eSIM automated network switching strategy management method based on multi-source data fusion

Abstract

The present invention discloses an eSIM automated network switching strategy management system and method based on multi-source data fusion, which relates to the field of network switching management. The main scheme is: by analyzing network status data, calculating basic performance indicators of network status, and judging whether the network to be accessed meets the basic performance evaluation standards; when the network to be accessed meets the basic performance evaluation standards, obtaining network security data, and analyzing the network security data to judge whether the network security indicators meet the network security standards; when all network security indicators meet the network security standards, executing the network switching instruction; solving the problem that the existing technology only focuses on network performance indicators and ignores network security, and the device switches to a network that seems to have good performance but has security risks, such as a fake network maliciously built by an attacker or a network with vulnerabilities, thereby increasing the risk of data leakage, network attacks, etc., and bringing serious privacy and security problems to users.

Description

Technical Field

[0001] The present invention relates to the technical field of network switching management, and in particular to an eSIM automated network switching strategy management method based on multi-source data fusion. Background Art

[0002] With the rapid development of the Internet of Things (IoT), eSIM technology, as one of its key supporting technologies, is being widely adopted in a wide range of fields, including smartphones, smartwatches, smart cars, and industrial IoT devices. Market research firms predict that the global eSIM market is expected to continue to grow in size, and its application prospects in the field of network connectivity are extremely broad.

[0003] In the consumer electronics sector, consumers are increasingly demanding seamless network connectivity. They expect their devices to automatically switch to the optimal network based on network conditions, ensuring smooth browsing, calls, video playback, and more. For example, when users move between different carrier coverage areas, the eSIM's automated network switching feature allows devices to automatically switch to a stronger, faster network, enhancing the user experience. In the Industrial Internet of Things (IIoT), numerous devices require stable and reliable network connections to transmit data and receive commands. Automated network switching technology ensures that these devices, even in complex and changing network environments, can promptly switch to the best network quality, ensuring production continuity and efficiency while mitigating the risk of production downtime and financial losses caused by network failures.

[0004] Currently, most automated eSIM network switching solutions focus solely on network performance metrics, such as signal strength and speed, to determine network switching decisions, while insufficiently considering network security. This can result in devices switching to seemingly healthy networks that harbor security risks, such as those maliciously set up by attackers or networks with vulnerabilities. This increases the risk of data leakage and cyberattacks, posing serious privacy and security risks to users. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides an eSIM automated network switching policy management method based on multi-source data fusion, which at least solves the problem that the existing technology only focuses on network performance indicators and ignores network security, thereby increasing the risk of data leakage and network attacks, and posing serious security risks to users.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: an eSIM automated network switching strategy management method based on multi-source data fusion includes:

[0009] Step 1: Obtain real-time network status data and calculate basic performance indicators of the network status by analyzing the network status data;

[0010] Step 2: Evaluate basic performance indicators to determine whether the network to be connected meets the basic performance evaluation standards;

[0011] Step 3: When the network to be accessed meets the basic performance evaluation criteria, obtain network security data, analyze the network security data, and calculate network security indicators;

[0012] Step 4: Determine whether the network security indicators meet the network security standards; if the network security indicators meet the network security standards, execute the network switching instruction;

[0013] Step 5: When the network to be accessed meets the basic performance evaluation standards but does not meet the network security standards, if an emergency security warning instruction is received, the exemption instruction can be temporarily activated, the evaluation results of the network security evaluation can be ignored, and the instruction to switch networks can be executed. After the emergency security warning instruction is eliminated, the network evaluation strategy of steps 1 to 4 can be referred to to determine whether to switch to other networks.

[0014] In the preferred solution of the above-mentioned eSIM automatic network switching policy management method based on multi-source data fusion, the basic performance indicators include the network quality assessment index, the roaming cost index and the reference signal theoretical received power.

[0015] In the preferred solution of the above-mentioned eSIM automated network switching policy management method based on multi-source data fusion, the method for obtaining the network quality evaluation index is:

[0016] The measurement configuration is obtained through the measurement control message in the system information block (SIB) broadcast by the base station, and then the theoretical received power of the reference signal is measured. After receiving the signal from the base station, the signal-to-noise ratio is calculated based on the received useful signal power as well as the interference and noise floor power. In wireless networks, the round-trip delay is measured by measuring the round-trip time of RRC signaling or the round-trip time of the TCP transmission control protocol connection.

[0017] The theoretical received power, signal-to-noise ratio, and round-trip delay of the reference signal are normalized and preprocessed to obtain the received signal score, signal-to-noise ratio score, and round-trip delay score, respectively. A comprehensive analysis is then performed using the network quality assessment value index calculation model to obtain the network quality assessment index.

[0018] In the preferred solution of the above-mentioned eSIM automatic network switching policy management method based on multi-source data fusion, the method for obtaining the roaming cost index is:

[0019] By connecting to the billing system of network operators through API, users' real-time tariff values ​​in different network environments can be obtained in real time. By querying the tax database of the target country or region, the applicable tax rate value of communication services can be obtained. The service contract signed between the enterprise and the operator can be obtained through the enterprise's contract management system, and the benchmark tariff value can be extracted from the service contract. The real-time tariff value, applicable tax rate value and benchmark tariff value are normalized and pre-processed, and a comprehensive analysis is performed through the roaming cost calculation model to obtain the roaming cost index.

[0020] In the preferred embodiment of the eSIM automated network switching strategy management method based on multi-source data fusion, the method for obtaining the theoretical received power of the reference signal is as follows:

[0021] Use the meteorological API to obtain the real-time rainfall in the area and the altitude difference between the terminal and the base station. Obtain the material attenuation coefficient of the obstructions or obstacles in the area. Normalize the real-time rainfall, altitude difference, and material attenuation coefficient, and analyze and calculate the environmental compensation factor. The formula is as follows:

[0022] ;

[0023] Where: HBC represents the environmental compensation factor; JY represents the real-time rainfall in the area; HB represents the altitude difference between the area and the base station; JZ j represents the material attenuation coefficient of the jth obstruction in the area; kr is the rain attenuation coefficient, which represents the dynamic influence weight of rainfall on signal attenuation; kt is the terrain coefficient, which represents the nonlinear compensation factor of terrain undulation on signal attenuation; β j Represents the weight of different building materials in the environment, and the calculation formula is: β j = Material attenuation coefficient of the jth obstruction / the sum of the material attenuation coefficients of all obstructions; j represents the serial number of the obstruction, which is a positive integer;

[0024] The theoretical received power of the reference signal is calculated based on the original signal strength and the environmental compensation factor. Specifically, the theoretical received power of the reference signal = original signal strength - environmental compensation factor.

