the active candidate configuration file is not yet stored, it needs to be downloaded first. The technical significance of this approach is that by first ensuring the availability of active candidate configuration files, and then further evaluating whether the activation opportunity has truly arrived, it avoids processing "whether the target configuration file is available" and "whether the migration opportunity is met" in the same stage.

[0137] In one embodiment, in response to the fact that the active candidate configuration file is not stored on the target terminal, the remote configuration management platform can issue a corresponding configuration file download control command to the target terminal, and the target terminal can then execute the download of the active candidate configuration file. For the underlying actions during the download process, such as authentication, fragmented transmission, integrity checks, and installation / writing, conventional remote configuration management techniques in the art can be used. The key point here is not how to define the download protocol, but rather that the active candidate configuration file is not immediately activated after the download is completed.

[0138] In one implementation, after the active candidate profile is downloaded, the activation of the active candidate profile is delayed, and the network access state sequence corresponding to the target terminal continues to be acquired. "Delayed activation" here means that after downloading, the active candidate profile is not immediately switched to the active profile, but rather the current active profile is maintained to continue to handle current network access, and the network access behavior is observed for an additional period of time.

[0139] This is done because, although the determination of the active candidate profile has been completed based on previous state samples, the location and network access status of the target terminal may continue to change in cross-regional deployment scenarios. If the target profile is activated immediately upon completion of the download, premature switching to the active candidate profile may still occur due to instantaneous fluctuations near the boundary, short-term coverage overlap, or local network jitter. By continuing the observation process after the download is completed, it is possible to further confirm whether the migration judgment obtained in the previous stage still holds true.

[0140] In one implementation, the continued observation phase after download completion can correspond to a post-download observation window. This post-download observation window can be set based on the target terminal's movement speed, the area transition length, and the intensity of access fluctuations near the boundary. If the target terminal's movement speed is high and the area boundary is clear, the post-download observation window can be appropriately shortened to avoid unnecessary delays in effective migration; if the target terminal is in a low-speed advancement, slow movement, or significantly overlapping boundary coverage scenario, the post-download observation window can be appropriately extended to preserve more sufficient subsequent access status. For example, the post-download observation window can be set to 5 to 60 seconds. For instance, in continuous cross-area passage scenarios, 8 or 10 seconds can be used; in low-speed slow movement, port queuing, or slow coverage transition scenarios, 20, 30, or 40 seconds can be used. These example values ​​are not isolated values ​​but correspond to the area transition rhythm and the level of caution in continued confirmation.

[0141] In one implementation, after acquiring network access state sequences within the post-download observation window, the handover stability corresponding to the active candidate profile can be re-determined based on these acquired sequences. This updated handover stability does not involve recalculating from the initial stage across all candidate profiles; instead, it re-confirms the current migration stability of the already determined active candidate profile by incorporating new state samples after the download. This separates the process of first determining the active candidate profile based on existing states from the process of re-confirming whether the active candidate profile is still suitable for activation after the download, thus forming a two-stage decision process.

[0142] In one embodiment, the updated handover stability can be determined using the same or a method consistent with the aforementioned handover stability determination. For example, the continuous attachment duration, access identifier changes, and connection persistence percentage can be statistically analyzed within the observation window after download completion. Based on a pre-set threshold, the updated handover stability can be categorized as high, medium, or low. Further, if the updated handover stability continuously meets the updated triggering conditions within the post-download observation window, the target terminal is controlled to activate an active candidate profile; if the updated handover stability fails to continuously meet the updated triggering conditions, the current active profile is maintained, and a retry can be postponed or the process can wait for the next candidate stage for re-evaluation.

[0143] In one specific embodiment, the target terminal can be a T-Box installed in a vehicle traveling across regions. The platform has already identified configuration file P2 as an active candidate configuration file in the previous stage, but detects that P2 is not yet stored in the target terminal. Therefore, it first controls the terminal to download P2. Assuming P2 is downloaded after 12 seconds, the platform does not immediately activate P2, but instead keeps the current active configuration file P1 active and continues to collect the network access state sequence for the following 20 seconds. If, within these 20 seconds, the handover stability obtained from the update around P2 remains high and continues to meet the migration triggering conditions updated in the previous stage, then the platform controls the terminal to activate P2. If, within these 20 seconds, the network access state corresponding to P2 experiences a significant bounce or connection instability, then P1 is maintained, and activation is not performed. This process avoids the risk of premature migration caused by activating immediately after downloading.

