Low-orbit satellite beam switching method and communication terminal

Abstract

The invention discloses a low-orbit satellite beam switching method and a communication terminal, and belongs to the technical field of satellite communication, and the method comprises the steps: synchronously collecting real-time multi-source data; preprocessing the real-time multi-source data to obtain a real-time multi-dimensional feature vector, and based on the real-time multi-dimensional feature vector, performing link quality prediction on links of all visible beams in a period of time in the future by adopting a prediction model to obtain an updated candidate beam set, and providing a time window for decision making; and constructing a comprehensive decision function fused with a multi-dimensional optimization target, calculating a comprehensive score of the updated candidate beam set, identifying a global optimal beam, and performing switching judgment and execution. According to the invention, the switching strategy conversion from passive response to active optimization is realized, the time delay and interruption probability of beam switching are obviously reduced, and the system resource utilization rate and the communication quality are improved.

Description

Technical Field

[0001] This application relates to the field of satellite communications, and in particular to a low-Earth orbit satellite beam switching method and communication terminal. Background Technology

[0002] With the rapid development of Low Earth Orbit (LEO) satellite constellations, giant satellite constellations are building a globally covered space-air communication network. LEO satellite communication, with its advantages of low transmission latency, wide-area coverage, and large system capacity, has become key to filling gaps in terrestrial communication and constructing an integrated space-air-ground network. However, the high-speed movement of LEO satellites relative to ground communication terminals results in rapid movement of their ground-based points, and a short period of visibility between a single satellite and a fixed ground point. This fundamental characteristic causes communication link states to change rapidly over time, requiring user terminals to frequently switch beams between different satellite beams, and even between different beams of the same satellite, to maintain communication continuity.

[0003] In existing technologies, beam switching mainly employs a threshold-based reactive strategy. These methods typically rely on a single real-time measurement parameter, such as the reference received power (RSRP) or signal-to-noise ratio (SNR), to initiate a switching request to a beam with a stronger signal when the signal quality of the current serving link falls below a preset threshold. However, this passive switching mechanism has significant technical bottlenecks: First, it is only triggered after the link quality has deteriorated to a critical point, resulting in an extremely short decision window, which can easily lead to switching failure or communication interruption in high-speed satellite scenarios. Second, the decision-making is based on a single information dimension, relying solely on instantaneous signal strength and failing to fully utilize multi-dimensional information such as accurate satellite ephemeris, real-time terminal location, and Quality of Service (QoS) requirements, resulting in a crude decision-making process. Third, the rapid fluctuations in beam edge signals can easily trigger a ping-pong effect, where the terminal repeatedly and unnecessarily switches between two or more beams, increasing network signaling overhead and severely degrading the user's communication experience. Finally, existing terminals typically follow the principle of selecting the strongest signal to communicate with a single satellite, failing to coordinate the resource potential of multiple visible satellites at the system level, such as load balancing and link persistence, thus leading to low overall system resource utilization efficiency. Summary of the Invention

[0004] This application aims to provide a low-Earth orbit (LEO) satellite beam switching method and communication terminal. By designing a link quality prediction model and a multi-objective integrated decision function, it realizes a shift from passive response to active optimization in the switching strategy, significantly reducing the latency and interruption probability of beam switching, improving system resource utilization and communication quality, and solving the problems of frequent beam switching, communication interruption, and low resource utilization efficiency caused by the high-speed movement of satellites in LEO satellite communication.

[0005] To achieve the above objectives, the technical solution of this application is as follows: A method for low-Earth orbit satellite beam switching, comprising: Simultaneously collect satellite ephemeris data, real-time channel measurements, and global navigation satellite system position and service quality requirements of communication terminals from multiple satellites to form real-time multi-source data; Real-time multi-source data is preprocessed to obtain real-time multi-dimensional feature vectors. Based on the real-time multi-dimensional feature vectors, a prediction model is used to predict the link quality of all visible beams in the future, and an updated candidate beam set is obtained to provide a time window for decision-making. A comprehensive decision function integrating multi-dimensional optimization objectives is constructed, the comprehensive score of the updated candidate beam set is calculated, the globally optimal beam is identified, and switching decisions and execution are performed.

[0006] Optionally, based on real-time multidimensional feature vectors, a prediction model is used to predict the link quality of all visible beams within a future period, resulting in an updated candidate beam set, specifically including: The real-time multidimensional feature vector is input into the preset link quality prediction model to obtain the link quality prediction result sequence and confidence interval of all visible beams in the future period. Based on the link quality prediction result sequence, if the link quality prediction result of a certain beam is consistently higher than the communication threshold and the predicted available duration is greater than the minimum service quality requirement, it is retained in the candidate beam set; otherwise, it is removed from the candidate beam set to obtain the updated candidate beam set.