[0025] In the preferred embodiment of the eSIM automated network switching policy management method based on multi-source data fusion, a comprehensive evaluation is performed on the network quality evaluation index, roaming cost index, and theoretical received power of the reference signal to determine whether the basic performance evaluation standards are met. The specific method is as follows:

[0026] Set the network quality index threshold, roaming cost index threshold, and receiving power threshold; compare the network quality assessment index, roaming cost index, and environmental compensation factor with the network quality index threshold, roaming cost index threshold, and environmental status threshold, respectively. When the network quality assessment index is greater than or equal to the network quality index threshold, the indicator is considered qualified; when the roaming cost index is less than or equal to the roaming cost index threshold, the indicator is considered qualified; when the reference signal theoretical receiving power is greater than or equal to the receiving power threshold, the indicator is considered qualified;

[0027] When all three indicators are qualified, the network to be accessed is determined to meet the basic performance evaluation standards.

[0028] In the preferred embodiment of the eSIM automatic network switching strategy management method based on multi-source data fusion, the network security data includes signal spatiotemporal conflict index, Doppler frequency shift deviation, and behavior pattern entropy.

[0029] In the preferred embodiment of the eSIM automated network switching strategy management method based on multi-source data fusion, the method for obtaining the signal spatiotemporal conflict index is as follows:

[0030] The terminal's GPS positioning coordinates are obtained through the terminal's built-in GNSS module, and the base station's registered geographic location coordinates are queried through the core network MME database. The terminal's moving speed is obtained through the GNSS Doppler frequency shift or inertial navigation system, and the base station's location update interval is obtained through the base station's broadcast system information block SIB. The base station authentication level value of the operator's core network for the base station's identity is obtained. The terminal's GPS positioning coordinates, the base station's registered geographic location coordinates, the terminal's moving speed, the base station's location update interval, and the base station authentication level value are normalized and pre-processed, and the signal spatiotemporal conflict index is calculated based on the following formula:

[0031] ;

[0032] Among them, STC represents the signal space-time conflict index; L GPS Indicates the GPS positioning coordinates of the terminal; L CEL Indicates the geographical location coordinates of the base station registration; Indicates the base station location update interval; v indicates the terminal's moving speed; ZHD indicates the base station certification level value;

[0033] The baseband processor uses a phase-locked loop (PLL) to measure the carrier frequency offset value actually measured by the terminal in real time. The carrier center frequency is obtained from the carrier frequency parameters broadcast by the base station, and the Doppler shift value is calculated based on the carrier center frequency. The formula is: , in the formula f carepresents the Doppler frequency shift value, f0 represents the carrier center frequency, v represents the terminal's moving speed, θ represents the angle between the base station's moving direction and the base station, and c represents the speed of light. The terminal's built-in magnetometer and accelerometer are used to obtain the angle between the moving direction and the base station's connection line, and the cosine value of the angle is calculated as the moving direction weight. The carrier frequency offset value, carrier center frequency, Doppler frequency shift value, and moving direction weight are normalized and preprocessed, and the processed parameters are input into the Doppler frequency shift deviation calculation model to calculate the Doppler frequency shift deviation.

[0034] The proportion of each signaling type within a unit period is counted to obtain the probability of occurrence of the kth type of signaling within the time window; the federated learning weight coefficient is dynamically adjusted based on the number of terminals through the cloud; the federated learning server API is called, and the local entropy value and time series characteristics are input to obtain the anomaly probability as the federated learning anomaly score; the local entropy value is calculated as follows: ;

[0035] The occurrence probability of the k-th type of signaling within the time window, the federated learning weight coefficient and the abnormal probability are normalized and preprocessed, and the behavior pattern entropy is calculated according to the behavior pattern entropy calculation model.

[0036] In the preferred embodiment of the eSIM automated network switching policy management method based on multi-source data fusion, the method for determining whether the network security indicators meet the network security standards is as follows:

[0037] The signal space-time conflict threshold, Doppler frequency shift deviation threshold and behavior pattern entropy threshold are set respectively; the signal space-time conflict index, Doppler frequency shift deviation and behavior pattern entropy are compared with the corresponding thresholds respectively; when the signal space-time conflict index is less than the signal space-time conflict threshold, this indicator meets the safety standard; when the Doppler frequency shift deviation is less than the Doppler frequency shift deviation threshold, this indicator meets the safety standard; when the behavior pattern entropy threshold is less than the behavior pattern entropy threshold, this indicator meets the safety standard.

[0038] (3) Beneficial effects

[0039] The present invention provides an eSIM automated network switching strategy management method based on multi-source data fusion, which has the following beneficial effects:

[0040] (1) Steps 1 and 2 together constitute the technical points for a comprehensive evaluation of the basic performance of the network. By obtaining real-time network status data and calculating basic performance indicators, it is determined whether the network to be accessed meets the basic performance evaluation standards. This process can accurately screen out networks that meet the performance standards. Compared with existing methods that only make judgments based on a single or a few network parameters, this solution combines multiple real-time network status data to determine basic performance, greatly enhancing the comprehensiveness and accuracy of network performance evaluation, and effectively avoiding the misjudgment of poor access network performance due to only referring to partial data, thereby laying a solid foundation for subsequent more in-depth network selection and switching decisions, and ensuring that user equipment can give priority to accessing networks with truly reliable performance.