[0144] In one implementation, after an active candidate profile is activated, the current active profile is not immediately and completely deactivated; instead, it is kept in an inactive reserved state. The inactive reserved state means that the current active profile is no longer used as the active profile actually carrying the current network access, but it remains in the target terminal and maintains a recoverable and rollbackable usable state. Thus, in cross-regional migration scenarios, even if an active candidate profile has been activated and successfully undertaken the current network access, it does not mean that this state will necessarily remain stable for a period of time. If the original profile is deleted, deactivated, or irrecoverably deactivated immediately after activation, the ability to quickly restore the original connection path will be lost if the new profile experiences a bounce, instability, or inability to continuously attach shortly after activation.

[0145] In one implementation, a set of configuration file states can be maintained within the target terminal, including at least an active state and an inactive reserved state. For example, before the active candidate configuration file P2 is activated, the original active configuration file P1 is in an active state; when P2 is activated, P2 becomes active, while P1 becomes inactive and reserved. At this time, although P1 no longer carries the current connection, it can still exist as a subsequent fallback target. This state organization method allows the migration process to no longer be a simple activation of a new configuration file followed by the complete exit of the old configuration file, but rather a two-stage migration structure where the new configuration file takes over first, and the old configuration file is temporarily retained.

[0146] In one implementation, the duration of the inactive retention state can be set based on the access fluctuation intensity, service continuity requirements, and allowable fallback window length of the target terminal's scenario. If the coverage transition near the boundary is short and the new profile usually stabilizes quickly after activation, the duration of the inactive retention state can be relatively short; if the boundary transition is slow, the terminal is in a low-speed advance, or the risk of short-term bounce is high, the duration of the inactive retention state can be appropriately extended. For example, the duration of the inactive retention state can be set from 10 seconds to 180 seconds. For instance, 20 seconds or 30 seconds can be used in normal continuous cross-area scenarios; 60 seconds, 90 seconds, or 120 seconds can be used in low-speed slow-moving scenarios, boundary overlapping coverage scenarios, or scenarios with high bounce risk. The above example values ​​are not isolated values, but rather protection windows that match actual fallback requirements and fluctuation risks.

[0147] For example, the target terminal is a T-Box in a vehicle traveling across regions. After the platform determines that P2 meets the migration conditions and activates P2, the original active configuration file P1 does not immediately exit but enters an inactive reserved state. Subsequently, within a preset reservation window, if the access status of P2 remains stable, P1 continues to remain in the inactive reserved state and waits for subsequent rollback; if P2 becomes unstable shortly after activation, the platform can prioritize using P1 to perform a fast rollback. This process can significantly reduce the recovery cost of anomalies occurring shortly after migration.

[0148] In one implementation, even if active candidate profiles have been determined based on switching stability at at least two different time scales, the cross-scale consistency levels among different active candidate profiles may still differ. If the same set of fixed triggering conditions is applied to all active candidate profiles, the following situation may occur: for candidates with high cross-scale consistency and strong stability, migration triggering may be too conservative; while for candidates with weak cross-scale consistency and good performance only in short time scales, migration triggering may be too aggressive. Therefore, it is necessary to further adjust the migration triggering timing and the original configuration retention duration based on cross-scale consistency to ensure that the triggering conditions are compatible with the stability of the active candidate profiles.

[0149] In one implementation, the switching trigger timing refers to the duration of time required from the determination of the active candidate profile to the actual permission for the migration action to occur; the original configuration retention duration refers to the minimum duration for which the currently active profile must remain active, given that the migration decision already has a trend basis. By adjusting these two parameters separately, it is possible to simultaneously control when switching is allowed and how long the original configuration must be retained, thereby avoiding premature triggering of the migration action and unnecessary migration execution delays.