[0007] Optionally, the input data of the link quality prediction model includes: time-series features, spatial geometric features, and environmental features; the time-series features include: satellite ephemeris of multiple satellites and real-time channel quality of multiple satellites; the spatial geometric features include: the global navigation satellite system position of the communication terminal; the environmental features include: relevant parameters of the satellite-to-ground transmission environment and specific environmental parameters of the communication terminal.

[0008] Optionally, the link quality prediction results include: reference signal received power.

[0009] Optionally, a comprehensive decision function integrating multi-dimensional optimization objectives is constructed, the comprehensive score of the updated candidate beam set is calculated, the globally optimal beam is identified, and switching decisions and execution are performed, specifically including: Perform multi-objective intelligent decision-making, construct a comprehensive decision function that integrates multi-dimensional optimization objectives, adaptively select weight configuration according to the current business type, calculate the comprehensive score of each candidate beam in parallel for the updated candidate beam set, and identify the globally optimal beam. The handover decision and execution are performed as follows: if the globally optimal beam is not the currently serving beam, and the gain of the comprehensive score of the globally optimal beam relative to the comprehensive score of the currently serving beam exceeds a dynamic hysteresis threshold, then a beam handover is determined to be performed, and a handover command is generated to execute the beam handover; otherwise, the existing connection is maintained.

[0010] Optionally, the comprehensive decision function is expressed as follows: in, This represents the weighted summation of the overall score obtained from the i-th candidate beam; This represents the signal quality gain of the i-th candidate beam; This represents the service sustainability benefit of the i-th candidate beam; This represents the load balancing benefit of the i-th candidate beam; This represents the service matching benefit of the i-th candidate beam; This represents the switching overhead penalty for the i-th candidate beam; Each item represents its corresponding weight, and appropriate weights are assigned to each item based on the current business type.

[0011] Optionally, the dynamic hysteresis threshold is represented as follows: in, Indicates the dynamic hysteresis threshold; Indicates the basic threshold; This indicates the adjustment weight, which is adjusted based on the confidence level obtained from the link quality prediction model; This represents the confidence interval obtained based on the link quality prediction model.

[0012] Optionally, real-time multi-source data can be preprocessed to obtain real-time multi-dimensional feature vectors, including: performing data time alignment on the real-time multi-source data and encapsulating it into real-time multi-dimensional feature vectors.

[0013] A low-Earth orbit satellite communication terminal includes: terminal hardware and software system; and connection between the terminal hardware and software system. The terminal hardware includes: an antenna unit, a radio frequency unit, a baseband processing unit, a central processing and control unit, and a network and user interface unit; the antenna unit, radio frequency unit, baseband processing unit, central processing and control unit, and network and user interface unit are connected in sequence; the central processing and control unit is connected to the software system; A low-Earth orbit satellite beam switching software system is used to execute a low-Earth orbit satellite beam switching method as described above, including: a data fusion module, a prediction and inference module, a decision arbitration module, a beam control drive module, and a strategy configuration module; The data fusion module has its first end connected to the first end of the predictive inference module, and its second end connected to the first end of the decision arbitration module. The predictive inference module has its second end connected to the first end of the beam control drive module. The decision arbitration module has its second end connected to the second end of the beam control drive module, and its third end connected to the first end of the strategy configuration module. Strategy configuration module; The beam control driver module has its third terminal connected to the second terminal of the strategy configuration module.

[0014] Optionally, a data fusion module is used to simultaneously receive satellite ephemeris data from multiple satellites, real-time channel measurements, and the global navigation satellite system position and service quality requirements of the communication terminal to form real-time multi-source data; and to preprocess the real-time multi-source data to obtain real-time multi-dimensional feature vectors. The predictive inference module is used to predict the link quality of all visible beams within a future period based on real-time multidimensional feature vectors and a predictive model, thereby obtaining an updated set of candidate beams and providing a time window for decision-making. The decision arbitration module is used to construct a comprehensive decision function that integrates multi-dimensional optimization objectives, calculate the comprehensive score of the updated candidate beam set, identify the globally optimal beam, and make a switching decision. The beam control drive module is used to convert the switching command into the control signal required by the antenna unit, return it to the central processing and control unit, and drive the antenna unit to transmit the control signal of the switching command to the satellite to perform beam switching; The strategy configuration module is used to configure predefined business profiles and multi-objective intelligent decision-making and switching decision weight parameters.

