Low-orbit satellite walkie-talkie seamless communication method and system based on star-ground switching pre-judgment

Abstract

The application discloses a low-orbit satellite intercom seamless communication method and system based on star-ground switching prediction, relates to the technical field of communication, and comprises the following steps: constructing a star-ground link quality degradation trajectory parameter, performing switching determination, generating a switching signal, performing access analysis according to the switching signal in combination with ephemeris data and channel state information, and constructing a candidate access network set; performing network switching evaluation based on the candidate access network set, calculating multiple candidate network switching adaptation degrees, performing positioning by traversing the candidate access network set according to the multiple candidate network switching adaptation degrees, and determining a target access network; performing context migration through the target access network to perform seamless communication switching, generating a seamless switching result for backtracking verification, and updating the target access network according to a switching verification result. The application solves the technical problem of poor communication continuity during star-ground switching of a low-orbit satellite intercom in the prior art, and achieves the technical effect of improving communication continuity during star-ground switching.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and specifically to a seamless communication method and system for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction. Background Technology

[0002] Low-Earth orbit (LEO) satellite walkie-talkies rely on LEO satellite networks to achieve long-distance, wide-coverage communication, suitable for scenarios with insufficient terrestrial network coverage or emergency communication. However, due to the high-speed movement of LEO satellites relative to ground terminals, the visibility time between the satellite and the walkie-talkie is limited, and link quality fluctuates continuously with changes in satellite elevation angle, relative speed, obstruction, and channel conditions. When the coverage of the serving satellite decreases or a handover between the satellite network and the terrestrial network is required, failure to anticipate link degradation trends and promptly select a suitable target network can easily lead to delayed handover timing, discontinuous service context migration, and consequently, voice interruptions, packet loss, or connection recovery delays, affecting the communication continuity of LEO satellite walkie-talkies in mobile communication scenarios. Summary of the Invention

[0003] This application provides a seamless communication method and system for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, which is used to address the technical problem of poor communication continuity during satellite-to-ground handover in existing low-Earth orbit satellite walkie-talkies.

[0004] In view of the above problems, this application provides a method and system for seamless communication of low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction.

[0005] The first aspect of this application provides a method for seamless communication of low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, the method comprising: Real-time motion status information and orbital parameters of low-Earth orbit satellites are retrieved and mapped to a prediction time window for calculation to construct satellite-to-ground link quality degradation trajectory parameters. Based on these parameters, a handover determination is made, generating a handover signal. Access analysis is performed using the handover signal in conjunction with ephemeris data and channel status information to construct a candidate access network set. Network handover evaluation is conducted based on this set, calculating the handover suitability of multiple candidate networks. The candidate access network set is then traversed according to these suitability scores to determine the target access network. Seamless communication handover is performed via context migration through the target access network, generating a seamless handover result for backtracking verification. The target access network is then updated based on the handover verification result.

[0006] A second aspect of this application provides a seamless communication system for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, the system comprising: The calculation module retrieves real-time motion status information and low-Earth orbit satellite orbit parameters, maps them to a prediction time window, and constructs satellite-to-ground link quality degradation trajectory parameters. The handover determination module determines handover based on these parameters, generates a handover signal, and performs access analysis using the handover signal in conjunction with ephemeris data and channel status information to construct a candidate access network set. The handover evaluation module evaluates the network handover based on the candidate access network set, calculates the handover suitability of multiple candidate networks, traverses the candidate access network set according to these suitability values, locates the target access network, and determines the target access network. The communication handover module performs seamless communication handover through context migration via the target access network, generates a seamless handover result for backtracking verification, and updates the target access network based on the handover verification result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application retrieves real-time motion state information and low-Earth orbit satellite orbital parameters, maps them to a prediction time window for calculation, and constructs satellite-to-ground link quality degradation trajectory parameters. Based on these parameters, a handover determination is made, a handover signal is generated, and access analysis is performed using the handover signal in conjunction with ephemeris data and channel state information to construct a candidate access network set. Based on this set, network handover evaluation is conducted, calculating the handover suitability of multiple candidate networks. The candidate access network set is then traversed according to these suitability values ​​to determine the target access network. Seamless communication handover is performed through context migration via the target access network, and a seamless handover result is generated for backtracking verification. The target access network is then updated based on the handover verification result. This invention solves the technical problem of poor communication continuity during satellite-to-ground handover in existing low-Earth orbit satellite walkie-talkies. By constructing satellite-to-ground link quality degradation trajectory parameters for handover prediction and determining the target access network for context migration, it achieves the technical effect of improving communication continuity during satellite-to-ground handover. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction provided in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a seamless communication system for low-orbit satellite walkie-talkies based on satellite-to-ground switching prediction, provided in an embodiment of this application.

[0010] Explanation of reference numerals in the attached diagram: Calculation module 11, Switching determination module 12, Switching evaluation module 13, Communication switching module 14. Detailed Implementation

[0011] This application provides a seamless communication method and system for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction. It addresses the technical problem of poor communication continuity during satellite-to-ground handover in existing technologies by constructing satellite-to-ground link quality degradation trajectory parameters for handover prediction and determining the target access network to perform context migration, thereby improving the technical effect of communication continuity during satellite-to-ground handover.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, this application provides a seamless communication method for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, the method comprising: Step S100: Retrieve real-time motion status information and low-orbit satellite orbit parameters, map them to the prediction time window for calculation, and construct the satellite-to-ground link quality degradation trajectory parameters.

[0015] In this embodiment, real-time motion state information, including inertial measurement data and satellite positioning data, is first obtained by real-time operation sensing based on a low-Earth orbit satellite intercom. Then, the real-time motion state information is retrieved, and the orbital parameters of the low-Earth orbit satellite are introduced. The inertial measurement data, satellite positioning data, and the orbital parameters of the low-Earth orbit satellite are mapped to a prediction time window for calculation, obtaining motion prediction trajectory parameters, satellite elevation angle sequence, and relative radial velocity sequence. Next, a time-series convolutional network is used for signal prediction based on the satellite elevation angle sequence and relative radial velocity sequence to generate a reference signal received power prediction sequence. The signal attenuation rate and interruption prediction time data are then calculated based on the reference signal received power prediction sequence. Finally, the signal attenuation rate is integrated according to the interruption prediction time data to construct the satellite-to-ground link quality degradation trajectory parameters.

[0016] Furthermore, the method provided in the application embodiment, which retrieves real-time motion state information and maps the orbital parameters of low-Earth orbit satellites to a prediction time window for calculation, and constructs satellite-to-ground link quality degradation trajectory parameters, also includes: Based on real-time operational sensing of a low-Earth orbit (LEO) satellite intercom, the real-time motion state information is obtained, including inertial measurement data and satellite positioning data. A prediction time window is constructed, and the inertial measurement data and satellite positioning data are mapped to the prediction time window to predict the motion of the LEO satellite intercom, obtaining motion prediction trajectory parameters. The orbital parameters of the LEO satellite are introduced, and the motion prediction trajectory parameters and the orbital parameters are mapped to the prediction time window for sampling to obtain satellite elevation angle sequence and relative radial velocity sequence. Based on the satellite elevation angle sequence and the relative radial velocity sequence, a temporal convolutional network signal prediction is performed to generate a reference signal received power prediction sequence. Based on the reference signal received power prediction sequence, signal attenuation analysis is performed to calculate the signal attenuation rate. Based on the reference signal received power prediction sequence, interruption prediction is performed to calculate interruption prediction time data. The signal attenuation rate is integrated according to the interruption prediction time data to construct the satellite-to-ground link quality degradation trajectory parameters.

[0017] In this embodiment, when performing real-time motion sensing based on a low-Earth orbit (LEO) satellite intercom, the LEO satellite intercom collects inertial measurement data and satellite positioning data according to a preset sampling period. The inertial measurement data includes triaxial acceleration, triaxial angular velocity, and attitude data corresponding to each sampling time. The satellite positioning data includes longitude, latitude, altitude, velocity, and timestamp corresponding to each sampling time. After collection, the inertial measurement data and satellite positioning data are sorted according to the timestamp. Using the timestamp of the satellite positioning data as a reference, linear interpolation is performed on the data between two adjacent sampling times of inertial measurement data, so that each timestamp of satellite positioning data corresponds to a set of inertial measurement data, thus obtaining the real-time motion state information corresponding to the time.

[0018] Next, a prediction time window is constructed, mapping inertial measurement data and satellite positioning data to the prediction time window to predict the motion of the low-Earth orbit satellite walkie-talkie. In this process, the prediction start time, prediction end time, and sampling interval are determined, and multiple prediction sampling times are generated according to the sampling interval. The longitude, latitude, and altitude corresponding to the prediction start time are used as the starting position, the velocity corresponding to the prediction start time is used as the starting velocity, and the attitude data corresponding to the prediction start time is used as the starting attitude. For two adjacent prediction sampling times, based on the attitude data of the previous prediction sampling time, the three-axis acceleration within the sampling interval is converted to the same spatial direction as the satellite positioning data. Then, the velocity of the previous prediction sampling time is added to the product of the acceleration and the sampling interval to obtain the velocity of the next prediction sampling time. The position of the previous prediction sampling time is added to the product of the velocity and the sampling interval, plus half of the product of the acceleration and the square of the sampling interval, to obtain the position of the next prediction sampling time. The attitude data of the previous prediction sampling time is added to the product of the three-axis angular velocity and the sampling interval to obtain the attitude data of the next prediction sampling time. The position, velocity, and attitude of all predicted sampling moments within the prediction time window are calculated sequentially and arranged in chronological order to obtain the motion prediction trajectory parameters.