[0150] In one implementation, the following adjustment method can be adopted. If the cross-scale consistency of an active candidate configuration file is high, it indicates that the candidate configuration file exhibits a relatively consistent and stable state under both short and long time scales. In this case, the switching trigger timing can be appropriately advanced, and the original configuration retention time can be shortened. For example, the switching trigger timing can be set to a shorter retention time range, and the original configuration retention time can be set to a shorter retention time range. Conversely, if the cross-scale consistency of an active candidate configuration file is medium, it indicates that although the candidate configuration file is superior to other candidates, its cross-time scale stability has not yet fully converged. In this case, the switching trigger timing can be appropriately delayed, and the original configuration retention time can be extended to suppress premature migration caused by short-term occasional fluctuations.

[0151] For example, the handover trigger timing and the original configuration retention duration can be set as tiered parameters. For instance, when cross-scale consistency is high, the handover trigger timing can be set to 5-10 seconds, and the original configuration retention duration can be set to 8-15 seconds; when cross-scale consistency is medium, the handover trigger timing can be set to 15-30 seconds, and the original configuration retention duration can be set to 20-40 seconds. Furthermore, in scenarios with slow boundary coverage, slow advancement, or port queuing, an additional extension can be added, for example, an additional extension of 5-20 seconds.

[0152] Thus, higher cross-scale consistency indicates that the candidate exhibits a strong and stable migration trend across different observation ranges, thereby shortening the confirmation wait time; lower cross-scale consistency requires a longer confirmation time in exchange for higher migration reliability.

[0153] For example, if the active candidate profile P2 exhibits high handover stability across both short and long timescales, its cross-scale consistency can be determined as high, and the handover trigger timing can be updated to 8 seconds, while the original configuration hold duration can be updated to 12 seconds. If, in another candidate phase, the active candidate profile P3 exhibits high stability across short and medium stability across long timescales, its cross-scale consistency can be determined as medium, and the handover trigger timing can be updated to 20 seconds, while the original configuration hold duration can be updated to 25 seconds. Through this process, the migration trigger thresholds corresponding to different active candidate profiles are no longer entirely identical, but can be dynamically adjusted according to the differences in cross-scale consistency, thus better reflecting the stability of real-world local network behavior.

[0154] In one implementation, after the active candidate profile is activated and the original active profile is in an inactive reserved state, it is still necessary to further determine when the rollback capability can be deactivated. This is because "the new profile has been activated" and "the new profile has stabilized to the point that no further rollback operation is needed" are not the same concept. Especially in scenarios with overlapping boundary coverage, short-term jitter, or low-speed advancement, although the new profile has been activated, its access status may still be at risk of rollback for a period of time afterward.

[0155] Therefore, before the original active profile is changed from an inactive reserved state to a non-rollback state, it is necessary to obtain the subsequent network access state again based on the active candidate profile and extract the corresponding subsequent state fragments to determine whether the consistency of its subsequent attachment results and the continuity of access identifiers have met the conditions for lifting the rollback.

[0156] In one implementation, the network access state sequence corresponding to the active candidate profile can be acquired after the active candidate profile is activated, and subsequent state segments can be extracted from it. The term "subsequent state segment" here refers to a continuous state sample that is located after the activation time of the active candidate profile and is used to reflect the local access behavior after the profile is activated.

[0157] For example, the subsequent state segment can be a continuous state sample within the range of 5 to 60 seconds after the activation time. For instance, in scenarios with continuous cross-regional coverage and relatively clear boundaries, 10 seconds can be used; in scenarios with slow boundary coverage transition, high risk of backflip, or slow terminal advancement, 20, 30, or 40 seconds can be used. The above values ​​are not isolated values, but correspond to the length of the observation window after activation and the degree of caution in the rollback cancellation.

[0158] In one implementation, the consistency of the attachment result and the continuity of the access identifier can be determined based on the subsequent state segment. The determination logic here can be connected with the aforementioned local state analysis stage, but its focus has changed from "whether it is suitable for migration" to "whether it is sufficient to remove the rollback". For example, if the attachment success state is maintained continuously in the subsequent state segment and there are no attachment failure samples exceeding a preset number consecutively, the consistency of the attachment result can be determined as high; if the access identifier in the subsequent state segment remains the same target home identifier after activation and does not jump back to the access identifier corresponding to the old configuration, the continuity of the access identifier can be determined as high.