[0015] The low-Earth orbit (LEO) satellite beam switching method and communication terminal proposed in this application upgrade beam switching from a passive response to an active prediction by introducing a link quality prediction model based on machine learning. This significantly extends the decision window and reduces the risk of switching failures and communication interruptions caused by high-speed satellite motion. By designing a multi-objective comprehensive decision function that integrates signal quality, service duration, satellite load, service matching, and switching overhead, and adopting a dynamic weighting mechanism, the method improves the resource utilization efficiency and load balance of the entire satellite constellation while ensuring user service experience. This achieves intelligent, smooth, precise, and adaptive beam switching management, solving the problems of frequent beam switching, communication interruptions, and low resource utilization efficiency caused by high-speed satellite motion in LEO satellite communication.

[0016] To make the above-mentioned features and advantages of the application more apparent and understandable, specific embodiments are provided below, and detailed descriptions are given in conjunction with the accompanying drawings. Attached Figure Description

[0017] Figure 1 This is a flowchart of the low-orbit satellite beam switching method proposed in this application.

[0018] Figure 2 This is a structural block diagram of the low-orbit satellite communication terminal proposed in this application. Detailed Implementation

[0019] To make the objectives and technical solutions of the embodiments of this application clearer, 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 some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the described embodiments of this application without creative effort are within the scope of protection of this application.

[0020] In one embodiment of this application, a low-Earth orbit satellite beam switching method is provided. Please refer to [link to relevant documentation]. Figure 1 , Figure 1 This is a flowchart of the low-Earth orbit satellite beam switching method proposed in this application, which includes steps S1 to S3.

[0021] Step S1: Simultaneously collect satellite ephemeris data, real-time channel measurements, and GNSS position and QoS requirements of communication terminals from multiple satellites to form real-time multi-source data; Step S2: Preprocess the real-time multi-source data to obtain real-time multi-dimensional feature vectors. Based on the real-time multi-dimensional feature vectors, use a prediction model to predict the future time period. Link quality prediction is performed on all visible beam links to obtain an updated candidate beam set, providing a time window for decision-making; Step S3: Construct a comprehensive decision function that integrates multi-dimensional optimization objectives, calculate the comprehensive score of the updated candidate beam set, identify the globally optimal beam, and make switching decisions and execute them.

[0022] The low-Earth orbit (LEO) satellite beam switching method proposed in this application upgrades beam switching from a passive response to an active prediction by introducing a link quality prediction model based on machine learning. This significantly extends the decision window and reduces the risk of switching failures and communication interruptions caused by high-speed satellite motion. By designing a multi-objective integrated decision function that integrates signal quality, service duration, satellite load, service matching, and switching overhead, and adopting a dynamic weighting mechanism, the method improves the resource utilization efficiency and load balance of the entire satellite constellation while ensuring user service experience. This achieves intelligent, smooth, precise, and adaptive beam switching management, solving the problems of frequent beam switching, communication interruptions, and low resource utilization efficiency caused by high-speed satellite motion in LEO satellite communication.

[0023] In step S1, please refer to Figure 1 In step S1, satellite ephemeris, real-time channel quality, global navigation satellite system position, and current service quality requirements of multiple satellites are collected simultaneously to form real-time multi-source data.

[0024] As an example, the process involves collecting satellite ephemeris data from multiple satellites, including parsing the satellite ephemeris (TLE files) of multiple satellites and calculating the orbits of future visible satellites; collecting real-time channel quality data from multiple satellites, including measuring channel quality indicators such as the reference signal received power and signal-to-noise ratio of the current serving beam and neighboring satellite beams; collecting the global navigation satellite system position of the communication terminal, including obtaining the communication terminal's precise latitude and longitude, moving speed and corresponding time, as well as the elevation angle, azimuth angle, distance and corresponding rate of change between the satellite and the communication terminal; and collecting the service quality requirements of the communication terminal, including obtaining the current service quality requirement level of the satellites.

[0025] As an example, the real-time multi-source data collected also includes: parameters related to the satellite-to-ground transmission environment, such as tropospheric parameters, ionospheric parameters, atmospheric absorption loss, etc., as well as specific environmental parameters of the communication terminal, such as environmental type classification, multipath reflection characteristics, ground reflection coefficient, etc.