[0019] Subsequently, the orbital parameters of the low-Earth orbit (LEO) satellite are introduced. The motion prediction trajectory parameters are mapped to the orbital parameters and sampled within a prediction time window. The ephemeris time, satellite spatial position, and satellite velocity of the LEO satellite in the period adjacent to the prediction time window are read. For each prediction sampling moment, the LEO satellite intercom position and velocity are extracted from the motion prediction trajectory parameters. Based on the time ratio between two adjacent ephemeris times, the satellite spatial position and velocity corresponding to the two adjacent ephemeris times are linearly interpolated to obtain the LEO satellite spatial position and velocity. The LEO satellite intercom position is subtracted from the LEO satellite spatial position to obtain the satellite-to-ground connection. The ground plane direction and vertical direction are determined based on the longitude, latitude, and altitude of the LEO satellite intercom. The satellite-to-ground connection is decomposed into horizontal and altitude components. The altitude component is divided by the length of the satellite-to-ground connection and the corresponding angle is calculated to obtain the satellite elevation angle. The relative velocity is obtained by subtracting the low-Earth orbit satellite intercom velocity from the satellite velocity. The direction of the satellite-to-ground connection is obtained by dividing the satellite-to-ground connection line by its length. The relative radial velocity is obtained by taking the component of the relative velocity along the satellite-to-ground connection direction. The satellite elevation angle sequence and the relative radial velocity sequence are obtained by calculating and arranging the results for each predicted sampling time.

[0020] Next, signal prediction using a temporal convolutional network is performed based on the satellite elevation angle sequence and relative radial velocity sequence. In this process, historical satellite elevation angle sequences, historical relative radial velocity sequences, and measured reference signal received power sequences from historical communication records are used. The historical satellite elevation angle and historical relative radial velocity corresponding to the same historical sampling time are used as inputs, and the measured reference signal received power corresponding to the same historical sampling time is used as the output. This is used to train the temporal convolutional network, resulting in a trained temporal convolutional network. During prediction, the satellite elevation angle and relative radial velocity corresponding to the same prediction sampling time within the prediction time window are used as a set of input data and input into the trained temporal convolutional network in chronological order of the prediction sampling times. The temporal convolutional network reads the satellite elevation angle and relative radial velocity at multiple consecutive prediction sampling times, multiplies them by the trained convolutional weights, sums them, and adds the trained bias to output the predicted reference signal received power value for each prediction sampling time. The predicted reference signal received power values ​​are then arranged in chronological order to generate a predicted reference signal received power sequence.

[0021] Then, signal attenuation analysis is performed based on the reference signal received power prediction sequence. The predicted reference signal received power values ​​for adjacent prediction sampling times are read sequentially. When the predicted reference signal received power value for the later prediction sampling time is less than that for the earlier prediction sampling time, the power decrease is obtained by subtracting the predicted reference signal received power value for the later prediction sampling time from the earlier prediction sampling time. The power decrease is then divided by the sampling interval between the two adjacent prediction sampling times to obtain the signal attenuation rate. When the predicted reference signal received power value for the later prediction sampling time is greater than or equal to that for the earlier prediction sampling time, the signal attenuation rate is recorded as zero. This process is repeated for all adjacent prediction sampling times within the prediction time window.

[0022] Subsequently, interruption prediction is performed based on the reference signal received power prediction sequence, and the interruption prediction time data is calculated. In this process, the minimum reference signal received power required for the low-Earth orbit satellite intercom to maintain satellite-to-ground communication is determined, and each predicted reference signal received power value is compared with the minimum reference signal received power in chronological order. When the predicted reference signal received power value at a certain prediction sampling time is less than or equal to the minimum reference signal received power, that prediction sampling time is determined as the interruption prediction time data. When the predicted reference signal received power values ​​of two adjacent prediction sampling times change from above the minimum reference signal received power to below the minimum reference signal received power, the difference between the predicted reference signal received power value of the previous prediction sampling time and the minimum reference signal received power is calculated. Then, the difference between the predicted reference signal received power value of the previous prediction sampling time and the predicted reference signal received power value of the next prediction sampling time is calculated. The two differences are divided and multiplied by the sampling interval to obtain the time elapsed for the predicted reference signal received power value to decrease to the minimum reference signal received power. This time is then added to the previous prediction sampling time to obtain the interruption prediction time data.

[0023] Finally, the signal attenuation rate is integrated according to the interruption prediction time data. Based on the order of the predicted sampling times, the satellite elevation angle sequence, relative radial velocity sequence, reference signal received power prediction sequence, and signal attenuation rate are arranged accordingly, so that each predicted sampling time corresponds to a satellite elevation angle, a relative radial velocity, and a reference signal received power prediction value, and adjacent predicted sampling times correspond to a signal attenuation rate. The interruption position in the reference signal received power prediction sequence is determined based on the interruption prediction time data. The reference signal received power prediction value and signal attenuation rate before the interruption position are extracted and mapped to the satellite elevation angle and relative radial velocity at the corresponding predicted sampling time to construct the satellite-to-ground link quality degradation trajectory parameters.

[0024] Step S200: Based on the satellite-to-ground link quality degradation trajectory parameters, a handover determination is made, a handover signal is generated, and access analysis is performed according to the handover signal in combination with ephemeris data and channel state information to construct a candidate access network set.

[0025] In this embodiment, when determining handover based on the satellite-to-ground link quality degradation trajectory parameters, handover prediction conditions are first set, and channel quality analysis is performed using the terrestrial communication network to obtain available signal quality parameters. Then, the estimated interruption time data in the satellite-to-ground link quality degradation trajectory parameters is compared with the prediction time window, and the available signal quality parameters are compared with a preset quality threshold. When the estimated interruption time data is within the prediction time window and the available signal quality parameters are lower than the preset quality threshold, the satellite-to-ground link quality degradation trajectory parameters are determined to meet the handover prediction conditions, and a handover signal is generated. Simultaneously, an Earth shadow model is constructed, and based on the orbital parameters of the low-Earth orbit satellite combined with the Earth shadow model, it is determined whether the low-Earth orbit satellite walkie-talkie enters the shadow area of ​​the Earth shadow model during the estimated interruption time. When it is during the estimated interruption time, the satellite-to-ground link quality degradation trajectory parameters are determined to meet the handover prediction conditions, and a handover signal is generated.

[0026] Next, when performing access analysis based on the switching signal, ephemeris data, and channel state information, ephemeris data of adjacent low-Earth orbit (LEO) satellites are introduced, and motion prediction trajectory parameters are combined to screen LEO satellite walkie-talkies, obtaining a subset of candidate satellites that meet the access conditions. Simultaneously, signal reception calculations are performed based on the terrestrial communication network to obtain signal reception power data, which is then compared with the access threshold. When the reference signal reception power data exceeds the access threshold, a subset of candidate terrestrial networks is formed. Subsequently, the candidate satellite subset and the candidate terrestrial network subset are merged to generate a candidate access network set.

[0027] Furthermore, in the method provided in the application embodiments, the handover determination based on the satellite-to-ground link quality degradation trajectory parameters and the generation of a handover signal further include: Based on the degradation trajectory parameters of the satellite-to-ground link quality, a handover determination is made, and handover prediction conditions are set: a terrestrial communication network is introduced, and channel quality analysis is performed on the terrestrial communication network to obtain available signal quality parameters; if the estimated interruption time data is within the prediction time window and the available channel quality parameters are lower than a preset quality threshold, then the degradation trajectory parameters of the satellite-to-ground link quality are determined to meet the handover prediction conditions, and a handover signal is generated; or, an Earth shadow model is constructed, and based on the orbital parameters of the low-Earth orbit satellite and the Earth shadow model, it is determined whether the low-Earth orbit satellite walkie-talkie enters the shadow area of ​​the Earth shadow model during the estimated interruption time data. If it does, then the degradation trajectory parameters of the satellite-to-ground link quality are determined to meet the handover prediction conditions, and a handover signal is generated.

[0028] In this embodiment, when determining handover based on satellite-to-ground link quality degradation trajectory parameters, the predicted time window, estimated interruption time data, reference signal received power prediction sequence, and signal attenuation rate in the satellite-to-ground link quality degradation trajectory parameters are first read, and handover prediction conditions are set. The predicted time window includes a start time and an end time. The estimated interruption time data is the time when the predicted reference signal received power drops to the minimum reference signal received power required for the low-Earth orbit satellite intercom to maintain satellite-to-ground communication. After reading, the estimated interruption time data is compared with the start and end times of the predicted time window. First, it is determined whether the estimated interruption time data is greater than or equal to the start time of the predicted time window, and then it is determined whether the estimated interruption time data is less than or equal to the end time of the predicted time window. If both determinations are true, it is determined that the estimated interruption time data is within the predicted time window.