[0041] (2) Steps 3 and 4 focus on the assessment of network security. On the premise that the network to be accessed meets the basic performance assessment standards, further network security data is obtained and network security indicators are calculated to determine whether it meets the network security standards. This technical point realizes a comprehensive and in-depth consideration of the network. Existing technologies often have a separation in performance and security assessment, or pay insufficient attention to security factors. This solution organically combines the two to ensure that even if the network performance is excellent, it will not be switched rashly if there are security risks. The network switching command will only be executed when the network security indicators also meet the standards. This effectively improves the overall quality and reliability of the device access to the network, greatly reduces the probability of user devices being subject to network attacks, data leaks and other security risks, and ensures the information security and system stability of users during network use.

[0042] (3) The emergency security warning and exemption command mechanism involved in step 5 is that when an emergency security warning is received, the exemption command can be temporarily activated to ignore the network security assessment results and execute the network switching command. After the emergency security warning is eliminated, the regular network assessment strategy is used to determine whether to switch to another network. This mechanism shows extremely high flexibility and practicality in the face of sudden network security threats or special emergency business needs.

[0043] (4) The entire solution, from step one to step five, systematically constructs a set of eSIM network switching strategy management processes based on multi-source data fusion. Through the fusion and application of multi-source data, it not only comprehensively considers the two key dimensions of network performance and security, but also introduces an emergency response mechanism, so that network switching decisions are no longer mechanical and single judgments, but can make flexible and intelligent responses based on real-time and changing network conditions and special emergency needs. Compared with the traditional more rigid and single-factor eSIM automatic network switching solution, this solution greatly optimizes the intelligence of network switching, enabling devices to more accurately match the network that best suits current actual usage needs. Whether it is pursuing smooth and stable daily use or ensuring basic connection in emergency situations, it has a usage strategy and is more practical. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Schematic diagram of the steps of the eSIM automatic network switching strategy management method based on multi-source data fusion of the present invention. DETAILED DESCRIPTION

[0045] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] Example 1

[0047] See also Figure 1 The present invention provides an eSIM automated network switching strategy management method based on multi-source data fusion, including:

[0048] Step 1: Obtain real-time network status data, analyze the network status data, and calculate basic performance indicators of the network status.

[0049] Step 101: Obtain the measurement configuration through the measurement control message in the system information block (SIB) broadcast by the base station, and then measure the theoretical received power of the reference signal. After receiving the signal from the base station, calculate the signal-to-noise ratio based on the received useful signal power as well as the interference and noise floor power. In a wireless network, the round-trip delay can be estimated by measuring the round-trip time of the RRC signaling or the round-trip time of the TCP transmission control protocol connection.

[0050] It should be noted that the specific method for measuring the theoretical received power of the reference signal according to the measurement configuration is: based on the information in the measurement configuration, the reference signal is received at the specified SSB time-frequency position, and then the energy of the CRS / SSB resource unit is integrated to calculate the theoretical received power RSR of the reference signal. The formula is as follows:

[0051] ;

[0052] Among them, r s It represents the received amplitude of the sth reference signal, s represents the sequence number of the reference signal, which is a positive integer, and S represents the number of reference signals.

[0053] This formula calculates a comprehensive theoretical received power of a reference signal by integrating the energy of the received amplitudes of multiple reference signals. s The RSR value represents the received amplitude of the sth reference signal. The RSR value is obtained by summing the squared amplitudes of all reference signals, averaging them, and performing a logarithmic transformation. This provides a more accurate received signal strength indicator for network quality assessment, enabling more reliable judgment of network reception quality in subsequent network handover decisions. This avoids problems such as frequent handovers or switching to networks with poor signal quality due to inaccurate signal strength assessments, and avoids the problem of assessing network reception quality based solely on the strength of a single reference signal or partial signal characteristics, which inaccurately reflects the overall received power. By integrating multiple reference signals and employing an energy integration method, errors caused by signal fluctuations and interference are effectively reduced, improving the accuracy and stability of received power assessments.

[0054] It should be noted that the signal-to-noise ratio (SIN) is calculated based on the received useful signal power, interference, and noise floor power. The formula is:

[0055] ;

[0056] Among them, P singnal Indicates the useful signal power; P noise Represents the noise floor power; P interference Indicates the co-channel / adjacent-channel interference power.

[0057] The formula calculates the signal-to-noise ratio (SNR) based on the ratio of the received useful signal power to the sum of the interference and noise floor powers. This provides an accurate indicator of signal quality and a more reliable basis for network switching decisions. When switching networks, network selection can be based on the SNR, thereby improving communication stability and reliability and reducing communication quality issues such as bit error rates. This avoids the problem of relying solely on signal strength to assess network quality, which fails to effectively reflect signal quality. Even with high signal strength, if interference and noise floor power are also high, actual communication quality may still be poor. This solution comprehensively assesses signal quality by comprehensively considering useful signal, interference, and noise floor power, addressing the one-sided nature of traditional signal quality assessments.

[0058] In step 102, the theoretical received power, signal-to-noise ratio, and round-trip delay of the reference signal are normalized and preprocessed to obtain a received signal score, a signal-to-noise ratio score, and a round-trip delay score, respectively. A comprehensive analysis is then performed using a network quality evaluation index calculation model to obtain a network quality evaluation index, based on the following formula:

[0059] ;

[0060] Among them, NQI represents the network quality assessment index; RSR represents the received signal score; SIN represents the signal-to-noise ratio score; RTT represents the round-trip delay score; α1 represents the weight coefficient of the received signal score; α2 represents the signal-to-noise ratio score; α3 represents the weight coefficient of the round-trip delay score. It can be adjusted according to user needs, and α1+α2+α3=1. The values ​​here can be: α1=0.3, α2=0.3, α3=0.4.

[0061] This formula normalizes and pre-processes three key network metrics: theoretical reference signal received power, signal-to-noise ratio, and round-trip delay. This formula calculates the network quality assessment index (NQI). This provides a more comprehensive and accurate quantitative assessment of network quality, providing a comprehensive quantitative basis for network handover decisions. During the automated eSIM network handover process, this index enables a more scientific and rational selection of the optimal network, thereby improving the success rate of network handovers, communication quality, and user experience.