[0026] As an example, the plurality of satellites includes all visible satellites whose current satellite beam coverage area includes the communication terminal used to perform the low-Earth orbit satellite beam switching method.

[0027] In step S2, please refer to Figure 1 Step S2 in the process involves preprocessing real-time multi-source data to obtain real-time multi-dimensional feature vectors. Based on these real-time multi-dimensional feature vectors, a prediction model is then used to predict future data for a given period. Link quality prediction is performed on all visible beam links to obtain an updated candidate beam set. The decision window is expanded, and global optimum is achieved through multi-objective optimization.

[0028] As an example, real-time multi-source data is preprocessed to obtain real-time multi-dimensional feature vectors, including: real-time time alignment of real-time multi-source data and encapsulation into real-time multi-dimensional feature vectors.

[0029] Furthermore, based on real-time multidimensional feature vectors, a prediction model is used to predict the future period. Link quality prediction is performed on all visible beam links to obtain an updated candidate beam set, specifically including steps S21 to S22.

[0030] Step S21: Input the real-time multidimensional feature vector into the preset link quality prediction model to obtain the results for a future period of time. Link quality prediction sequence and confidence intervals for all visible beams .

[0031] Step S22: Based on the link quality prediction result sequence, if the link quality prediction result of a certain beam is consistently higher than the communication threshold and the predicted available duration is greater than the minimum service quality requirement, it is retained in the candidate beam set; otherwise, it is removed from the candidate beam set to obtain an updated candidate beam set, ensuring the feasibility and effectiveness of the updated candidate beam set.

[0032] As an example, in step S21, the link quality prediction result includes: reference signal received power.

[0033] Specifically, the link quality prediction model employs a Long Short-Term Memory (LSTM) network. During training, the link quality prediction model is pre-trained based on massive historical satellite-to-ground link quality measurement data, and then lightweighted using model distillation technology before deployment on the communication terminal, supporting online fine-tuning to adapt to the specific environment of the communication terminal. The input layer of the link quality prediction model receives input data including temporal features, spatial geometric features, and environmental features; the LSTM network layer of the link quality prediction model captures the temporal dependence of link quality and the regular changes caused by satellite motion through an internal gating mechanism including forget gates, input gates, and output gates; the output layer of the link quality prediction model outputs the future time period. The sequence of reference signal received power prediction values ​​for all visible beams is used as the link quality prediction result sequence.

[0034] In actual prediction, the input data of the link quality prediction model includes: temporal features in the real-time multidimensional feature vector, spatial geometric features in the real-time multidimensional feature vector, and environmental features in the real-time multidimensional feature vector; Specifically, the temporal characteristics include: historical channel quality sequences of multiple satellites (such as historical observations of reference signal received power and signal-to-noise ratio); spatial geometric characteristics include: the global navigation satellite system position of the communication terminal, the real-time position of the satellites (calculated from ephemeris), and the relative geometric parameters between the satellites and the ground (including elevation angle, azimuth angle, and slant range); environmental characteristics include: parameters related to the satellite-to-ground transmission environment, such as tropospheric parameters, ionospheric parameters, atmospheric absorption loss, etc., as well as specific environmental parameters of the communication terminal, such as environmental type classification, multipath reflection characteristics, and ground reflection coefficient.

[0035] In one embodiment of this application, in step S21, a future period of time It lasts for 30 seconds.

[0036] In step S3, please refer to Figure 1 In step S3, a comprehensive decision function integrating multi-dimensional optimization objectives is constructed, the comprehensive score of the updated candidate beam set is calculated, the globally optimal beam is identified, and a switching decision and execution are performed.

[0037] As an example, a two-stage strategy of separating prediction and decision-making is adopted. First, the future link quality status of all visible beams is evaluated through the link quality prediction model. Then, the comprehensive score of all beams is calculated synchronously through the comprehensive decision function, and the best beam is switched.

[0038] As an example, step S3 specifically includes steps S31 to S32.

[0039] Step S31: Perform multi-objective intelligent decision-making, construct a comprehensive decision function that integrates multi-dimensional optimization objectives, adaptively select weight configuration according to the current service type, calculate the comprehensive score of each candidate beam in parallel for the updated candidate beam set, identify the globally optimal beam, and overcome the limitations of single satellite signal strength decision-making through multi-dimensional feature information fusion.

[0040] Step S32: Perform a handover decision and execution. If the globally optimal beam is not the currently serving beam, and the gain of the overall score of the globally optimal beam relative to the overall score of the currently serving beam exceeds a dynamic hysteresis threshold. If the signal is strong enough, a beam switching operation is initiated, and a switching command is generated to execute the beam switching; otherwise, the existing connection is maintained.