[0029] Next, a terrestrial communication network is introduced, and channel quality analysis is performed. Within the coverage area of ​​the terrestrial communication network corresponding to the current location of the low-Earth orbit satellite walkie-talkie, the signal received power, signal-to-noise ratio (SNR), and channel occupancy rate of the terrestrial communication network are collected according to the sampling time of the prediction time window. Subsequently, the signal received power, SNR, and channel occupancy rate are numerically processed. The signal received power is divided by a preset received power reference value to obtain the signal received power ratio, the SNR is divided by a preset SNR reference value to obtain the SNR ratio, and the channel occupancy rate is subtracted from 1 to obtain the channel idle ratio. Then, the signal received power ratio, SNR ratio, and channel idle ratio are multiplied by preset weights and summed to obtain the available signal quality parameters at the corresponding sampling time. The above calculations are performed for multiple sampling times within the prediction time window, and the smallest available signal quality parameter is taken as the available channel quality parameter of the current terrestrial communication network.

[0030] The available channel quality parameters are then compared with a preset quality threshold. If the available channel quality parameters are less than the preset quality threshold, the channel quality judgment result of the ground communication network is recorded as below the preset quality threshold. Simultaneously, combining the aforementioned time comparison results, if the estimated interruption time data is within the prediction time window and the available channel quality parameters are below the preset quality threshold, the satellite-to-ground link quality degradation trajectory parameters are determined to meet the handover prediction conditions, and a handover signal is generated. When generating the handover signal, the prediction time window, the estimated interruption time data, the available channel quality parameters, and the preset quality threshold are written into the handover signal, enabling subsequent access analysis to continue according to the handover signal.

[0031] When making another switching decision, an Earth shadow model is constructed. Specifically, the Earth's center position, Earth's radius, and the direction of solar incidence are determined, and the direction of the shadow area in the Earth shadow model is defined by the reverse extension of the direction of solar incidence. The boundary of the shadow area is then determined using the Earth's radius, forming the Earth shadow model. Subsequently, the orbital parameters of the low-Earth orbit (LEO) satellites are read. These orbital parameters include the satellite's spatial position at each sampling time within the prediction time window. When the orbital parameters provide the satellite's spatial position for adjacent ephemeris times, linear interpolation is performed according to the proportion of time occupied by the sampling time between adjacent ephemeris times to obtain the satellite's spatial position at each sampling time within the prediction time window.

[0032] Finally, when determining the shadow area based on the orbital parameters of the low-Earth orbit satellite and the Earth's shadow model, the spatial position of the low-Earth orbit satellite at each sampling time within the prediction time window is first read, along with the Earth's center position, Earth's radius, and the direction of solar incidence from the Earth's shadow model. For any sampling time, the Earth's center position is subtracted from the low-Earth orbit satellite's spatial position to obtain the direction data from the Earth's center to the low-Earth orbit satellite's spatial position. This direction data is then compared with the direction of solar incidence. When the low-Earth orbit satellite's spatial position is on the side opposite to the direction of solar incidence, the center line of the shadow area is determined from the Earth's center position along the opposite direction of solar incidence. Subsequently, the distance from the low-Earth orbit satellite's spatial position to the center line of the shadow area is calculated. First, the projection length of the direction data from the Earth's center to the low-Earth orbit satellite's spatial position onto the direction of the shadow area's center line is calculated. Then, the distance is determined based on the length of the direction data from the Earth's center to the low-Earth orbit satellite's spatial position and the projection length. The square root of the squared difference between degrees is calculated to obtain the vertical distance from the low-Earth orbit satellite's spatial position to the center line of the shadow area. The vertical distance is compared with the Earth's radius. When the vertical distance is less than or equal to the Earth's radius, it is determined that the low-Earth orbit satellite walkie-talkie has entered the shadow area of ​​the Earth's shadow model at that sampling time. Then, the sampling time of entering the shadow area is compared with the interruption prediction time data. When the sampling time of entering the shadow area is equal to the interruption prediction time data, or the interruption prediction time data is located between the sampling time of entering the shadow area and the next sampling time, it is determined that the low-Earth orbit satellite walkie-talkie entered the shadow area of ​​the Earth's shadow model at the interruption prediction time data. Then, it is determined that the trajectory parameters of the satellite-to-ground link quality degradation meet the switching prediction conditions, a switching signal is generated, and the interruption prediction time data, the sampling time of entering the shadow area, the shadow area judgment result of the Earth's shadow model, and the corresponding low-Earth orbit satellite orbit parameters are written into the switching signal.

[0033] Furthermore, in the method provided in the application embodiments, the method for constructing a candidate access network set by performing access analysis based on the switching signal combined with ephemeris data and channel state information further includes: By incorporating ephemeris data from adjacent low-Earth orbit (LEO) satellites and combining it with the motion prediction trajectory parameters, a satellite selection process is performed on LEO satellite walkie-talkies to form a candidate satellite subset. Signal reception power data is obtained based on the terrestrial communication network and compared with an access threshold. When the reference signal reception power data exceeds the access threshold, a candidate terrestrial network subset is formed. The candidate satellite subset and the candidate terrestrial network subset are merged to generate the candidate access network set.

[0034] In this embodiment, when using ephemeris data from adjacent low-Earth orbit (LEO) satellites in conjunction with motion prediction trajectory parameters to screen LEO satellite walkie-talkies, the system reads the prediction time window and interruption prediction time data from the switching signal, and also reads the ephemeris data of adjacent LEO satellites. The ephemeris data includes the satellite identifier, ephemeris time, satellite spatial position, and satellite velocity corresponding to each sampling time within the prediction time window. Simultaneously, the motion prediction trajectory parameters are read, including the position and velocity data of the LEO satellite walkie-talkie corresponding to each sampling time within the prediction time window. First, the ephemeris time of adjacent LEO satellites is matched with the sampling time of the motion prediction trajectory parameters according to the sampling time within the prediction time window. When the ephemeris time and sampling time are inconsistent, the time ratio of the sampling time relative to the previous and subsequent ephemeris times is calculated, and the satellite spatial position and velocity corresponding to the previous and subsequent ephemeris times are linearly interpolated according to this time ratio to obtain the satellite spatial position and velocity corresponding to the sampling time. For each adjacent low-Earth orbit (LEO) satellite, the satellite's spatial position is subtracted from the LEO satellite intercom position to obtain the satellite-to-ground connection. The differences between the satellite-to-ground connection in the three coordinate directions are squared, summed, and the square root is taken to obtain the satellite-to-ground distance. The ground plane direction is determined based on the longitude, latitude, and altitude of the LEO satellite intercom. The satellite-to-ground connection is decomposed into horizontal and altitude components. The altitude component is divided by the length of the satellite-to-ground connection to obtain the elevation angle calculation ratio, and the satellite elevation angle is obtained based on this elevation angle calculation ratio. The satellite velocity is then subtracted from the LEO satellite intercom velocity to obtain the relative velocity. The satellite-to-ground connection direction is obtained by dividing the satellite-to-ground connection by the satellite-to-ground distance. The relative velocity and the satellite-to-ground connection direction are multiplied and summed to obtain the relative radial velocity. The satellite elevation angle, satellite-to-ground distance, and relative radial velocity are calculated for each sampling moment within the prediction time window. The number of sampling moments where the satellite elevation angle is continuously higher than the preset elevation angle threshold is counted. The number of sampling moments is multiplied by the sampling interval to obtain the coverage duration. The satellite elevation angle is compared with a preset elevation angle threshold, the satellite-to-ground distance is compared with a preset communication distance, and the coverage duration is compared with a preset duration. When an adjacent low-Earth orbit satellite meets the following conditions, such as a satellite elevation angle higher than the preset elevation angle threshold, a satellite-to-ground distance less than the preset communication distance, and a coverage duration greater than the preset duration, the adjacent low-Earth orbit satellite is added to the candidate satellite subset.

[0035] Next, signal reception calculations are performed based on the terrestrial communication network to obtain signal reception power data, which is then compared with the access threshold. Specifically, the channel state information of the terrestrial communication network is read, including the terrestrial communication network identifier, transmit power, transmit antenna gain, receive antenna gain, operating frequency, terrestrial communication network location, and access threshold. The location data of the low-Earth orbit satellite intercom at each sampling time within the prediction time window is also read from the motion prediction trajectory parameters. For each terrestrial communication network, the location of the terrestrial communication network is subtracted from the location of the low-Earth orbit satellite intercom to obtain the position difference between them. The values ​​of the position difference in the three coordinate directions are squared, summed, and the square root is taken to obtain the distance between the terrestrial communication network and the low-Earth orbit satellite intercom. Propagation loss is calculated based on the operating frequency and distance, which is equal to the sum of the logarithmic terms of the operating frequency, distance, and a constant term. The transmit power, transmit antenna gain, and receive antenna gain are then summed, and the propagation loss is subtracted to obtain the signal reception power data corresponding to that sampling time. The above calculations are repeated for each sampling time within the prediction time window to obtain the signal reception power data sequence of the terrestrial communication network within the prediction time window. Each signal received power data in the signal received power data sequence is compared with the access threshold in chronological order. When the signal received power data is greater than the access threshold, the corresponding sampling time is recorded as the sampling time that satisfies the access threshold.