[0062] Step 103: By connecting to the billing system of the network operator through an API, the real-time tariff value of the user in different network environments is obtained in real time; by querying the tax database of the target country or region, the applicable tax rate value of the communication service is obtained; the service contract signed between the enterprise and the operator is obtained through the enterprise's contract management system, and the benchmark tariff value is extracted from the service contract.

[0063] Step 104: normalize and pre-process the real-time tariff value, applicable tax rate value, and base tariff value, and perform comprehensive analysis using a roaming cost calculation model to obtain a roaming cost index. The calculation formula is as follows:

[0064] ;

[0065] Among them, RCL represents the roaming cost index; SZ represents the real-time tariff value; SL represents the applicable tax rate; and QJ represents the benchmark tariff value.

[0066] By connecting to the billing system of network operators through APIs, users' tariff information in different network environments can be obtained in real time. At the same time, the applicable tax rate is obtained by combining the tax database of the target country or region, and the benchmark tariff is extracted from the service contract signed between the enterprise and the operator. These tariff-related data are normalized and pre-processed for subsequent comprehensive analysis; this provides a reliable data basis for the accurate calculation of subsequent roaming costs, so that the roaming cost index can more truly reflect the user's actual roaming costs, which is conducive to fully considering cost factors when performing eSIM network switching, selecting more cost-effective networks, and reducing users' communication costs; traditional methods often find it difficult to obtain tariff information in different network environments in real time and accurately, and the assessment of tariffs may be relatively lagging or inaccurate, and cannot fully consider the impact of factors such as tax rates and service contracts on actual tariffs. This solution solves the problem of untimely and incomplete acquisition of tariff information through API connection and multi-source data fusion, and improves the accuracy of tariff assessment.

[0067] Step 105: Obtain the real-time rainfall in the area through the meteorological API, and obtain the altitude difference between the terminal and the base station; obtain the material attenuation coefficient of the obstructions or obstacles in the area.

[0068] It should be noted that the material attenuation coefficients of different materials are obtained based on known public data such as the city-level BIM database, and the building information in the area is obtained, and different material attenuation coefficients are corresponding to the type of building; for example, the value of reinforced concrete wall can be 0.8-1.2, and the value of double-layer insulating glass wall can be 0.3-0.5, etc.

[0069] Step 106: Normalize the real-time rainfall, altitude difference, and material attenuation coefficient, and obtain the environmental compensation factor after analysis and calculation. The calculation formula is as follows:

[0070] ;

[0071] Where: HBC represents the environmental compensation factor; JY represents the real-time rainfall in the area; HB represents the altitude difference between the area and the base station; JZ jrepresents the material attenuation coefficient of the jth obstruction in the area; kr is the rain attenuation coefficient, which represents the dynamic influence weight of rainfall on signal attenuation; kt is the terrain coefficient, which represents the nonlinear compensation factor of terrain undulation on signal attenuation; β j Represents the weight of different building materials in the environment, and the calculation formula is: β j = Material attenuation coefficient of the jth obstruction / sum of the material attenuation coefficients of all obstructions; j represents the serial number of the obstruction, which is a positive integer.

[0072] During rainfall, raindrops absorb and scatter electromagnetic wave signals. As electromagnetic wave signals propagate through the atmosphere, water molecules in raindrops absorb signal energy, weakening signal strength. Generally speaking, the greater the rainfall, the greater the number of raindrops per unit volume, and the greater the water content, the greater the signal attenuation. Large altitude differences typically increase the signal propagation path and may increase the number of times a signal passes through different atmospheric layers, exacerbating atmospheric refraction and scattering, further increasing signal attenuation. Different building materials have different signal attenuation capabilities. The thickness, density, and electromagnetic properties of obstructions all affect the material's attenuation coefficient. Thicker, denser materials typically have larger attenuation coefficients and attenuate signals more severely. By calculating an environmental compensation factor and applying it to the theoretical received power of a reference signal, a more accurate prediction of the actual signal strength received by a terminal can be achieved. Considering the environmental compensation factor during communication network planning can more accurately assess signal coverage in different areas.

[0073] The real-time rainfall, altitude difference, and material attenuation coefficient are normalized to eliminate the influence of different dimensions and data ranges. By calculating the environmental compensation factor, the impact of environmental factors on the signal reception power can be quantified, providing an accurate compensation basis for the subsequent calculation of the theoretical reception power of the reference signal, thereby more accurately predicting the actual signal reception situation and improving the scientific nature of network switching decisions.

[0074] It should be noted that the rain attenuation coefficient can be dynamically predicted by the LSTM model and can take a value of 0.5-1.5. The calculation formula based on the LSTM model is:

[0075] ;

[0076] in, is the measured received power of the reference signal of the i-th historical sample; The reference signal predicted received power of the i-th historical sample; R represents the actual rainfall of the i-th historical sample; and the constraint is kr∈[0.5, 1.5], i represents the sequence number of the historical sample, and the value is a positive integer.

[0077] The LSTM model dynamically predicts the rain attenuation coefficient. The model is trained using historical sample data (including the measured and predicted received power of the reference signal, as well as actual rainfall) to determine the optimal rain attenuation coefficient. This improves the accuracy of signal attenuation assessment in rainy environments, allowing the environmental compensation factor to more accurately reflect the impact of rainfall on the signal, further enhancing the accuracy of the theoretical received power calculation of the reference signal. The impact of rainfall on signal attenuation is dynamic, and traditional fixed rain attenuation coefficients cannot accurately reflect this change. This solution leverages the dynamic prediction capabilities of the LSTM model to adjust the rain attenuation coefficient in real time, resolving the issue of inaccurate signal attenuation assessment caused by a fixed rain attenuation coefficient.

[0078] It should be noted that before calculating the formula, the parameters need to be normalized and preprocessed. The calculation formula is: Where Ptx represents the base station's transmit power; PL0 represents the reference path loss; γ represents the path loss exponent, which can be set according to industry standards, such as 3.7 for urban areas and 3.0 for suburban areas; d represents the actual physical distance between the receiver and transmitter; and d0 represents the reference distance between the receiver and transmitter, providing a standardized path loss calculation benchmark, usually taking the typical distance of free space propagation. ant Indicates the directional gain of the antenna in dBi, which is obtained based on the antenna performance parameter description.