[0159] In one implementation, the preset rollback release condition can be set as follows: the consistency of the attachment result and the continuity of the access identifier corresponding to the subsequent state segment continuously meet the high-level judgment condition. Here, "continuously meet" means that within the entire observation window of the subsequent state segment, neither of the above two types of judgment results drops below the preset release threshold.

[0160] For example, it can be stipulated that the current active profile can only be changed from an inactive reserved state to a non-reversible state if both the consistency of the attachment result in the subsequent state segment and the continuity of the access identifier are high; if neither of these conditions is met, the original active profile will remain in an inactive reserved state. The technical significance of this is that the rollback capability of the old profile is only truly removed when the new profile continues to exhibit stable and consistent access behavior after activation.

[0161] In one specific embodiment, the target terminal is a T-Box in a vehicle traveling across regions. After the platform activates the active candidate profile P2, it continues to acquire the network access state sequence within the next 30 seconds and uses this 30-second state sample as a subsequent state segment. If, within this subsequent state segment, P2 continues to attach successfully without more than two consecutive attachment failures, and the access identifier remains the target home access identifier without returning to the access identifier corresponding to the old profile P1, then it can be determined that the consistency of the attachment result and the continuity of the access identifier continuously meet the preset rollback release conditions. At this time, P1 can be adjusted from an inactive reserved state to a non-rollback state. Conversely, if P2 experiences a short-term bounce or the access identifier returns to the home corresponding to P1 within the observation window, P1 continues to be kept in an inactive reserved state to retain the subsequent rapid rollback capability. Through this process, this application forms a phased migration structure of "first activating the new profile, then observing its subsequent stability, and finally releasing the rollback capability of the old profile," thereby further reducing the recovery risk caused by a short-term bounce after migration.

[0162] In one embodiment, before determining the handover stability of a candidate profile based on preceding and following state segments, regional change information can be used to determine whether there are queuing passage sections in the regional change trajectories corresponding to the preceding and / or following state segments where continuous dwell states and short-distance advance states alternate. This determination is introduced because in cross-regional scenarios, especially at border crossings, toll booths, toll gates, or inspection queues, the target terminal does not cross the boundary at a continuous, uniform speed, but often exhibits an alternating movement pattern of "dwelling—short-distance advance—dwelling again." During this process, the terminal may repeatedly experience attachment success events within locally overlapping coverage areas. If such scenarios are not distinguished from general continuous passage scenarios, attachment success events in subsequent preceding and following state segments may be mistakenly considered evidence of genuine stable migration, thereby inflating the handover stability of the corresponding candidate profile.

[0163] In one embodiment, a queuing passage segment refers to a local segment in the area's trajectory that meets the following motion characteristics: the target terminal exhibits at least one alternation between a continuous stationary state and a short-distance advancing state within this segment, and this alternation repeats a preset number of times within a preset time window. A continuous stationary state refers to a state where the target terminal's position change is less than a stationary threshold at several consecutive sampling times, or where the corresponding movement speed is consistently lower than a stationary speed threshold. A short-distance advancing state refers to a state where, within a preset advancing window after the continuous stationary state ends, the target terminal's cumulative advancing distance along the area's trajectory is greater than the stationary threshold but less than the preset advancing threshold. By alternating between these two states, a true queuing slow-moving scenario can be distinguished from a normal continuous passage scenario.

[0164] For example, queuing passage sections can be identified as follows: First, a regional change trajectory is formed based on regional change information, and the position change or velocity value between adjacent sampling times is calculated in chronological order. If the position change for N_1 consecutive sampling times is less than the dwell threshold, or the velocity for N_1 consecutive sampling times is less than the dwell velocity threshold, then the time period is marked as a continuous dwell state. Subsequently, if, within a preset advancement window following this continuous dwell state, the cumulative advancement distance of the target terminal is greater than the dwell threshold but less than the preset advancement threshold, then the advancement process is marked as a short-distance advancement state. If, within a preset time window, the continuous dwell state and the short-distance advancement state alternate for a preset number of times, such as 2 times, 3 times, or more, then the corresponding regional change trajectory section can be determined as a queuing passage section.