[0036] When the reference signal received power data exceeds the access threshold, a subset of candidate ground networks is formed. Specifically, the signal received power data corresponding to each sampling time is used as the reference signal received power data, and the sampling times are judged according to their order within the prediction time window. For the same ground communication network, the reference signal received power data is read one by one, and the access threshold is subtracted from the reference signal received power data to obtain the threshold difference. When the threshold difference is greater than zero, the sampling time is recorded as meeting the access threshold. The number of sampling times of the ground communication network that meet the access threshold within the prediction time window is counted, and the number of sampling times is multiplied by the sampling interval to obtain the duration for which the reference signal received power data exceeds the access threshold. Ground communication networks that meet the access threshold are written into the candidate ground network subset, and the corresponding reference signal received power data, access threshold, threshold difference, and duration are retained.

[0037] Finally, the candidate satellite subset and the candidate ground network subset are merged. Adjacent low-Earth orbit (LEO) satellites in the candidate satellite subset are read, and ground communication networks in the candidate ground network subset are read. For each adjacent LEO satellite in the candidate satellite subset, the access object identifier, access object type, sampling time, ephemeris data, satellite elevation angle, satellite-to-ground distance, relative radial velocity, and coverage duration are recorded. For each ground communication network in the candidate ground network subset, the access object identifier, access object type, sampling time, channel state information, reference signal received power data, access threshold, threshold difference, and duration are recorded. The data in the candidate satellite subset and the candidate ground network subset are arranged according to the access object identifier. Data with the same access object identifier are merged into one access object record, while data with different access object identifiers are retained separately. After arrangement, the access object records corresponding to the candidate satellite subset and the candidate ground network subset are written into the same data list to generate a candidate access network set. As one implementation method, the access object records in the candidate access network set can be stored according to Table 1.

[0038] Table 1. Example Data Table of Candidate Access Network Sets ; ; As shown in Table 1, adjacent low-Earth orbit satellites that meet the screening conditions of satellite elevation angle, satellite-to-ground distance and coverage duration are written into the candidate satellite subset, and terrestrial communication networks that meet the condition that the reference signal received power data exceeds the access threshold are written into the candidate terrestrial network subset. They are then combined to generate the candidate access network set.

[0039] Step S300: Based on the candidate access network set, perform network handover evaluation, calculate the handover adaptability of multiple candidate networks, traverse the candidate access network set according to the multiple candidate network handover adaptability to locate the target access network.

[0040] In this embodiment, when evaluating network handover based on a candidate access network set, network handover evaluation parameters corresponding to each candidate network are extracted, and handover adaptation analysis is performed based on the network handover evaluation parameters to calculate the handover adaptation degree of multiple candidate networks. Then, the candidate access network set is sorted in descending order according to the handover adaptation degree of multiple candidate networks to generate a candidate network ranking list. Subsequently, the candidate network ranking list is traversed, and the first-order candidate network is extracted as the candidate network with the highest handover adaptation degree, and a network to be confirmed is generated. Service quality constraint verification is performed on the network to be confirmed. When the service quality constraint verification of the network to be confirmed passes, the network to be confirmed is located as the target access network. When the service quality constraint verification of the network to be confirmed fails, the next candidate network is traversed based on the candidate network ranking list as the network to be confirmed, and the service quality constraint verification is repeated until the network to be confirmed meets the service quality constraints, and the target access network is determined.

[0041] Furthermore, in the method provided in the application embodiments, the method further includes: performing network handover evaluation based on the candidate access network set, calculating multiple candidate network handover adaptability, traversing the candidate access network set according to the multiple candidate network handover adaptability to locate and determine the target access network; and also includes: Based on the candidate access network set, network handover evaluation is performed, network handover evaluation parameters are extracted for handover adaptation analysis, and the handover adaptation degree of multiple candidate networks is calculated. The candidate access network set is sorted in descending order according to the multiple candidate network handover adaptation degrees to generate a candidate network ranking list. The candidate network ranking list is traversed to extract the first-ranked candidate network, which is selected as the candidate network with the highest handover adaptation degree to generate a network to be confirmed. The network to be confirmed is subjected to quality of service constraint verification. If the quality of service constraint verification of the network to be confirmed passes, the network to be confirmed is located as the target access network. If the quality of service constraint verification of the network to be confirmed fails, the next candidate network is traversed based on the candidate network ranking list as the network to be confirmed, and the quality of service constraint verification is repeated until the network to be confirmed meets the quality of service constraints, and the target access network is determined.

[0042] In this embodiment, when evaluating network handover based on a candidate access network set, the candidate access network set is used as the evaluation object, and network handover evaluation parameters are used as evaluation indicators to construct a multi-attribute decision matrix. Then, the evaluation indicators in the multi-attribute decision matrix are normalized in the same direction to generate a standardized decision matrix. Next, a discrete analysis is performed based on the numerical distribution of each candidate network in the candidate access network set on the evaluation indicators, and entropy weight is calculated based on the indicator dispersion to obtain objective weight coefficients. Simultaneously, the service type of the low-orbit satellite walkie-talkie is retrieved, and service weight mapping is performed in conjunction with the evaluation indicators to calculate service bias weight coefficients. Then, the objective weight coefficients and service bias weight coefficients are weighted and fused to obtain a comprehensive weight vector, and the approximation of ideal solutions is sorted based on the comprehensive weight vector, where the ideal solutions include positive and negative ideal solutions. Based on the comprehensive weight vector, a proximity analysis is performed between the candidate access network set and the positive and negative ideal solutions respectively, calculating the first relative proximity and the second relative proximity. Finally, data integration is performed based on the first relative proximity and the second relative proximity to generate the handover adaptability corresponding to each candidate network.

[0043] Next, the candidate access network set is sorted in descending order according to the multiple candidate network handover adaptability scores. The candidate network handover adaptability scores corresponding to each candidate network in the candidate access network set are read, and the candidate networks are associated with their corresponding candidate network handover adaptability scores. Then, the candidate access network set is arranged in descending order of candidate network handover adaptability scores to obtain the candidate network sorting list.

[0044] Then, the candidate network sorting list is traversed to extract the first-order candidate network. The first-order candidate network is selected as the candidate network with the highest handover adaptability. That is, the first-order candidate network is read according to the sorting order of the candidate network sorting list, and the candidate network handover adaptability, signal received power data, coverage duration, access duration and network handover evaluation parameters corresponding to the first-order candidate network are extracted. The first-order candidate network is selected as the network to be confirmed.

[0045] Next, when performing service quality constraint verification on the network to be verified, the network handover evaluation parameters corresponding to the network to be verified are read, and each network handover evaluation parameter is compared with the service quality constraints. Specifically, the transmission delay of the network to be verified is compared with the delay constraint, the available bandwidth of the network to be verified is compared with the bandwidth constraint, the packet loss rate of the network to be verified is compared with the packet loss rate constraint, the coverage duration or access duration of the network to be verified is compared with the duration constraint, and the signal received power data of the network to be verified is compared with the signal received power constraint. When the transmission delay is not higher than the delay constraint, the available bandwidth is not lower than the bandwidth constraint, the packet loss rate is not higher than the packet loss rate constraint, the coverage duration or access duration is not lower than the duration constraint, and the signal received power data is not lower than the signal received power constraint, the service quality constraint verification of the network to be verified is deemed to have passed, and the network to be verified is identified as the target access network.

[0046] If the service quality constraint verification of the network to be confirmed fails, the next candidate network is traversed based on the candidate network sorting list as the network to be confirmed, and the service quality constraint verification is repeated. During this process, if any network handover evaluation parameter of the network to be confirmed does not meet the corresponding service quality constraint, the next candidate network is read according to the order of the candidate network sorting list and is taken as the new network to be confirmed. Then, the transmission delay, available bandwidth, packet loss rate, coverage duration or access duration, and signal received power data of the new network to be confirmed are compared with the corresponding service quality constraints item by item until the network to be confirmed meets the service quality constraints, and the target access network is determined.

[0047] Furthermore, in the method provided in the application embodiments, the process of performing network handover evaluation based on the candidate access network set, extracting network handover evaluation parameters for handover adaptation analysis, and calculating the handover adaptation degree of multiple candidate networks further includes: Using the candidate access network set as the evaluation object and the network handover evaluation parameters as the evaluation indicators, a multi-attribute decision matrix is ​​constructed. The evaluation indicators are normalized in the same direction by traversing the multi-attribute decision matrix to generate a standardized decision matrix. Discrete analysis is performed on the evaluation indicators based on the candidate access network set, and entropy weight calculation is performed based on the indicator dispersion to obtain objective weight coefficients. The service types of low-orbit satellite walkie-talkies are retrieved and combined with the evaluation indicators to perform service weight mapping, calculating service bias weight coefficients. The objective weight coefficients and the service bias weight coefficients are weighted and fused to obtain a comprehensive weight vector, which is then used to approximate the ideal solution and sort it: the ideal solution includes positive and negative ideal solutions. Based on the comprehensive weight vector, the candidate access network set is analyzed for proximity to the positive ideal solution to calculate a first relative proximity. Based on the comprehensive weight vector, the candidate access network set is analyzed for proximity to the negative ideal solution to calculate a second relative proximity. Data is integrated based on the first and second relative proximity to generate a handover adaptability score.

[0048] In this embodiment, the candidate access network set is used as the evaluation object, and network handover evaluation parameters are used as evaluation indicators. When constructing the multi-attribute decision matrix, each candidate network in the candidate access network set is read, and the network handover evaluation parameters corresponding to each candidate network are extracted. The network handover evaluation parameters include signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters. The candidate networks are used as rows of the multi-attribute decision matrix, and the network handover evaluation parameters are used as columns. The values ​​of the first candidate network under the signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters are sequentially written into the first row. The values ​​of the second candidate network under the above evaluation indicators are sequentially written into the second row, until all candidate networks in the candidate access network set are written, thus obtaining the multi-attribute decision matrix.