[0041] As an example, in step S31, the comprehensive decision function is expressed as follows: in, This represents the weighted summation of the overall score obtained from the i-th candidate beam; This represents the signal quality gain of the i-th candidate beam; This represents the service sustainability benefit of the i-th candidate beam; This represents the load balancing benefit of the i-th candidate beam; This represents the service matching benefit of the i-th candidate beam; This represents the switching overhead penalty for the i-th candidate beam; Each item represents its corresponding weight, and appropriate weights are assigned to each item based on the current business type.

[0042] in, and For the signal quality term of the comprehensive decision function, and To integrate decision function systems and business items, Switch the overhead term for the integrated decision function.

[0043] Specifically, the signal quality gain of the i-th candidate beam The link reliability is ensured by mapping the link quality prediction result of the i-th candidate beam to a gain score between 0 and 1 through a linear normalization function, as shown below: in, This represents the link quality prediction result for the i-th candidate beam; and These represent the minimum and maximum link quality, i.e., the reference signal received power reference values, as defined by the system.

[0044] Service continuity benefits of the i-th candidate beam The remaining visible time of the i-th candidate beam is calculated based on the precise locations of the satellite and communication terminal, and then normalized to avoid switching to a beam that is about to disappear. It is represented as follows: in, This represents the remaining visible time of the i-th candidate beam; This indicates the maximum visible time reference value.

[0045] Load balancing benefit of the i-th candidate beam By analyzing the load indication information of each satellite periodically broadcast by the low-Earth orbit satellite communication network, the load factor of each satellite is obtained, and the load reduction benefit of the i-th candidate beam is calculated. Satellites with lighter loads are selected first to improve the overall system capacity, as shown below: in, This represents the load factor of the i-th satellite obtained by analyzing the periodic broadcasts from the low-Earth orbit satellite communication network. This represents the system-defined full load threshold.

[0046] Specifically, the load indication information of each satellite periodically broadcast by the low-Earth orbit satellite communication network is obtained through the System Information Block (SIB) of the low-Earth orbit satellite communication network.

[0047] Service matching revenue of the i-th candidate beam The matching degree between the current business type and the inherent characteristics of the i-th candidate beam is calculated by querying a predefined business profile based on the current business service quality requirements, and is expressed as follows: in, The weights of each inherent characteristic of the candidate beam under service type b are represented. This represents the score vector of each inherent characteristic of the i-th candidate beam under service type b; the inherent characteristics of the i-th candidate beam include: theoretical bandwidth, propagation delay, link availability, and signal stability.

[0048] Specifically, the score vectors of each inherent characteristic of the i-th candidate beam under service type b. The following steps are used to obtain the scores: First, extract the quantized values ​​of the inherent characteristics of the i-th candidate beam from the real-time measurements and satellite system broadcast information; then, configure normalized scoring rules according to the actual low-Earth orbit satellite communication network's statistical or service quality requirements, mapping the original quantized values ​​of each parameter to standardized scores; next, based on the service quality requirements of service type b, perform nonlinear adaptation on the scores of key characteristics with hard threshold constraints or high sensitivity, such as hard truncation of the latency index for ultra-reliable low-latency communication services; finally, encapsulate the standardized scores of each inherent characteristic into a multi-dimensional scoring vector. As input to the comprehensive decision function, it is multiplied by the corresponding weight vector.

[0049] Specifically, business types b This includes: enhanced mobile broadband (eMBB) services, ultra-reliable low-latency communication (uRLLC) services, and massive machine-type communication (mMTC) services.

[0050] Switching overhead penalty for the i-th candidate beam A penalty factor is introduced to effectively suppress the ping-pong effect during handover; if the i-th candidate beam is the current beam, the handover overhead penalty for the i-th candidate beam is... The value is 0; if the i-th candidate beam is a new beam, then a base penalty value is applied, which is finely adjusted according to the link stability of the current i-th candidate beam, as follows: in, This represents a fixed penalty term to suppress the ping-pong effect of switching operations; This represents the variable penalty weights, relative to the link quality prediction results of the current serving beam. Inversely proportional to the current service beam link quality prediction results The worse the performance, the greater the variable penalty weight. The smaller the relative size.