[0079] It should be noted that the terrain coefficient is obtained through the regional optimization method under the federated learning framework, based on the following method:

[0080] To calculate the gradient of the terrain coefficient through the terminal device, it is necessary to preset the kt value according to the actual situation. For example, the initial setting for plain terrain is 0.8; the initial setting for hilly terrain is 1.2; the initial setting for mountainous terrain is 1.5, etc. The formula is as follows:

[0081] ;

[0082] in, represents the gradient of the mountain coefficient, ‖(RSRP 实测 -RSRP 预测 )‖ 2 Indicates that RSRP 实测 and RSRP 预测 Perform square error operation; It is a partial integral operation.

[0083] The terrain coefficient is calculated from the gradient of the terrain coefficient according to the following formula:

[0084]

[0085] Among them, kt old represents the terrain coefficient before updating; σ represents the learning rate, which is in the range of [0.001, 0.1] and can be selected based on historical experience, such as city / plain: σ=0.02, mountain / hill: σ=0.02, and so on.

[0086] By using a regional optimization method within a federated learning framework to obtain the terrain coefficient, the terrain coefficient better reflects the actual terrain's impact on the signal, improving the accuracy of the environmental compensation factor calculation. This in turn enhances the accuracy of the prediction of the theoretical received power of the reference signal, making network switching decisions more accurate and effective under varying terrain conditions. Because different terrains have varying effects on signal attenuation, traditional fixed terrain coefficients are unable to adapt to various complex terrain environments. This solution dynamically optimizes the terrain coefficient using a gradient descent method, addressing the issue of inaccurate signal attenuation assessments caused by a fixed terrain coefficient.

[0087] Step 107: Calculate the theoretical received power of the reference signal according to the environmental compensation factor, specifically: theoretical received power of the reference signal = original signal strength - environmental compensation factor.

[0088] Calculating the theoretical reference signal received power based on the calculated environmental compensation factor more accurately predicts the actual theoretical reference signal received power, providing a more realistic and reliable signal strength assessment metric for eSIM automated network switching strategies. During network switching, more accurate signal strength information enables more informed network selection decisions, improving network switching success rates, communication quality, and user experience. Traditional methods often fail to fully account for environmental factors when calculating the theoretical reference signal received power, resulting in discrepancies between the calculated results and the actual received power. This solution introduces an environmental compensation factor to correct the original signal strength, addressing the issue of inaccurate theoretical reference signal received power calculations due to environmental factors.

[0089] Step 2: Evaluate the basic performance indicators to determine whether the network to be accessed meets the basic performance evaluation standards.

[0090] Comprehensively evaluate the network quality assessment index, roaming cost index, and theoretical received power of the reference signal to determine whether the network meets the basic performance evaluation standards. The specific method is as follows:

[0091] Set the network quality index threshold, roaming cost index threshold, and receiving power threshold; compare the network quality assessment index, roaming cost index, and environmental compensation factor with the network quality index threshold, roaming cost index threshold, and environmental status threshold respectively. When the network quality assessment index is greater than or equal to the network quality index threshold, the indicator is considered qualified; when the roaming cost index is ≤ the roaming cost index threshold, the indicator is considered qualified; when the reference signal theoretical receiving power is ≥ the receiving power threshold, the indicator is considered qualified.

[0092] When all three indicators are qualified, the network to be accessed is determined to meet the basic performance evaluation standards.

[0093] It should be noted that the network quality index threshold refers to relevant standards and specifications for network quality in the communications industry. Alternatively, data on network quality assessment indexes across different scenarios and time periods can be collected, and the distribution and user feedback on network quality can be analyzed to determine a threshold that ensures a basic network experience. For example, the network quality assessment index over the past month can be calculated and the average or a specific percentile (such as the 90th percentile) can be used as the threshold. The roaming cost index threshold can be determined based on the enterprise's cost control objectives and user willingness to pay by analyzing the expected benefits and costs under different roaming costs. The receive power threshold can be set based on the signal receive power requirements of the adopted communications standards and protocols. Communications protocols typically specify minimum receive signal power in different scenarios to ensure communication reliability and compatibility.

[0094] Compliance with basic performance evaluation standards is determined by comprehensively evaluating the network quality assessment index, roaming cost index, and theoretical received power of the reference signal. This approach involves setting corresponding thresholds and comparing each indicator with the thresholds. When all three indicators meet the criteria, the network to be accessed is deemed to meet the basic performance evaluation standards. This approach more accurately assesses the overall performance of the network, providing a comprehensive and reliable evaluation result for network switching decisions, making them more scientific and reasonable, and improving user experience and network reliability. Traditional network evaluation methods often focus on a single indicator and fail to comprehensively consider the impact of multiple factors on network performance. This solution addresses the issue of incomplete network performance evaluation in existing technologies by constructing a comprehensive evaluation system, thereby improving the accuracy and scientific nature of network performance evaluation.

[0095] Step 3: When the network to be accessed meets the basic performance evaluation criteria, obtain network security data, analyze the network security data, and calculate network security indicators.

[0096] Step 301: Obtain the GPS positioning coordinates of the terminal through the built-in GNSS module of the terminal, and query the geographical location coordinates registered by the base station through the core network MME database; obtain the moving speed of the terminal through the GNSS Doppler frequency shift or inertial navigation system, and obtain the base station location update time interval through the base station broadcast system information block SIB; obtain the base station authentication level of the operator core network for the base station identity, 0 means unauthenticated, and 1 means fully authenticated.

[0097] Step 302: Normalize and pre-process the GPS positioning coordinates of the terminal, the geographical location coordinates registered by the base station, the moving speed of the terminal, the base station location update time interval, and the base station authentication level value; and calculate the signal spatiotemporal conflict index based on the pre-processed data, based on the following formula:

[0098] ;

[0099] Among them, STC represents the signal space-time conflict index; L GPS Indicates the GPS positioning coordinates of the terminal; L CEL Indicates the geographical location coordinates of the base station registration; Indicates the base station location update interval; v indicates the terminal's moving speed; ZHD indicates the base station authentication level value, which can be 0 or 1.