[0049] Next, the multi-attribute decision matrix is ​​traversed to perform forward normalization on the evaluation metrics. For signal received power data, coverage persistence parameters, resource availability parameters, and historical performance parameters, forward normalization is performed by first reading the maximum and minimum values ​​in the same evaluation metric column, then subtracting the minimum value from the current candidate network's metric value, and dividing by the difference between the maximum and minimum values ​​in that column to obtain the corresponding standardized value. For transmission delay and packet loss rate, reverse normalization is performed by first reading the maximum and minimum values ​​in the same evaluation metric column, then subtracting the current candidate network's metric value from the maximum value in that column, and dividing by the difference between the maximum and minimum values ​​in that column to obtain the corresponding standardized value. These standardized values ​​are then written back into the original positions of the candidate networks and evaluation metrics in the multi-attribute decision matrix to generate the standardized decision matrix.

[0050] Subsequently, a discrete analysis of the evaluation indicators is performed based on the candidate access network set. Entropy weights are calculated based on the indicator dispersion. For each evaluation indicator in the standardized decision matrix, the standardized values ​​of all candidate networks under that evaluation indicator are first summed to obtain the index column sum. Then, the standardized value of each candidate network under that evaluation indicator is divided by the index column sum to obtain the proportion of that candidate network under that evaluation indicator. Next, each proportion is multiplied by its natural logarithm, and the sum of the multiplication results for all candidate networks under the same evaluation indicator is calculated. This sum is then multiplied by -1 and the inverse of the natural logarithm of the number of candidate networks to obtain the information entropy of that evaluation indicator. Finally, the information entropy is subtracted from 1 to obtain the indicator dispersion of that evaluation indicator. After calculating the indicator dispersion of all evaluation indicators, the indicator dispersion of each evaluation indicator is divided by the sum of the indicator dispersions of all evaluation indicators to obtain the objective weight coefficient of the corresponding evaluation indicator.

[0051] Next, the service type of the low-Earth orbit satellite radio is retrieved and combined with evaluation indicators to perform service weight mapping. The service type corresponding to the current service carried by the low-Earth orbit satellite radio is read, and the evaluation indicators are read. The evaluation indicators include signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters. Following the order of the evaluation indicators—signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters—each evaluation indicator is first configured with the same basic weight; when there are six evaluation indicators, the basic weight corresponding to each evaluation indicator is 1 / 6. Subsequently, service-biased evaluation indicators are determined according to the service type. When the service type is a delay-sensitive service type, the coverage persistence parameter is determined as the service-biased evaluation indicator; when the service type is a bandwidth-sensitive service type, the resource availability parameter is determined as the service-biased evaluation indicator; when the service type is a high-reliability service type, the historical performance parameters are determined as the service-biased evaluation indicator. After determining the business bias evaluation indicators, multiply the base weight of each indicator by a preset enhancement factor to obtain the adjusted weight for that indicator; the remaining indicators retain their base weights. Add the adjusted weights to the base weights of the remaining indicators to obtain a total weight; then divide the weight of each indicator by the total weight to obtain the business bias weight coefficient for each indicator, ensuring that the sum of all business bias weight coefficients equals 1.

[0052] When weighting and fusing the objective weight coefficients and the service bias weight coefficients, the service type of the low-Earth orbit satellite walkie-talkie is first determined. Service types include latency-sensitive, bandwidth-sensitive, and high-reliability service types. When the service type is latency-sensitive, the service bias weight for coverage persistence parameters is increased to generate the first service bias weight. When the service type is bandwidth-sensitive, the service bias weight for resource availability parameters is increased to generate the second service bias weight. When the service type is high-reliability, the service bias weight for historical performance parameters is increased to generate the third service bias weight. Then, the first, second, or third service bias weights are linearly weighted with the objective weight coefficients according to a preset ratio to obtain the comprehensive weight vector corresponding to the network handover evaluation parameters.

[0053] Next, the ideal solutions are sorted, and the standardized decision matrix and comprehensive weight vector are read. The standardized decision matrix includes the standardized values ​​of each candidate network under signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters. The comprehensive weight vector includes the comprehensive weights corresponding to each evaluation index. According to the correspondence between candidate networks and evaluation indexes, the standardized value of each candidate network under each evaluation index is multiplied by the comprehensive weight corresponding to that evaluation index to obtain the weighted standardized value. Then, the weighted standardized values ​​of each candidate network are read column by column according to the evaluation index. Since the evaluation indexes have already undergone unidirectional normalization, the larger the value, the better the network switching conditions under the corresponding evaluation index. Therefore, the largest weighted standardized value in the same evaluation index column is determined as the positive ideal solution value corresponding to that evaluation index, and the smallest weighted standardized value in the same evaluation index column is determined as the negative ideal solution value corresponding to that evaluation index. Based on the evaluation indicators of signal received power data, coverage persistence parameters, resource availability parameters, transmission delay, packet loss rate, and historical performance parameters, the positive ideal solution value and negative ideal solution value corresponding to each evaluation indicator are determined in sequence to obtain the positive ideal solution and negative ideal solution.

[0054] Based on the comprehensive weight vector, when performing a proximity analysis between the candidate access network set and the positive ideal solution, for each candidate network in the candidate access network set, the weighted standardized value of the candidate network under each evaluation index is read, and the positive ideal solution value corresponding to each evaluation index in the positive ideal solution is also read. The weighted standardized value of the candidate network under the signal received power data is subtracted from the positive ideal solution value corresponding to the signal received power data to obtain the first index difference. Following the same method, the index differences corresponding to the coverage persistence parameter, resource availability parameter, transmission delay, packet loss rate, and historical performance parameter are calculated respectively. The squares of each index difference are squared and summed, and the square root of the sum is taken to obtain the first relative proximity degree of the candidate network relative to the positive ideal solution. The above calculations are performed one by one according to the order of the candidate networks in the candidate access network set to obtain the first relative proximity degree for each candidate network.

[0055] Based on the comprehensive weight vector, when performing a proximity analysis between the candidate access network set and the negative ideal solution, for each candidate network in the candidate access network set, the weighted standardized value of the candidate network under each evaluation index is read, and the negative ideal solution value corresponding to each evaluation index in the negative ideal solution is also read. The weighted standardized value of the candidate network under the signal received power data is subtracted from the negative ideal solution value corresponding to the signal received power data to obtain the second index difference. Following the same method, the index differences corresponding to the coverage persistence parameter, resource availability parameter, transmission delay, packet loss rate, and historical performance parameter are calculated respectively. The squares of each index difference are squared and summed, and the square root of the sum is taken to obtain the second relative proximity degree of the candidate network relative to the negative ideal solution. The above calculations are performed one by one according to the order of the candidate networks in the candidate access network set to obtain the second relative proximity degree for each candidate network.

[0056] Finally, data integration is performed based on the first and second relative proximity scores. For each candidate network in the candidate access network set, the first and second relative proximity scores are read, and the sum of the first and second relative proximity scores is obtained to get the total proximity score. Then, the second relative proximity score is divided by the total proximity score to obtain the handover fit score for that candidate network. The above data integration is performed one by one according to the order of the candidate networks in the candidate access network set to obtain multiple candidate network handover fit scores. The handover fit score is used to characterize the comprehensive proximity result of the candidate network relative to the positive ideal solution and the negative ideal solution.

[0057] Furthermore, in the method provided in the application embodiments, the objective weight coefficient and the business bias weight coefficient are weighted and fused to obtain a comprehensive weight vector, which further includes: The low-orbit satellite walkie-talkie's service types include latency-sensitive, bandwidth-sensitive, and high-reliability service types. If the service type is a latency-sensitive service type, the service bias weight of the coverage persistence parameter is increased to generate a first service bias weight. If the service type is a bandwidth-sensitive service type, the service bias weight of the resource availability parameter is increased to generate a second service bias weight. If the service type is a high-reliability service type, the service bias weight of the historical performance parameter is increased to generate a third service bias weight. The first service bias weight, the second service bias weight, or the third service bias weight are linearly weighted with the objective weight according to a preset proportional coefficient to obtain the comprehensive weight vector.

[0058] In this embodiment, the service types of the low-Earth orbit satellite walkie-talkie include latency-sensitive service types, bandwidth-sensitive service types, and high-reliability service types. The service type corresponding to the currently carried service of the low-Earth orbit satellite walkie-talkie is read, and the service bias weight coefficient and objective weight coefficient corresponding to each evaluation index are read in the order of evaluation indicators: signal received power data, coverage persistence parameters, resource availability parameters, transmission latency, packet loss rate, and historical performance parameters.

[0059] If the business type is a latency-sensitive business type, the business bias weight of the coverage persistence parameter is increased. The business bias weight coefficient corresponding to the coverage persistence parameter is multiplied by a preset increase factor to obtain the increased weight value corresponding to the coverage persistence parameter. The increased weight value is added to the business bias weight coefficients corresponding to the other evaluation indicators to obtain the weight sum value. Then, the increased weight value corresponding to the coverage persistence parameter is divided by the weight sum value, and the business bias weight coefficients corresponding to the other evaluation indicators are divided by the weight sum value respectively to obtain the first business bias weight.