[0051] Specifically, for enhancing mobile broadband services, significantly improving and The weighting prioritizes high signal quality and low network load; for ultra-reliable low-latency communication services, it significantly improves... and The weighting prioritizes low-latency paths and suppresses unnecessary handovers; for large-scale machine-type communication services: significantly improves... and The weighting prioritizes connection persistence and network coverage reliability.

[0052] In a specific embodiment of this application, for enhanced mobile broadband services, the weight vector corresponding to each sub-item is: For ultra-reliable low-latency communication services, the weight vector corresponding to each sub-item is: .

[0053] Furthermore, in the updated candidate beam set, the candidate beam with the highest comprehensive score is identified as the globally optimal beam.

[0054] As an example, in step S32, the dynamic hysteresis threshold It is expressed as follows: in, The basic threshold is determined by the service quality requirements of the current service type, the real-time moving speed and corresponding time of the terminal, the network load indication information, the handover overhead, and the specific environmental parameters of the communication terminal. This basic threshold is generated in real time by weighted fusion of the contribution values ​​of the above dimensions, providing a decision benchmark that matches the current scenario for the dynamic adjustment of the confidence level obtained in subsequent predictions. This indicates that the adjustment weights are adjusted based on the confidence level obtained from the link quality prediction model. When the prediction confidence level decreases, the adjustment weights are adjusted accordingly. The larger the value, the higher the dynamic hysteresis threshold. The larger the increment, the more conservative the switching decision, which can effectively suppress erroneous switching and ping-pong switching caused by inaccurate prediction. As the prediction confidence increases, the prediction confidence affects the dynamic hysteresis threshold. The weaker the influence, the more important the adjustment weight. The smaller the value, the more flexible the switching decision, ensuring timely switching when the link deteriorates and avoiding communication interruption; This represents the confidence interval obtained based on the link quality prediction model. Dynamic hysteresis threshold. It can adaptively adjust based on the confidence level of the link quality prediction results, ensuring more conservative decision-making when prediction uncertainty is high, and effectively suppressing ping-pong handover.

[0055] Furthermore, the switching decision context is updated to the historical database for subsequent switching decision optimization.

[0056] This application also provides a low-Earth orbit (LEO) satellite communication terminal for executing the aforementioned LEO satellite beam switching method. Please refer to... Figure 2 , Figure 2 The present application provides a structural block diagram of a low-Earth orbit (LEO) satellite communication terminal, which includes: terminal hardware 1 and software system 2; the terminal hardware 1 and software system 2 are connected.

[0057] As an example, terminal hardware 1 includes: antenna unit 11, radio frequency unit 12, baseband processing unit 13, central processing and control unit 14, and network and user interface unit 15; the antenna unit 11, radio frequency unit 12, baseband processing unit 13, central processing and control unit 14, and network and user interface unit 15 are connected in sequence; the central processing and control unit 14 is connected to the software system 2.

[0058] As an example, terminal hardware 1 uses a heterogeneous processing chip, which integrates an ARM Cortex-A78AE processor and a deep learning accelerator to run terminal hardware 1 and software system 2; wherein, the deep learning accelerator is used to efficiently run the link quality prediction model.

[0059] As an example, antenna element 11 includes: a multi-band phased array antenna array for supporting satellite communication frequency bands such as Ka or Ku, while forming multiple beams to receive signals and transmitting control signals to the satellite to perform beam switching.

[0060] The radio frequency unit 12 includes a low-noise amplifier, a power amplifier, a filter, and up / down converters, which are used to support the transmission and reception of various signals of the communication terminal.

[0061] The baseband processing unit 13 includes a modem and a global navigation satellite system signal receiver, used to provide centimeter-level positioning accuracy and nanosecond-level time synchronization, and to receive global navigation satellite system signals.

[0062] The central processing and control unit 14 includes: an intelligent beam switching manager and a beam control interface, used to execute the low-Earth orbit satellite beam switching software system 2 and send switching commands to the antenna unit 11; wherein, the intelligent beam switching manager is used to execute the low-Earth orbit satellite beam switching software system 2, and the beam control interface is used to send switching commands to the antenna unit 11.

[0063] The network and user interface unit 15 includes a 5G wireless network (Wi-Fi) and an Ethernet, used for service interaction between the communication terminal and the user.

[0064] As an example, all hardware components in the RF unit 12 are interconnected via a high-speed bus to ensure low latency in data processing and control.

[0065] In one specific embodiment of this application, the chip model of the Global Navigation Satellite System signal receiver specifically includes: u-blox ZED-F9P.