[0165] In one embodiment, the dwell threshold, dwell speed threshold, preset advance threshold, and alternation number threshold can be set according to the target terminal type, positioning accuracy, and application scenario. If the target terminal is a T-Box installed in a vehicle, the dwell threshold can be determined based on the positioning jitter range and typical coordinate drift when the vehicle is stationary; the dwell speed threshold can be determined based on the vehicle's speed measurement results in a scenario of "basically stationary but with slight pose jitter". For example, the dwell threshold can be set to 1 meter to 5 meters, and the dwell speed threshold can be set to 0.2 meters / second to 1 meter / second. For instance, in scenarios with high satellite positioning accuracy and good road conditions, the dwell threshold and dwell speed threshold can be set to 2 meters and 0.5 meters / second, respectively; in scenarios with more obstruction and slightly larger positioning jitter, the dwell threshold and dwell speed threshold can be set to 4 meters and 0.8 meters / second, respectively. The preset advance threshold is used to distinguish between "small advances in queues" and "continuous advances in normal continuous driving", and it can be determined based on the port lane length, the average queuing step distance of vehicles, and the road organization method near the boundary. For example, the preset advance threshold can be set to 5 meters to 50 meters. For example, 8 or 10 meters can be used in scenarios with slow-moving traffic at checkpoints; 20 or 30 meters can be used in scenarios with long convoys advancing in segments. The threshold for the number of alternations can be set to 2 to 5 times to ensure that the identified queuing sections have sufficient repeatability, rather than occasional pauses. The example values ​​above are not isolated values, but rather correspond to settings under different positioning accuracies, different vehicle advancement rhythms, and different scenario complexities.

[0166] In one embodiment, after determining that there are queuing passage sections in the regional change trajectories corresponding to the preceding and subsequent state segments, short-cycle repetitive attachment records in the preceding and subsequent state segments can be further identified. A short-cycle repetitive attachment record refers to an attachment record in which the change in the target terminal's position between adjacent successful attachment events is less than a preset advancement threshold, and the duration of each successful attachment is less than a preset stability duration. This definition is because if the change in position between adjacent successful attachment events is very small, it indicates that the target terminal has not truly traversed a long regional transition zone; simultaneously, if the duration of each successful attachment is very short, it indicates that the attachment may only be a temporary attachment due to local overlap or short-term jitter, rather than a stable migration. Only when both conditions are met simultaneously does it better conform to the characteristics of pseudo-stability.

[0167] In practice, all successful attachment events can be identified in the preceding and following state segments, and the start and end times of each successful attachment event, as well as the change in the target terminal's position during that event, can be recorded. Then, the trajectory advancement between two adjacent successful attachment events is calculated. If this advancement is less than a preset advancement threshold, and the duration of at least one of the two adjacent successful attachment events is less than a preset stabilization duration, then the adjacent event pair can be marked as a short-cycle repetitive attachment record. If multiple consecutive adjacent event pairs meet the above conditions, then these records can be marked as pseudo-stable records. The preset stabilization duration is used to distinguish between "short-term attachments that disappear immediately after establishment" and "real attachments that have been able to be maintained continuously."

[0168] For example, the preset stabilization time can be set to 3 to 20 seconds. For instance, in scenarios where the state sampling period is 1 second and access fluctuations are frequent near the boundary, 5 or 8 seconds can be used; in scenarios with high business continuity requirements and stricter exclusion of short-term attachments, 10 or 12 seconds can be used.

[0169] For example, the target terminal is a T-Box installed in a vehicle traveling across regions. The platform, in extracting preceding and following state segments around candidate temporal positions, detects that the vehicle exhibits a continuous movement pattern of "6 seconds of stillness – 7 meters of forward movement – ​​5 seconds of stillness – 6 meters of forward movement" within the last 40 seconds, and this pattern occurs more than twice. Therefore, the platform locally identifies this area's trajectory change as a queuing section. Further, the platform identifies three successful attachment events in the preceding and following state segments. The positional changes between two adjacent successful attachment events are 6 meters and 8 meters, respectively, both less than the currently set 10-meter advancement threshold. Furthermore, the durations of the three successful attachment events are 4 seconds, 5 seconds, and 6 seconds, respectively, all less than the currently set 8-second stability duration. At this point, the platform can mark the records corresponding to these successful attachment events as short-cycle repetitive attachment records, and further mark them as pseudo-stable records.