[0060] If the service type is a bandwidth-sensitive service type, the service bias weight of the resource availability parameter is increased. The service bias weight coefficient corresponding to the resource availability parameter is multiplied by a preset increase factor to obtain the increased weight value corresponding to the resource availability parameter. The increased weight value is added to the service bias weight coefficients corresponding to the other evaluation indicators to obtain the weight sum value. Then, the increased weight value corresponding to the resource availability parameter is divided by the weight sum value, and the service bias weight coefficients corresponding to the other evaluation indicators are divided by the weight sum value respectively to obtain the second service bias weight.

[0061] If the service type is a high-reliability service type, the service bias weight of the historical performance parameters is increased. The service bias weight coefficient corresponding to the historical performance parameter is multiplied by a preset increase factor to obtain the increased weight value corresponding to the historical performance parameter. The increased weight value is added to the service bias weight coefficient corresponding to the other evaluation indicators to obtain the weight sum value. Then, the increased weight value corresponding to the historical performance parameter is divided by the weight sum value, and the service bias weight coefficient corresponding to the other evaluation indicators is divided by the weight sum value respectively to obtain the third service bias weight.

[0062] Finally, the first business bias weight, the second business bias weight, or the third business bias weight is linearly weighted with the objective weight according to a preset ratio coefficient. The first business bias weight, the second business bias weight, or the third business bias weight is selected according to the business type. In accordance with the order of evaluation indicators, the objective weight coefficient and the selected business bias weight corresponding to the same evaluation indicator are read one by one. The objective weight coefficient is multiplied by the preset ratio coefficient, and the selected business bias weight is multiplied by 1 and subtracted from the preset ratio coefficient. The two products are added together to obtain the comprehensive weight of the corresponding evaluation indicator. The comprehensive weight corresponding to each evaluation indicator is calculated in sequence and arranged in the order of evaluation indicators to obtain the comprehensive weight vector.

[0063] Step S400: Perform context migration through the target access network to achieve seamless communication handover, generate a seamless handover result for backtracking verification, and update the target access network based on the handover verification result.

[0064] In this embodiment, when performing seamless communication handover via context migration through the target access network, the communication context in the current access network is first read. The communication context includes the terminal identifier, session identifier, bearer configuration parameters, security key parameters, data packet sequence number, acknowledgment status, and pending data buffer. Then, a target access link is established based on the access parameters of the target access network, and access authentication, resource allocation, and bearer configuration are completed on the target access link. After the target access link meets the bearer requirements, the communication context is written to the target access network, and the data packet sequence number, acknowledgment status, and pending data buffer are synchronized to the target access network, enabling the target access network to continue sending and receiving service data according to the continuous sequence number of the original communication session. During the synchronization period… The system maintains the current access network connection and writes newly generated service data into the buffer queue. When the target access network returns a context reception confirmation, the service data in the buffer queue is sent sequentially through the target access network according to the data packet sequence number. After the target access network completes continuous transmission and reception of service data, the current access network is released, completing the seamless communication handover. The system records the handover trigger time, the target access link establishment time, the context migration completion time, the time when the first service data is successfully transmitted and received through the target access network, the time when the current access network is released, the reference signal received power data and the service data transmission and reception status during the handover process, and associates the above records with the target access network identifier, session identifier, and continuous status of data packet sequence number to generate a seamless handover result.

[0065] After generating seamless handover results, backtracking verification is performed based on these results. First, a handover evaluation is conducted on the seamless handover results, constructing actual performance indicators including handover interruption time and average reference signal received power. Then, a handover evaluation prediction is performed based on the target access network, constructing predictive performance indicators including predicted interruption time and predicted reference signal received power. Subsequently, the handover interruption time is compared with the predicted interruption time to calculate the interruption time deviation value, and the average reference signal received power is compared with the predicted reference signal received power to calculate the signal quality deviation value. When the interruption time deviation value is greater than a preset interruption deviation threshold or the signal quality deviation value is greater than a preset signal deviation threshold, the handover verification result is determined to be a verification failure. When the interruption time deviation value is not greater than the preset interruption deviation threshold and the signal quality deviation value is not greater than the preset signal deviation threshold, the handover verification result is determined to be a verification success.

[0066] Finally, when updating the target access network based on the handover verification result, the handover verification result is read first. If the handover verification result is successful, the target access network is determined to meet the preset performance expectations, and the target access network is retained as the current access result. If the handover verification result is unsuccessful, the packet retransmission rate and link recovery delay data in the seamless handover result are extracted, and auxiliary judgment is made based on the packet retransmission rate and link recovery delay data. If the packet retransmission rate exceeds the preset retransmission rate threshold or the link recovery delay data exceeds the preset recovery delay threshold, the target access network is determined to not meet the preset performance expectations, and an update instruction is generated to update and optimize the target access network.

[0067] Furthermore, the method provided in the application embodiments, in generating seamless switching results for backtracking verification, also includes: Based on the seamless handover results, a handover evaluation is performed to construct actual performance indicators, including handover interruption time and average reference signal received power. Based on the target access network, a handover evaluation prediction is performed to construct predicted performance indicators, including predicted interruption time and predicted reference signal received power. The handover interruption time is compared with the predicted interruption time to calculate the interruption time deviation. The average reference signal received power is compared with the predicted reference signal received power to calculate the signal quality deviation. If the interruption time deviation is greater than a preset interruption deviation threshold or the signal quality deviation is greater than a preset signal deviation threshold, the handover verification result is determined to be a verification failure. If the interruption time deviation is not greater than the preset interruption deviation threshold and the signal quality deviation is not greater than the preset signal deviation threshold, the handover verification result is determined to be a verification success.

[0068] In this embodiment of the application, when evaluating the handover based on the seamless handover results and constructing the actual performance indicators, the handover start time, the time when the first service data is successfully transmitted and received, and the reference signal received power data at each sampling time during the handover process are read from the seamless handover results; the handover start time is subtracted from the time when the first service data is successfully transmitted and received to obtain the handover interruption time; the reference signal received power data corresponding to each sampling time during the handover process are added together and divided by the number of sampling times to obtain the average reference signal received power; the handover interruption time and the average reference signal received power are used as the actual performance indicators.

[0069] Subsequently, handover assessment and prediction are performed based on the target access network to construct predictive performance indicators. The predicted interruption time and predicted reference signal received power of the target access network during the network handover assessment phase are read. The predicted interruption time and predicted reference signal received power are used as predictive performance indicators and are correlated with the handover interruption time and average reference signal received power in the actual performance indicators.

[0070] Next, the switching interruption time is compared with the predicted interruption time to calculate the interruption time deviation. The switching interruption time is subtracted from the predicted interruption time, and the absolute value of the difference is taken to obtain the interruption time deviation value. The average reference signal received power is compared with the predicted reference signal received power. When calculating the signal quality deviation value, the average reference signal received power is subtracted from the predicted reference signal received power, and the absolute value of the difference is taken to obtain the signal quality deviation value.

[0071] Next, the interruption time deviation value is compared with a preset interruption deviation threshold, and the signal quality deviation value is compared with a preset signal deviation threshold. If the interruption time deviation value is greater than the preset interruption deviation threshold, or the signal quality deviation value is greater than the preset signal deviation threshold, the handover verification result is determined to be a failure. If the interruption time deviation value is not greater than the preset interruption deviation threshold, and the signal quality deviation value is not greater than the preset signal deviation threshold, the handover verification result is determined to be a success.

[0072] Furthermore, in the method provided in the application embodiments, updating the target access network based on the handover verification result further includes: When the handover verification result is successful, it is determined that the target access network meets the preset performance expectations; when the handover verification result is unsuccessful, the data packet retransmission rate and link recovery delay data are extracted, and auxiliary judgment is made based on the data packet retransmission rate and the link recovery delay data: if the data packet retransmission rate exceeds the preset retransmission rate threshold or the link recovery delay data exceeds the preset recovery delay threshold, it is determined that the target access network does not meet the preset performance expectations, and an update instruction is generated to update and optimize the target access network.

[0073] In this embodiment of the application, the handover verification result is read. When the handover verification result is successful, the target access network is marked as meeting the preset performance expectations, and the target access network is retained as the access network for the current communication session of the low-orbit satellite walkie-talkie.

[0074] When the handover verification result is "verification failed," the packet retransmission rate and link recovery delay data are extracted and used for auxiliary judgment. Specifically, the total number of packets sent and the total number of packet retransmissions are read from the seamless handover result. The total number of packet retransmissions is divided by the total number of packets sent to obtain the packet retransmission rate. The link interruption time and the link recovery time are read, and the link recovery time is subtracted from the link interruption time to obtain the link recovery delay data. Then, the packet retransmission rate is compared with a preset retransmission rate threshold, and the link recovery delay data is compared with a preset recovery delay threshold.