[0066] As an example, the low-Earth orbit satellite beam switching software system 2, used to execute the aforementioned low-Earth orbit satellite beam switching method, includes: a data fusion module 21, a prediction and inference module 22, a decision arbitration module 23, a beam control drive module 24, and a strategy configuration module 25.

[0067] The data fusion module 21 has its first end connected to the first end of the prediction and inference module 22, and its second end connected to the first end of the decision arbitration module 23. It is used to synchronously receive satellite ephemeris data, real-time channel measurements, and the global navigation satellite system position and service quality requirements of the communication terminal from multiple satellites to form real-time multi-source data. The real-time multi-source data is preprocessed to obtain a real-time multi-dimensional feature vector.

[0068] Predictive inference module 22, the second end of which is connected to the first end of beam control drive module 24, is used to predict future time based on real-time multidimensional feature vectors using a predictive model. Link quality prediction is performed on all visible beam links to obtain an updated candidate beam set, providing a time window for decision-making.

[0069] The decision arbitration module 23 is connected at its second end to the second end of the beam control drive module 24 and at its third end to the first end of the strategy configuration module 25. It is used to construct a comprehensive decision function that integrates multi-dimensional optimization objectives, calculate the comprehensive score of the updated candidate beam set, identify the globally optimal beam, and make a switching decision.

[0070] The beam control drive module 24, whose third end is connected to the second end of the strategy configuration module 25, is used to convert the switching command into the control signal required by the antenna unit 11 and return it to the intelligent beam switching manager of the central processing and control unit 14. The control signal that drives the antenna unit 11 to transmit the switching command to the satellite through the beam control interface is used to perform beam switching.

[0071] The strategy configuration module 25 is used to configure predefined business profiles and multi-objective intelligent decision-making and switching judgment weight parameters.

[0072] As an example, converting the switching command into the control signal required by the antenna element 11 includes: converting the switching command into the control signal required by the antenna element 11 by means of phase and amplitude representation.

[0073] The low-Earth orbit (LEO) satellite beam switching method and communication terminal provided in this application upgrade beam switching from a passive response to an active prediction by introducing a link quality prediction model based on machine learning. This significantly extends the decision window and reduces the risk of switching failures and communication interruptions caused by high-speed satellite motion. By designing a multi-objective comprehensive decision function that integrates signal quality, service duration, satellite load, service matching, and switching overhead, and adopting a dynamic weighting mechanism, the method improves the resource utilization efficiency and load balancing of the entire satellite constellation while ensuring user service experience. This achieves intelligent, smooth, precise, and adaptive beam switching management, solving the problems of frequent beam switching, communication interruptions, and low resource utilization efficiency caused by high-speed satellite motion in LEO satellite communication.

[0074] Although this application has been disclosed above with reference to embodiments, it is not intended to limit this application. Anyone skilled in the art may make some modifications and refinements without departing from the spirit and scope of this application. Therefore, the scope of protection of this application shall be determined by the appended claims.

Claims

1. A method for switching beams on low-Earth orbit satellites, characterized in that, include, Simultaneously collect satellite ephemeris data, real-time channel measurements, and global navigation satellite system position and service quality requirements of communication terminals from multiple satellites to form real-time multi-source data; Real-time multi-source data is preprocessed to obtain real-time multi-dimensional feature vectors. Based on the real-time multi-dimensional feature vectors, a prediction model is used to predict the link quality of all visible beams in the future, and an updated candidate beam set is obtained to provide a time window for decision-making. A comprehensive decision function integrating multi-dimensional optimization objectives is constructed, the comprehensive score of the updated candidate beam set is calculated, the globally optimal beam is identified, and switching decisions and execution are performed.

2. The low-orbit satellite beam switching method as described in claim 1, characterized in that, Based on real-time multidimensional feature vectors, a prediction model is used to predict the link quality of all visible beams within a future period, resulting in an updated candidate beam set, specifically including: The real-time multidimensional feature vector is input into the preset link quality prediction model to obtain the link quality prediction result sequence and confidence interval of all visible beams in the future period. Based on the link quality prediction result sequence, if the link quality prediction result of a certain beam is consistently higher than the communication threshold and the predicted available duration is greater than the minimum service quality requirement, it is retained in the candidate beam set; otherwise, it is removed from the candidate beam set to obtain the updated candidate beam set.

3. The low-orbit satellite beam switching method as described in claim 2, characterized in that, The input data of the link quality prediction model includes: time series features, spatial geometric features, and environmental features; the time series features include: satellite ephemeris of multiple satellites and real-time channel quality of multiple satellites; the spatial geometric features include: the global navigation satellite system position of the communication terminal; the environmental features include: relevant parameters of the satellite-to-ground transmission environment and specific environmental parameters of the communication terminal.