[0170] In this way, state samples that have successfully attached but have not formed a stable migration in queuing and slow-moving scenarios can be distinguished from ordinary stable samples.

[0171] This application does not simply treat all successful attachment events as positive evidence for improved handover stability. Instead, it first identifies the existence of queuing passage sections, and then further identifies short-period repeated attachment records composed of short-distance advances and short-term attachments, marking them as pseudo-stable records. The technical advantage of this approach is that it can eliminate or weaken pseudo-stable samples that might mislead stability determinations under passage conditions with obvious scene characteristics, such as border crossings and slow-moving queues, thus providing a prerequisite for more accurate calculation of handover stability.

[0172] In one implementation, after identifying pseudo-stable records, instead of directly using all state samples from the original preceding and following state segments to calculate the continuity of access identifiers and the consistency of attachment results, the contribution weight of the state samples corresponding to the pseudo-stable records in the calculation is reduced to obtain corrected continuity of access identifiers and consistency of attachment results. Based on the corrected results, the handover stability of the corresponding candidate profile is determined. The reason for not simply deleting pseudo-stable records is that, although pseudo-stable records in queuing scenarios are not equivalent to evidence of real stable migration, they may still contain certain effective trend information. For example, the target access direction reflected in a pseudo-stable record may not be entirely meaningless; its stability is simply significantly lower than that of real continuous attachment. Therefore, completely zeroing its contribution may result in the loss of useful local trend information; while reducing its contribution weight can retain certain trend information while avoiding an excessively positive amplification effect on stability determination.

[0173] In one implementation, the contribution weight refers to the relative proportion of a certain state sample or a certain type of state sample in the determination of the continuity of access identifiers and the consistency of attachment results. For ease of implementation, ordinary state samples can be assigned a base weight of 1, while state samples corresponding to pseudo-stable records can be assigned a reduced weight of less than 1. For example, the reduced weight of state samples corresponding to pseudo-stable records can be set to 0 to 0.6. For instance, in scenarios where strict suppression of queuing interference is required, a weight of 0 or 0.2 can be used; in scenarios where a small amount of trend information needs to be retained, a weight of 0.4 or 0.5 can be used. The above values ​​are not isolated values, but correspond to the intensity of queuing interference and the proportion of trend information to be retained.

[0174] In one implementation, when calculating access identifier continuity, state samples belonging to pseudo-stable records in the preceding and following state segments can be identified first, and then the contribution of these state samples to the continuity determination can be reduced. For example, when calculating whether the access identifier change near the boundary between preceding and following segments exhibits a unidirectional continuous transition, state samples belonging to pseudo-stable records are not counted as continuity support times as ordinary samples, but are counted with a reduced weight. Thus, if the access identifier continuity of a candidate profile is mainly contributed by short-cycle repeated attachments in queuing scenarios, its corrected access identifier continuity will be significantly reduced; conversely, if its continuity mainly comes from truly continuous access transitions, the impact of the corrected result will be smaller.

[0175] Furthermore, a similar approach can be used when calculating the consistency of attachment results. For example, when analyzing whether successful attachment states are maintained consistently in subsequent state segments and whether consecutive failure samples appear near the boundary, short-lived successful attachment samples belonging to pseudo-stable records are not counted in the same successful attachment duration as ordinary successful attachment samples, but are accumulated using a reduced weight. Thus, when the "successful attachment maintenance" of a candidate profile mainly consists of multiple short-lived repeated attachment events, the consistency of its corrected attachment results will no longer be unduly inflated by these pseudo-stable attachments.