[0075] If the packet retransmission rate exceeds a preset retransmission rate threshold or the link recovery delay data exceeds a preset recovery delay threshold, the target access network is determined to not meet the preset performance expectations, and an update instruction is generated to update and optimize the target access network. Specifically, when the packet retransmission rate is greater than the preset retransmission rate threshold or the link recovery delay data is greater than the preset recovery delay threshold, the target access network is determined to not meet the preset performance expectations. When generating the update instruction, the target access network identifier, packet retransmission rate, preset retransmission rate threshold, link recovery delay data, preset recovery delay threshold, and handover verification results are written into the update instruction, and the network handover evaluation of the candidate access network set is retried according to the update instruction to update and optimize the target access network.

[0076] In summary, the embodiments of this application have at least the following technical effects: This application retrieves real-time motion state information and low-Earth orbit satellite orbital parameters, maps them to a prediction time window for calculation, and constructs satellite-to-ground link quality degradation trajectory parameters. Based on these parameters, a handover determination is made, a handover signal is generated, and access analysis is performed using the handover signal in conjunction with ephemeris data and channel state information to construct a candidate access network set. Based on this set, network handover evaluation is conducted, calculating the handover suitability of multiple candidate networks. The candidate access network set is then traversed according to these suitability values ​​to determine the target access network. Seamless communication handover is performed through context migration via the target access network, and a seamless handover result is generated for backtracking verification. The target access network is then updated based on the handover verification result. This invention solves the technical problem of poor communication continuity during satellite-to-ground handover in existing low-Earth orbit satellite walkie-talkies. By constructing satellite-to-ground link quality degradation trajectory parameters for handover prediction and determining the target access network for context migration, it achieves the technical effect of improving communication continuity during satellite-to-ground handover.

[0077] Example 2 is based on the same inventive concept as the low-orbit satellite walkie-talkie seamless communication method based on satellite-to-ground switching prediction in the previous examples, such as... Figure 2 As shown, this application provides a seamless communication system for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction. The system and method embodiments in this application are based on the same inventive concept. The system includes: The calculation module 11 is used to retrieve real-time motion state information and low-orbit satellite orbit parameters and map them to a prediction time window for calculation to construct satellite-to-ground link quality degradation trajectory parameters; the handover determination module 12 is used to determine handover based on the satellite-to-ground link quality degradation trajectory parameters, generate a handover signal, and perform access analysis based on the handover signal combined with ephemeris data and channel state information to construct a candidate access network set; the handover evaluation module 13 is used to perform network handover evaluation based on the candidate access network set, calculate the handover suitability of multiple candidate networks, traverse the candidate access network set according to the multiple candidate network handover suitability to locate and determine the target access network; the communication handover module 14 is used to perform seamless communication handover by performing context migration through the target access network, generate seamless handover results for backtracking verification, and update the target access network according to the handover verification results.

[0078] Furthermore, the system is also used to implement the following functions: Based on real-time operational sensing of a low-Earth orbit (LEO) satellite intercom, the real-time motion state information is obtained, including inertial measurement data and satellite positioning data. A prediction time window is constructed, and the inertial measurement data and satellite positioning data are mapped to the prediction time window to predict the motion of the LEO satellite intercom, obtaining motion prediction trajectory parameters. The orbital parameters of the LEO satellite are introduced, and the motion prediction trajectory parameters and the orbital parameters are mapped to the prediction time window for sampling to obtain satellite elevation angle sequence and relative radial velocity sequence. Based on the satellite elevation angle sequence and the relative radial velocity sequence, a temporal convolutional network signal prediction is performed to generate a reference signal received power prediction sequence. Based on the reference signal received power prediction sequence, signal attenuation analysis is performed to calculate the signal attenuation rate. Based on the reference signal received power prediction sequence, interruption prediction is performed to calculate interruption prediction time data. The signal attenuation rate is integrated according to the interruption prediction time data to construct the satellite-to-ground link quality degradation trajectory parameters.

[0079] Furthermore, the system is also used to implement the following functions: Based on the degradation trajectory parameters of the satellite-to-ground link quality, a handover determination is made, and handover prediction conditions are set: a terrestrial communication network is introduced, and channel quality analysis is performed on the terrestrial communication network to obtain available signal quality parameters; if the estimated interruption time data is within the prediction time window and the available channel quality parameters are lower than a preset quality threshold, then the degradation trajectory parameters of the satellite-to-ground link quality are determined to meet the handover prediction conditions, and a handover signal is generated; or, an Earth shadow model is constructed, and based on the orbital parameters of the low-Earth orbit satellite and the Earth shadow model, it is determined whether the low-Earth orbit satellite walkie-talkie enters the shadow area of ​​the Earth shadow model during the estimated interruption time data. If it does, then the degradation trajectory parameters of the satellite-to-ground link quality are determined to meet the handover prediction conditions, and a handover signal is generated.

[0080] Furthermore, the system is also used to implement the following functions: By incorporating ephemeris data from adjacent low-Earth orbit (LEO) satellites and combining it with the motion prediction trajectory parameters, a satellite selection process is performed on LEO satellite walkie-talkies to form a candidate satellite subset. Signal reception power data is obtained based on the terrestrial communication network and compared with an access threshold. When the reference signal reception power data exceeds the access threshold, a candidate terrestrial network subset is formed. The candidate satellite subset and the candidate terrestrial network subset are merged to generate the candidate access network set.

[0081] Furthermore, the system is also used to implement the following functions: Based on the candidate access network set, network handover evaluation is performed, network handover evaluation parameters are extracted for handover adaptation analysis, and the handover adaptation degree of multiple candidate networks is calculated. The candidate access network set is sorted in descending order according to the multiple candidate network handover adaptation degrees to generate a candidate network ranking list. The candidate network ranking list is traversed to extract the first-ranked candidate network, which is selected as the candidate network with the highest handover adaptation degree to generate a network to be confirmed. The network to be confirmed is subjected to quality of service constraint verification. If the quality of service constraint verification of the network to be confirmed passes, the network to be confirmed is located as the target access network. If the quality of service constraint verification of the network to be confirmed fails, the next candidate network is traversed based on the candidate network ranking list as the network to be confirmed, and the quality of service constraint verification is repeated until the network to be confirmed meets the quality of service constraints, and the target access network is determined.

[0082] Furthermore, the system is also used to implement the following functions: Using the candidate access network set as the evaluation object and the network handover evaluation parameters as the evaluation indicators, a multi-attribute decision matrix is ​​constructed. The evaluation indicators are normalized in the same direction by traversing the multi-attribute decision matrix to generate a standardized decision matrix. Discrete analysis is performed on the evaluation indicators based on the candidate access network set, and entropy weight calculation is performed based on the indicator dispersion to obtain objective weight coefficients. The service types of low-orbit satellite walkie-talkies are retrieved and combined with the evaluation indicators to perform service weight mapping, calculating service bias weight coefficients. The objective weight coefficients and the service bias weight coefficients are weighted and fused to obtain a comprehensive weight vector, which is then used to approximate the ideal solution and sort it: the ideal solution includes positive and negative ideal solutions. Based on the comprehensive weight vector, the candidate access network set is analyzed for proximity to the positive ideal solution to calculate a first relative proximity. Based on the comprehensive weight vector, the candidate access network set is analyzed for proximity to the negative ideal solution to calculate a second relative proximity. Data is integrated based on the first and second relative proximity to generate a handover adaptability score.

[0083] Furthermore, the system is also used to implement the following functions: The low-orbit satellite walkie-talkie's service types include latency-sensitive, bandwidth-sensitive, and high-reliability service types. If the service type is a latency-sensitive service type, the service bias weight of the coverage persistence parameter is increased to generate a first service bias weight. If the service type is a bandwidth-sensitive service type, the service bias weight of the resource availability parameter is increased to generate a second service bias weight. If the service type is a high-reliability service type, the service bias weight of the historical performance parameter is increased to generate a third service bias weight. The first service bias weight, the second service bias weight, or the third service bias weight are linearly weighted with the objective weight according to a preset proportional coefficient to obtain the comprehensive weight vector.

[0084] Furthermore, the system is also used to implement the following functions: Based on the seamless handover results, a handover evaluation is performed to construct actual performance indicators, including handover interruption time and average reference signal received power. Based on the target access network, a handover evaluation prediction is performed to construct predicted performance indicators, including predicted interruption time and predicted reference signal received power. The handover interruption time is compared with the predicted interruption time to calculate the interruption time deviation. The average reference signal received power is compared with the predicted reference signal received power to calculate the signal quality deviation. If the interruption time deviation is greater than a preset interruption deviation threshold or the signal quality deviation is greater than a preset signal deviation threshold, the handover verification result is determined to be a verification failure. If the interruption time deviation is not greater than the preset interruption deviation threshold and the signal quality deviation is not greater than the preset signal deviation threshold, the handover verification result is determined to be a verification success.

[0085] Furthermore, the system is also used to implement the following functions: When the handover verification result is successful, it is determined that the target access network meets the preset performance expectations; when the handover verification result is unsuccessful, the data packet retransmission rate and link recovery delay data are extracted, and auxiliary judgment is made based on the data packet retransmission rate and the link recovery delay data: if the data packet retransmission rate exceeds the preset retransmission rate threshold or the link recovery delay data exceeds the preset recovery delay threshold, it is determined that the target access network does not meet the preset performance expectations, and an update instruction is generated to update and optimize the target access network.