[0100] The formula comprehensively assesses the spatiotemporal signal conflict by calculating the square of the Euclidean distance between the terminal's GPS positioning coordinates and the base station's registered geographic coordinates, dividing the square by the product of the terminal's mobile speed and the base station's location update interval, and finally multiplying the square by the base station's certification level. This allows for a more accurate assessment of the spatiotemporal conflict risk during signal transmission, providing a quantitative basis for the calculation of network security indicators. During network switching, this index can be used to more accurately determine the security of the target network, prioritizing networks with low spatiotemporal conflict risk, improving overall network security and reducing the likelihood of user devices being vulnerable to security threats such as malicious attacks or data tampering. Traditional network security assessment methods struggle to fully consider the impact of multiple factors, such as terminal location, base station location updates, and authentication, on network security. This can lead to inaccurate and inaccurate network security assessments and an inability to effectively identify potential security risks. By integrating this multi-source data, this solution addresses the issue of incomplete and inaccurate network security assessments in existing technologies and improves the ability to identify network security threats.

[0101] Step 303: Use the baseband processor to measure the carrier frequency offset value actually measured by the terminal in real time through the phase-locked loop (PLL); obtain the carrier center frequency from the carrier frequency parameters broadcast by the base station, and calculate the Doppler frequency shift value based on the carrier center frequency; use the terminal's built-in magnetometer and accelerometer to obtain the angle area between the motion direction and the base station connection line, and calculate the cosine value of the angle as the motion direction weight.

[0102] Step 304: normalize and pre-process the carrier frequency offset value, carrier center frequency, Doppler frequency shift value, and motion direction weight, and input the processed parameters into a Doppler frequency shift deviation calculation model to calculate the Doppler frequency shift deviation. The formula based on the Doppler frequency shift deviation calculation model is as follows:

[0103] ;

[0104] Where DSD represents the Doppler shift deviation, f ob Indicates the carrier frequency offset value; f ca represents the Doppler frequency shift value; f0 represents the carrier center frequency, and YD represents the motion direction weight;

[0105] The carrier frequency offset value is calculated based on the carrier center frequency according to the following formula: , where θ represents the angle between the base station's moving direction and the base station, and c represents the speed of light.

[0106] Doppler shift deviation is comprehensively assessed by calculating the difference between the terminal's actual carrier frequency offset and the Doppler shift value calculated based on the carrier center frequency, dividing the difference by the carrier center frequency, and multiplying the result by the motion direction weight. This allows for more accurate calculation of Doppler shift deviation, providing a more precise quantitative basis for calculating network security indicators. During network handover, this deviation can be used to more accurately determine the signal stability of the target network, prioritizing networks with smaller Doppler shift deviations. This improves overall network stability, reduces signal demodulation errors and communication quality degradation caused by larger Doppler shift deviations, and enhances user experience. Traditional methods for Doppler shift assessment often ignore the impact of the terminal's motion direction on the frequency shift, resulting in inaccurate assessment results. Furthermore, in complex environments, the relative speed between the terminal and the base station varies rapidly, making traditional fixed-parameter calculations difficult to adapt. This solution, by integrating multiple data sources including carrier frequency offset, Doppler shift, and motion direction weights, addresses the existing issues of inaccurate Doppler shift deviation assessment and insufficient consideration of the impact of motion direction, thereby improving the accuracy and reliability of Doppler shift deviation assessment.

[0107] Step 305: Count the proportions of each signaling type in the last hour to obtain the probability of occurrence of the kth type of signaling within the time window; dynamically adjust the federated learning weight coefficient based on the number of terminals through the cloud; call the federated learning server API, input the local entropy value and time series characteristics, and obtain the anomaly probability, which is set to [0, 1], as the federated learning anomaly score; the local entropy value is calculated as follows: .

[0108] Traditional methods struggle to accurately identify abnormal behavior in real time, especially in complex network environments where it can be concealed and prone to missed or false positives. This solution leverages federated learning technology to integrate local entropy and time series features, addressing the inaccurate identification of abnormal behavior in existing technologies and improving both accuracy and timeliness.

[0109] Step 306: The occurrence probability of the k-th type of signaling within the time window, the federated learning weight coefficient, and the abnormal probability are normalized and preprocessed, and then the behavior pattern entropy is calculated using the behavior pattern entropy calculation model. The formula is as follows:

[0110] ;

[0111] Among them, BPE represents behavior pattern entropy; p k represents the probability of occurrence of the kth type of signaling within the time window; k represents the serial number of the signaling type, which is a positive integer; ρ represents the federated learning weight coefficient, which is calculated as follows: ρ=0.5+0.1*log(N 终端 ), in the formula, N 终端 represents the number of terminals; YC represents the federated learning anomaly score, which takes a value of [0, 1].

[0112] In different network environments, the number of terminals varies significantly. Traditional fixed weighting coefficients cannot adapt to these variations, resulting in reduced adaptability and accuracy of network security assessment results. Dynamically adjusting the weighting coefficients overcomes the fixed weighting coefficient issue in existing technologies, improving the flexibility and adaptability of network security assessments. This allows network security assessments to better reflect the impact of terminal scale in actual network environments, enhancing the reliability and accuracy of assessment results. In networks with a large number of terminals, potential security threats can be more effectively identified, ensuring network security and stability.

[0113] Traditional network security assessment methods primarily focus on the physical layer parameters of network signals, while ignoring the impact of network behavior patterns on network security. This solution addresses the existing problem of insufficient consideration of network behavior patterns by integrating multi-source data such as signaling occurrence probability, federated learning weight coefficients, and anomaly probability. This makes network security assessments more comprehensive and accurate, enabling more precise assessment of network security status and providing a comprehensive quantitative basis for the calculation of network security indicators. During network switching, the security of the target network can be more accurately judged based on the entropy of the behavior pattern, thereby prioritizing networks with low behavior entropy, improving overall network security, and reducing the likelihood of user devices being subject to malicious attacks or abnormal behavior interference.