4. The low-orbit satellite beam switching method as described in claim 2, characterized in that, The link quality prediction results include: reference signal received power.

5. The low-orbit satellite beam switching method as described in claim 1, characterized in that, A comprehensive decision function integrating multi-dimensional optimization objectives is constructed, the comprehensive score of the updated candidate beam set is calculated, the globally optimal beam is identified, and switching decisions and execution are performed, specifically including: Perform multi-objective intelligent decision-making, construct a comprehensive decision function that integrates multi-dimensional optimization objectives, adaptively select weight configuration according to the current business type, calculate the comprehensive score of each candidate beam in parallel for the updated candidate beam set, and identify the globally optimal beam. The handover decision and execution are performed as follows: if the globally optimal beam is not the currently serving beam, and the gain of the comprehensive score of the globally optimal beam relative to the comprehensive score of the currently serving beam exceeds a dynamic hysteresis threshold, then a beam handover is determined to be performed, and a handover command is generated to execute the beam handover; otherwise, the existing connection is maintained.

6. The low-orbit satellite beam switching method as described in claim 5, characterized in that, The comprehensive decision function is expressed as follows: in, This represents the weighted summation of the overall score obtained from the i-th candidate beam; This represents the signal quality gain of the i-th candidate beam; This represents the service sustainability benefit of the i-th candidate beam; This represents the load balancing benefit of the i-th candidate beam; This represents the service matching benefit of the i-th candidate beam; This represents the switching overhead penalty for the i-th candidate beam; Each item represents its corresponding weight, and appropriate weights are assigned to each item based on the current business type.

7. The low-Earth orbit satellite beam switching method as described in claim 5, characterized in that, The dynamic hysteresis threshold is represented as follows: in, Indicates the dynamic hysteresis threshold; Indicates the basic threshold; This indicates the adjustment weight, which is adjusted based on the confidence level obtained from the link quality prediction model; This represents the confidence interval obtained based on the link quality prediction model.

8. The low-orbit satellite beam switching method as described in claim 1, characterized in that, Real-time multi-source data is preprocessed to obtain real-time multi-dimensional feature vectors, including: real-time time alignment of real-time multi-source data and encapsulation into real-time multi-dimensional feature vectors.

9. A low-orbit satellite communication terminal, characterized in that, include: Terminal hardware and software systems; Terminal hardware and software system connection; The terminal hardware includes: an antenna unit, a radio frequency unit, a baseband processing unit, a central processing and control unit, and a network and user interface unit; the antenna unit, radio frequency unit, baseband processing unit, central processing and control unit, and network and user interface unit are connected in sequence; the central processing and control unit is connected to the software system; A low-Earth orbit satellite beam switching software system for executing a low-Earth orbit satellite beam switching method as described in any one of claims 1 to 8, comprising: a data fusion module, a prediction and inference module, a decision arbitration module, a beam control drive module, and a strategy configuration module; The data fusion module has its first end connected to the first end of the predictive inference module, and its second end connected to the first end of the decision arbitration module. The predictive inference module has its second end connected to the first end of the beam control drive module. The decision arbitration module has its second end connected to the second end of the beam control drive module, and its third end connected to the first end of the strategy configuration module. Strategy configuration module; The beam control driver module has its third terminal connected to the second terminal of the strategy configuration module.

10. The low-orbit satellite communication terminal as described in claim 9, characterized in that, The data fusion module is used to simultaneously receive satellite ephemeris data, real-time channel measurements, and global navigation satellite system position and service quality requirements from multiple satellites, forming real-time multi-source data; and to preprocess the real-time multi-source data to obtain real-time multi-dimensional feature vectors. The predictive inference module is used to predict the link quality of all visible beams within a future period based on real-time multidimensional feature vectors and a predictive model, thereby obtaining an updated set of candidate beams and providing a time window for decision-making. The decision arbitration module is used to construct a comprehensive decision function that integrates multi-dimensional optimization objectives, calculate the comprehensive score of the updated candidate beam set, identify the globally optimal beam, and make a switching decision. The beam control drive module is used to convert the switching command into the control signal required by the antenna unit, return it to the central processing and control unit, and drive the antenna unit to transmit the control signal of the switching command to the satellite to perform beam switching; The strategy configuration module is used to configure predefined business profiles and multi-objective intelligent decision-making and switching decision weight parameters.

Images