[0176] In one implementation, the handover stability of the corresponding candidate profile can be re-determined based on the corrected access identifier continuity and the corrected attachment result consistency. For ease of implementation, the aforementioned hierarchical determination method can be used. For example, if both the corrected access identifier continuity and the corrected attachment result consistency are high, the handover stability can be determined as high; if one of them decreases to medium, the handover stability can decrease accordingly to medium; if the access identifier continuity or attachment result consistency decreases to low due to the reduction of pseudo-stable records, the handover stability also decreases accordingly to low. The technical significance of this is that the handover stability no longer only reflects the stability presented on the surface of the original state, but further reflects the true stability corresponding to the candidate profile after excluding the pseudo-stability amplification effect in queuing scenarios.

[0177] For example, using the aforementioned T-Box queuing scenario, the platform has identified three short-cycle repeated attachment records in the preceding and following state segments and marked them as pseudo-stable records. Then, when calculating access identifier continuity, the contribution weight of the state samples corresponding to these pseudo-stable records is reduced from 1 to 0.3; when calculating attachment result consistency, the attachment maintenance contribution corresponding to these pseudo-stable attachment samples is also accumulated only with a weight of 0.3. If a candidate profile, before correction, has both access identifier continuity and attachment result consistency initially judged as high, but after correction, one of them decreases to medium, its handover stability can be adjusted from high to medium; if both decrease, its handover stability can be adjusted from high to low. Thus, candidate profiles that initially exhibited high stability due to multiple short-term attachments in the queuing scenario will return to a stability level more consistent with the real scenario after correction.

[0178] In this way, in queuing scenarios with alternating characteristics of continuous dwell and short-distance advance, the artificially high impact of short-cycle repeated attachment records on the continuity of access identifiers and the consistency of attachment results can be reduced, thereby making the handover stability more accurately reflect the true migration stability of candidate profiles and further reducing the risk of misjudgment caused by pseudo-stable samples.

[0179] Based on the same inventive concept, this application also provides a remote configuration management system based on the embedded SIM specification, which corresponds to the remote configuration management method based on the embedded SIM specification. Since the principle of the system in this application is similar to the remote configuration management method based on the embedded SIM specification described above, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be described again.

[0180] Reference Figure 4 The diagram shown is a schematic of a remote configuration management system based on the embedded SIM specification provided in an embodiment of this application. The system includes:

[0181] The acquisition module 10 is used to acquire regional change information and network access status sequence of the target terminal within a preset time period;

[0182] The generation module 20 is used to generate at least one candidate switching prompt based on the regional change information and preset configuration rules. The candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file.

[0183] The processing module 30 is configured to extract local state segments corresponding to each candidate segment from the network access state sequence according to the candidate handover prompts, and determine the handover stability corresponding to each candidate configuration file based on the local state segments; determine the active candidate configuration file based on each handover stability, and update the triggering conditions for configuration state migration based on the active candidate configuration file.

[0184] Control module 40 is used to control the target terminal to migrate from the current active configuration file to the active candidate configuration file in response to the current active configuration file meeting the updated triggering condition.

[0185] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A remote configuration management method based on the embedded SIM specification, characterized in that, include: Acquire regional change information and network access status sequence of the target terminal within a preset time period; Based on the regional change information and preset configuration rules, at least one candidate switching prompt is generated. The candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file. Based on the candidate handover prompts, local state segments corresponding to each candidate segment are extracted from the network access state sequence, and the handover stability corresponding to each candidate configuration file is determined based on the local state segments. Based on the switching stability described above, an active candidate configuration file is determined, and the triggering conditions for configuration state migration are updated based on the active candidate configuration file. In response to the current active profile meeting the updated triggering conditions, the target terminal is controlled to migrate from the current active profile to the active candidate profile.

2. The remote configuration management method based on the embedded SIM specification according to claim 1, characterized in that, The generation of at least one candidate switching suggestion includes: Based on the regional change information, determine the regional change trajectory and extract the adjacent segments of the regional change trajectory. Based on the temporal position of each of the attribution change adjacent segments in the regional change trajectory and the preset configuration rules, candidate configuration files associated with each of the attribution change adjacent segments are determined, and candidate switching prompts are generated for each of the candidate configuration files.