[0114] Step 4: Determine whether the network security indicators meet the network security standards; when the network security indicators all meet the network security standards, execute the network switching instruction.

[0115] Step 401: Set the signal space-time conflict threshold, Doppler frequency shift deviation threshold and behavior pattern entropy threshold respectively; compare the signal space-time conflict index, Doppler frequency shift deviation and behavior pattern entropy with the corresponding thresholds respectively; when the signal space-time conflict index is less than the signal space-time conflict threshold, this indicator meets the safety standard; when the Doppler frequency shift deviation is less than the Doppler frequency shift deviation threshold, this indicator meets the safety standard; when the behavior pattern entropy threshold is less than the behavior pattern entropy threshold, this indicator meets the safety standard.

[0116] It should be noted that the signal spatiotemporal conflict threshold can be determined by collecting data on the signal spatiotemporal conflict index in different scenarios over the past, analyzing its distribution and changing trends, and determining a threshold that ensures network security. For example, the signal spatiotemporal conflict index can be calculated over the past month in different regions and time periods, and the average or a percentile (such as the 95th percentile) can be used as the threshold. Alternatively, by studying the signal spatiotemporal conflict index during past network security incidents, the index range associated with security incidents can be identified and the threshold can be set slightly below the lower limit of this range to promptly trigger the security assessment mechanism when similar incidents occur. For the Doppler shift deviation threshold, in actual network environments, field testing can be conducted to record the network quality assessment index under different Doppler shift deviations. The Doppler shift deviations that meet the standard for network quality assessment index are then calculated and the average or a percentile (such as the 95th percentile) can be used as the threshold. The behavioral pattern entropy threshold can be determined by collecting and analyzing known abnormal behavior samples, calculating their behavioral pattern entropy, and setting the threshold at the critical point between normal and abnormal behavior entropy.

[0117] Step 402: When all three indicators are qualified, it is determined that the network to be accessed meets the security assessment standard, and the network switching instruction is executed.

[0118] When all three network security indicators are qualified, the network to be accessed is quickly determined to meet the security assessment standards, and the network switching instruction is immediately executed to ensure that the user device can be connected to a safe and reliable network in a timely manner; the user device can connect to a safer and higher-quality network more quickly, reducing problems such as communication quality degradation or data loss caused by network switching delays, thereby improving user satisfaction and communication experience.

[0119] Step 5: When the network to be accessed meets the basic performance evaluation standards but does not meet the network security standards, if an emergency security warning instruction is received, the exemption instruction can be temporarily activated, the evaluation results of the network security evaluation can be ignored, and the instruction to switch networks can be executed. After the emergency security warning instruction is eliminated, the network evaluation strategy of steps 1 to 4 can be referred to to determine whether to switch to other networks.

[0120] It should be noted that the emergency security warning instruction can be an emergency instruction issued in response to official orders, such as a red security warning (war and natural disasters, etc.). It is necessary to verify the digital signature certificate PKI and the physical key HSM at the same time. After determining that the instruction is authentic, the exemption instruction is triggered.

[0121] If the network to be connected meets the basic performance assessment criteria but does not meet the network security criteria, if an emergency security warning is received, a temporary exemption command can be activated, ignoring the network security assessment results and executing the network switching command. After the emergency security warning is eliminated, the network assessment strategy in steps 1 to 4 will be used to determine whether to switch to another network.

[0122] In certain special emergency situations (such as war, natural disasters, etc.), network security standards may conflict with the timeliness and continuity of communications. Traditional network switching strategies will refuse switching when network security assessments do not meet standards. However, in emergency situations, this practice may lead to communication interruptions, affecting the transmission of critical information and emergency rescue work. This solution introduces an emergency exemption mechanism to ensure the authenticity of instructions. It solves the problem of the inability to flexibly switch networks in emergency situations in existing technologies, thereby improving the system's emergency response capabilities and flexibility. It can quickly establish communication connections in emergency situations, ensure the continuity of critical services and the timely transmission of information, reduce switching delays caused by network security assessments, and reduce the risk of communication interruptions. It provides strong support for emergency rescue, command and dispatch, etc. At the same time, it also ensures the security of the system through the instruction verification mechanism, preventing malicious attacks from exploiting the exemption mechanism to cause damage.

[0123] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0124] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0125] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. An eSIM automated network switching strategy management method based on multi-source data fusion is characterized by: include: Step 1: Obtain real-time network status data and analyze it to calculate basic network performance indicators. Basic performance indicators include the network quality assessment index, roaming cost index, and reference signal theoretical received power. The reference signal theoretical received power is obtained as follows: Use the meteorological API to obtain the real-time rainfall in the area and the altitude difference between the terminal and the base station. Obtain the material attenuation coefficient of the obstructions or obstacles in the area. Normalize the real-time rainfall, altitude difference, and material attenuation coefficient, and analyze and calculate the environmental compensation factor. The formula is as follows: ； Where: HBC represents the environmental compensation factor; JY represents the real-time rainfall in the area; HB represents the altitude difference between the area and the base station; JZ j represents the material attenuation coefficient of the jth obstruction in the area; kr is the rain attenuation coefficient, which represents the dynamic influence weight of rainfall on signal attenuation; kt is the terrain coefficient, which represents the nonlinear compensation factor of terrain undulation on signal attenuation; β j Represents the weight of different building materials in the environment, and the calculation formula is: β j = Material attenuation coefficient of the jth obstruction / the sum of the material attenuation coefficients of all obstructions; j represents the serial number of the obstruction, which is a positive integer; The theoretical received power of the reference signal is calculated based on the original signal strength and the environmental compensation factor. Specifically, the theoretical received power of the reference signal = original signal strength - environmental compensation factor. Step 2: Evaluate basic performance indicators to determine whether the network to be connected meets the basic performance evaluation standards; Step 3: When the network to be accessed meets the basic performance evaluation criteria, obtain network security data, analyze the network security data, and calculate network security indicators; Step 4: Determine whether the network security indicators meet the network security standards; if the network security indicators meet the network security standards, execute the network switching instruction; Step 5: When the network to be accessed meets the basic performance evaluation standards but does not meet the network security standards, if an emergency security warning instruction is received, the exemption instruction can be temporarily activated, the evaluation results of the network security evaluation can be ignored, and the instruction to switch networks can be executed. After the emergency security warning instruction is eliminated, the network evaluation strategy of steps 1 to 4 can be referred to to determine whether to switch to other networks.

2. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 1 is characterized in that: The method for obtaining the network quality evaluation index is as follows: The measurement configuration is obtained through the measurement control message in the system information block (SIB) broadcast by the base station, and then the theoretical received power of the reference signal is measured. After receiving the signal from the base station, the signal-to-noise ratio is calculated based on the received useful signal power as well as the interference and noise floor power. In wireless networks, the round-trip delay is measured by measuring the round-trip time of RRC signaling or the round-trip time of the TCP transmission control protocol connection. The theoretical received power, signal-to-noise ratio, and round-trip delay of the reference signal are normalized and preprocessed to obtain the received signal score, signal-to-noise ratio score, and round-trip delay score, respectively. A comprehensive analysis is then performed using the network quality assessment value index calculation model to obtain the network quality assessment index.

3. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 2 is characterized in that: The method to obtain the roaming cost index is as follows: By connecting to the network operator's billing system through an API, users can obtain real-time tariff values ​​in different network environments. By querying the tax database of the target country or region, the applicable tax rate value for communication services can be obtained. The service contract signed between the enterprise and the operator can be obtained through the enterprise's contract management system, and the benchmark tariff value can be extracted from the service contract. The real-time tariff value, applicable tax rate value and benchmark tariff value are normalized and pre-processed, and a comprehensive analysis is performed through the roaming cost calculation model to obtain the roaming cost index.

4. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 3 is characterized in that: Comprehensively evaluate the network quality assessment index, roaming cost index, and theoretical received power of the reference signal to determine whether the network meets the basic performance evaluation standards. The specific method is as follows: Set the network quality index threshold, roaming cost index threshold, and receiving power threshold; compare the network quality assessment index, roaming cost index, and environmental compensation factor with the network quality index threshold, roaming cost index threshold, and environmental status threshold, respectively. When the network quality assessment index is greater than or equal to the network quality index threshold, the indicator is considered qualified; when the roaming cost index is less than or equal to the roaming cost index threshold, the indicator is considered qualified; when the reference signal theoretical receiving power is greater than or equal to the receiving power threshold, the indicator is considered qualified; When all three indicators are qualified, the network to be accessed is determined to meet the basic performance evaluation standards.

5. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 4 is characterized in that: Cybersecurity data includes signal spatiotemporal conflict index, Doppler shift deviation, and behavior pattern entropy.

6. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 5 is characterized in that: The method for obtaining the signal space-time conflict index is as follows: The terminal obtains the GPS positioning coordinates of the terminal through the built-in GNSS module, and queries the geographical location coordinates registered by the base station through the core network MME database; obtains the moving speed of the terminal through the GNSS Doppler frequency shift or inertial navigation system, and obtains the base station location update interval through the base station broadcast system information block SIB; Obtain the base station authentication level value of the operator's core network for the base station identity, normalize and pre-process the terminal's GPS positioning coordinates, the base station's registered geographic location coordinates, the terminal's moving speed, the base station location update interval, and the base station authentication level value, and calculate the signal spatiotemporal conflict index based on the following formula: ； Among them, STC represents the signal space-time conflict index; L GPS Indicates the GPS positioning coordinates of the terminal; L CEL Indicates the geographical location coordinates of the base station registration; Indicates the base station location update interval; v indicates the terminal's moving speed; ZHD indicates the base station certification level value; The baseband processor uses a phase-locked loop (PLL) to measure the carrier frequency offset value actually measured by the terminal in real time. The carrier center frequency is obtained from the carrier frequency parameters broadcast by the base station, and the Doppler shift value is calculated based on the carrier center frequency. The formula is: , in the formula f ca represents the Doppler frequency shift value, f0 represents the carrier center frequency, v represents the terminal's moving speed, θ represents the angle between the base station's moving direction and the base station, c represents the speed of light, and YD represents the moving direction weight. The terminal's built-in magnetometer and accelerometer are used to obtain the angle between the moving direction and the base station connection, and the cosine value of the angle is calculated as the moving direction weight. The carrier frequency offset value, carrier center frequency, Doppler frequency shift value, and moving direction weight are normalized and preprocessed, and the processed parameters are input into the Doppler frequency shift deviation calculation model to calculate the Doppler frequency shift deviation. The proportion of each signaling type within a unit period is counted to obtain the probability of occurrence of the kth type of signaling within the time window; the federated learning weight coefficient is dynamically adjusted based on the number of terminals through the cloud; the federated learning server API is called, and the local entropy value and time series characteristics are input to obtain the anomaly probability as the federated learning anomaly score; the local entropy value is calculated as follows: ;p k represents the probability of occurrence of the k-th type of signaling within the time window; ρ represents the federated learning weight coefficient; YC represents the federated learning anomaly score; The occurrence probability of the k-th type of signaling within the time window, the federated learning weight coefficient and the abnormal probability are normalized and preprocessed, and the behavior pattern entropy is calculated according to the behavior pattern entropy calculation model.

7. The eSIM automated network switching strategy management method based on multi-source data fusion according to claim 6 is characterized in that: The method for determining whether network security indicators meet network security standards is as follows: The signal space-time conflict threshold, Doppler frequency shift deviation threshold and behavior pattern entropy threshold are set respectively; the signal space-time conflict index, Doppler frequency shift deviation and behavior pattern entropy are compared with the corresponding thresholds respectively; when the signal space-time conflict index is less than the signal space-time conflict threshold, this indicator meets the safety standard; when the Doppler frequency shift deviation is less than the Doppler frequency shift deviation threshold, this indicator meets the safety standard; when the behavior pattern entropy threshold is less than the behavior pattern entropy threshold, this indicator meets the safety standard.

Images