[0086] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A seamless communication method for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, characterized in that, The method includes: Real-time motion status information and orbital parameters of low-Earth orbit satellites are retrieved and mapped to a prediction time window for calculation to construct trajectory parameters for satellite-to-ground link quality degradation. Based on the satellite-to-ground link quality degradation trajectory parameters, a handover determination is made, a handover signal is generated, and access analysis is performed according to the handover signal in combination with ephemeris data and channel state information to construct a candidate access network set. Based on the candidate access network set, network handover evaluation is performed, multiple candidate network handover adaptability is calculated, and the candidate access network set is traversed according to the multiple candidate network handover adaptability to locate and determine the target access network. Seamless communication handover is performed by performing context migration through the target access network, a seamless handover result is generated for backtracking verification, and the target access network is updated based on the handover verification result.

2. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 1, characterized in that, The method involves retrieving real-time motion status information and mapping the orbital parameters of low-Earth orbit satellites to a prediction time window for calculation, thereby constructing trajectory parameters for satellite-to-ground link quality degradation. The method includes: The real-time motion state information is obtained by real-time operation sensing based on a low-orbit satellite walkie-talkie, and the real-time motion state information includes inertial measurement data and satellite positioning data. A prediction time window is constructed, and the inertial measurement data and the satellite positioning data are mapped to the prediction time window to perform motion prediction on the low-orbit satellite walkie-talkie and obtain motion prediction trajectory parameters. By introducing the orbital parameters of low-Earth orbit satellites, the motion prediction trajectory parameters are mapped to the prediction time window for sampling to obtain the satellite elevation angle sequence and relative radial velocity sequence. Based on the satellite elevation angle sequence and the relative radial velocity sequence, a time-series convolutional network signal prediction is performed to generate a reference signal received power prediction sequence. Based on the reference signal received power prediction sequence, signal attenuation analysis is performed to calculate the signal attenuation rate. Based on the reference signal received power prediction sequence, interruption prediction is performed, and interruption prediction time data is calculated. The signal attenuation rate is integrated according to the estimated interruption time data to construct the satellite-to-ground link quality degradation trajectory parameters.

3. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 2, characterized in that, Based on the satellite-to-ground link quality degradation trajectory parameters, a handover determination is made and a handover signal is generated. The method includes: Based on the satellite-to-ground link quality degradation trajectory parameters, a handover determination is made, and handover prediction conditions are set: A terrestrial communication network is introduced, and channel quality analysis is performed on the terrestrial communication network to obtain usable signal quality parameters; If the interruption prediction time data is within the prediction time window and the available channel quality parameter is lower than the preset quality threshold, then the satellite-to-ground link quality degradation trajectory parameter is determined to meet the handover prediction condition, and a handover signal is generated. or, A shadow model of the Earth is constructed. Based on the orbital parameters of the low-Earth orbit satellite and the shadow model of the Earth, it is determined whether the low-Earth orbit satellite walkie-talkie enters the shadow area of ​​the Earth shadow model at the time of the interruption prediction. If it does, the trajectory parameters of the satellite-to-ground link quality degradation meet the handover prediction conditions, and a handover signal is generated.

4. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 3, characterized in that, Based on the switching signal combined with ephemeris data and channel state information, an access analysis is performed to construct a candidate access network set. The method includes: By incorporating ephemeris data from adjacent low-Earth orbit (LEO) satellites and combining it with the motion prediction trajectory parameters, a subset of candidate satellites is selected for LEO satellite walkie-talkies. Signal reception calculations are performed based on the terrestrial communication network to obtain signal reception power data, and the signal reception power data is compared with the access threshold. When the received power data of the reference signal exceeds the access threshold, it constitutes a subset of candidate ground networks; The candidate satellite subset and the candidate ground network subset are merged to generate the candidate access network set.

5. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 1, characterized in that, Based on the candidate access network set, network handover evaluation is performed, multiple candidate network handover adaptability scores are calculated, and the candidate access network set is traversed according to the multiple candidate network handover adaptability scores to locate and determine the target access network. The method includes: Based on the candidate access network set, network handover evaluation is performed, network handover evaluation parameters are extracted for handover adaptation analysis, and handover adaptation degree of multiple candidate networks is calculated. The candidate access network set is sorted in descending order according to the switching adaptability of the multiple candidate networks to generate a candidate network sorting list. The candidate network sorting list is traversed to extract the first-order candidate network. The first-order candidate network is selected as the candidate network with the highest switching fitness to generate the network to be confirmed. The network to be confirmed is subjected to a quality of service constraint verification. If the quality of service constraint verification of the network to be confirmed passes, the network to be confirmed is identified as the target access network. If the service quality constraint verification of the network to be confirmed fails, the next candidate network is traversed based on the candidate network sorting list as the network to be confirmed, and the service quality constraint verification is repeated until the network to be confirmed meets the service quality constraints, and the target access network is determined.

6. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 5, characterized in that, Based on the candidate access network set, network handover evaluation is performed, network handover evaluation parameters are extracted for handover adaptation analysis, and the handover adaptability of multiple candidate networks is calculated. The method includes: Using the candidate access network set as the evaluation object and the network handover evaluation parameters as the evaluation index, a multi-attribute decision matrix is ​​constructed. The evaluation indicators are normalized in the same direction by traversing the multi-attribute decision matrix to generate a standardized decision matrix. Based on the candidate access network set, the evaluation index is discretely analyzed, and the entropy weight is calculated based on the index dispersion to obtain the objective weight coefficient. The service types of low-orbit satellite walkie-talkies are retrieved and combined with the evaluation indicators to perform service weight mapping, and the service bias weight coefficient is calculated. The objective weight coefficient and the business bias weight coefficient are weighted and fused to obtain a comprehensive weight vector, which is then used to approximate the ideal solution and sort it. The ideal solution includes positive ideal solutions and negative ideal solutions; Based on the comprehensive weight vector, the candidate access network set and the positive ideal solution are subjected to a proximity analysis to calculate the first relative proximity. Based on the comprehensive weight vector, the candidate access network set and the negative ideal solution are subjected to a proximity analysis to calculate the second relative proximity. Data is integrated based on the first relative proximity and the second relative proximity to generate a switching adaptation score.

7. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 6, characterized in that, The objective weight coefficient and the business bias weight coefficient are weighted and fused to obtain a comprehensive weight vector. The method includes: The service types of the low-orbit satellite walkie-talkie include latency-sensitive service types, bandwidth-sensitive service types, and high-reliability service types; If the service type is the latency-sensitive service type, then the service bias weight covering the persistence parameter is increased to generate the first service bias weight. If the service type is the bandwidth-sensitive service type, then the service bias weight of the resource availability parameter is increased to generate a second service bias weight; If the service type is the high-reliability service type, then the service bias weight of the historical performance parameters is increased to generate a third service bias weight; The first business bias weight, the second business bias weight, or the third business bias weight are linearly weighted with the objective weight according to a preset ratio coefficient to obtain the comprehensive weight vector.

8. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 1, characterized in that, The methods for generating seamless handover results and performing backtracking verification include: Based on the seamless handover results, a handover evaluation is performed to construct actual performance indicators, which include handover interruption time and average reference signal received power. Based on the target access network, a handover evaluation and prediction are performed to construct a prediction performance index, which includes the prediction interruption time and the prediction reference signal received power. The switching interruption time is compared with the predicted interruption time, and the interruption time deviation value is calculated. The average reference signal received power is compared with the predicted reference signal received power to calculate the signal quality deviation value; If the interruption time deviation value is greater than the preset interruption deviation threshold or the signal quality deviation value is greater than the preset signal deviation threshold, then the handover verification result is determined to be a verification failure. If the interruption time deviation value is not greater than the preset interruption deviation threshold and the signal quality deviation value is not greater than the preset signal deviation threshold, then the handover verification result is determined to be successful.

9. The seamless communication method for low-orbit satellite walkie-talkies based on satellite-to-ground handover prediction as described in claim 8, characterized in that, The target access network is updated based on the handover verification result, including the following methods: When the switching verification result is successful, it is determined that the target access network meets the preset performance expectations; When the handover verification result is a failure, the packet retransmission rate and link recovery delay data are extracted, and auxiliary determination is made based on the packet retransmission rate and link recovery delay data: If the data packet retransmission rate exceeds a preset retransmission rate threshold or the link recovery delay data exceeds a preset recovery delay threshold, then the target access network is determined to not meet the preset performance expectations, and an update instruction is generated to update and optimize the target access network.

10. A seamless communication system for low-Earth orbit satellite walkie-talkies based on satellite-to-ground handover prediction, characterized in that, The system is used to execute the low-orbit satellite walkie-talkie seamless communication method based on satellite-to-ground handover prediction as described in any one of claims 1-9, and the system includes: The calculation module is used to retrieve real-time motion status information and low-orbit satellite orbit parameters, map them to the prediction time window for calculation, and construct the trajectory parameters for satellite-to-ground link quality degradation. The handover determination module is used to determine the handover based on the satellite-to-ground link quality degradation trajectory parameters, generate a handover signal, and perform access analysis according to the handover signal in combination with ephemeris data and channel state information to construct a candidate access network set. The handover evaluation module is used to perform network handover evaluation based on the candidate access network set, calculate the handover adaptability of multiple candidate networks, traverse the candidate access network set according to the multiple candidate network handover adaptability to locate and determine the target access network. The communication switching module is used to perform seamless communication switching by performing context migration through the target access network, generate seamless switching results for backtracking verification, and update the target access network based on the switching verification results.

Images