3. The remote configuration management method based on the embedded SIM specification according to claim 1, characterized in that, Determining the switching stability of each candidate configuration file includes: Taking the timing position corresponding to each candidate switching prompt as the center, extract the preceding state segment and the following state segment from the network access state sequence; Determine the continuity of the access identifier and the consistency of the attachment result between the preceding state segment and the following state segment; Based on the continuity of the access identifier and the consistency of the attachment result, the handover stability of the corresponding candidate configuration file is determined.

4. The remote configuration management method based on the embedded SIM specification according to claim 1, characterized in that, The determination of active candidate configuration files includes: The switching stability of each candidate configuration file is determined at at least two different time scales; Active candidate configuration files are determined based on the cross-scale consistency of switching stability across different time scales.

5. The remote configuration management method based on the embedded SIM specification according to claim 1, characterized in that, The migration to the active candidate profile includes: In response to the fact that the active candidate configuration file is not stored on the target terminal, the target terminal is controlled to download the active candidate configuration file; After the active candidate configuration file is downloaded, the activation of the active candidate configuration file is delayed, and the network access status sequence corresponding to the target terminal is obtained again. The handover stability corresponding to the active candidate configuration file is updated based on the continuously acquired network access state sequence; In response to the updated switching stability continuously meeting the updated triggering conditions, the target terminal is controlled to activate the active candidate configuration file.

6. The remote configuration management method based on the embedded SIM specification according to claim 1, characterized in that, The migration to the active candidate profile includes: After the active candidate profile is activated, the currently active profile is kept in an inactive reserved state.

7. The remote configuration management method based on the embedded SIM specification according to claim 4, characterized in that, The triggering conditions for the configuration state migration include: The switching trigger timing and the original configuration retention duration in the triggering conditions are adjusted according to the cross-scale consistency.

8. The remote configuration management method based on the embedded SIM specification according to claim 6, characterized in that, The method further includes: Obtain the network access state sequence corresponding to the active candidate configuration file and extract it as a subsequent state segment; Determine the consistency of the attachment results and the continuity of the access identifiers corresponding to the subsequent state segments; In response to the continued satisfaction of the preset rollback release conditions for the consistency of the attachment results and the continuity of the access identifier, the current active configuration file is adjusted from the inactive retained state to the non-rollback state.

9. The remote configuration management method based on the embedded SIM specification according to claim 3, characterized in that, Also includes: Based on the regional change information, determine whether there are queuing passage sections in the regional change trajectories corresponding to the preceding state segment and / or the following state segment where continuous dwell state and short-distance advance state alternate; In response to the existence of the queuing passage section, short-period repetitive attachment records in the preceding state segment and the following state segment, where the position change between adjacent successful attachment events is less than a preset advancement threshold and the duration of successful attachment is less than a preset stable duration, are identified and marked as pseudo-stable records.

10. The remote configuration management method based on the embedded SIM specification according to claim 9, characterized in that, The determination of the switching stability of the corresponding candidate configuration file includes: The contribution weight of the state sample corresponding to the pseudo-stable record in calculating the continuity of the access identifier and the consistency of the attachment result is reduced to obtain the corrected continuity of the access identifier and the corrected consistency of the attachment result. Based on the continuity of the corrected access identifier and the consistency of the corrected attachment result, the handover stability of the corresponding candidate configuration file is determined.

11. A remote configuration management system based on the embedded SIM specification, used to implement the remote configuration management method based on the embedded SIM specification as described in any one of claims 1-10, characterized in that, include: The acquisition module is used to obtain information on regional changes and network access status sequences of the target terminal within a preset time period; The generation module is used to generate at least one candidate switching prompt based on the regional change information and preset configuration rules. The candidate switching prompt represents the configuration state migration of the candidate segment and corresponds to the candidate configuration file. The processing module is configured to extract local state segments corresponding to each candidate segment from the network access state sequence according to the candidate handover prompts, and determine the handover stability corresponding to each candidate configuration file based on the local state segments; determine the active candidate configuration file based on each handover stability, and update the triggering conditions for configuration state migration based on the active candidate configuration file; The control module is used to control the target terminal to migrate from the current active configuration file to the active candidate configuration file in response to the updated triggering condition of the current active configuration file.

Images