Low earth orbit satellite communication switching method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a low earth orbit satellite communication switching method and device, equipment and a storage medium. The method is applied to the technical field of low-orbit satellite wireless communication, and comprises the following steps: initializing parameter information of a multi-satellite coverage communication system, and determining satellite parameters, terminal distribution conditions and ground station coverage states; receiving a state attribute uploaded by a terminal in a coverage range of the communication system, and generating a corresponding state vector according to the uploaded state attribute; inputting the generated state vector into a reasoning model for reasoning, and determining a switching strategy of the terminal according to a reasoning result of the reasoning model; a terminal switching strategy is optimized through a multi-objective optimization algorithm, and inter-satellite switching is executed according to the optimized switching strategy, so that a dynamic network environment can be flexibly adapted, the time delay guarantee of high-priority services is improved, the system signaling overhead is reduced, and the overall switching efficiency is improved.

Description

Technical Field

[0001] This application relates to the technical field of low-earth orbit satellite wireless communication technology, and particularly to a low-earth orbit satellite communication handover method, device, equipment, and storage medium. Background Art

[0002] With the rapid development of communication technology, satellite communication systems, as the key link connecting the ground and space, play an irreplaceable role in realizing global seamless communication services. In the context of the increasing deepening of informatization and intelligence, low-earth orbit satellite networks, with their characteristics of low latency, high throughput, and wide-area coverage, have become an important technical means to support large-scale terminal connections. Specifically, the coverage area of a single low-earth orbit satellite can reach tens of thousands of square kilometers, which enables it to provide efficient and reliable network services for a wide range of user groups.

[0003] However, in low-earth orbit satellite networks, due to the high-speed movement of satellites far exceeding the terminal movement speed, the handover management between satellites and terminals has become a major technical problem. In scenarios with high terminal density and diverse service requirements, traditional handover strategies often face many challenges. The mechanism mainly based on single-user handover may lead to problems such as frequent handovers, a sharp increase in signaling overhead, and a decrease in handover success rate. Although the full-group handover can relieve the signaling pressure, it may affect the service quality and system stability due to uneven resource allocation. In addition, traditional methods lack flexibility and pertinence when dealing with high-priority services and are difficult to meet their requirements for low latency and high reliability.

[0004] The effectiveness of handover strategies is closely related to the intelligence of mode selection. However, there are still deficiencies in mode selection in current mainstream handover methods. The traditional mode division based on fixed thresholds or simple rules is difficult to adapt to the dynamic network environment, and the computational complexity increases significantly with the increase in the number of terminals, limiting its real-time performance. At the same time, existing solutions rarely pay attention to the differentiated requirements of service priorities. Especially in the two scenarios of having ground station coverage and not having ground station coverage, there is a lack of an adaptive collaborative optimization mechanism, resulting in low resource utilization and insufficient guarantee for key services. Summary of the Invention

[0005] This application provides a low-earth orbit satellite communication handover method, device, equipment, and storage medium to solve the deficiencies in mode selection of existing handover strategies.

[0006] In a first aspect, this application provides a low-earth orbit satellite communication handover method, which includes:

[0007] Initialize the parameter information of the multi-satellite coverage communication system, and determine satellite parameters, terminal distribution, and ground station coverage status;

[0008] Receive the status attributes uploaded by the terminals within the coverage of the communication system, and generate corresponding state vectors according to the uploaded status attributes;

[0009] Input the generated state vectors into the inference model for inference, and determine the handover strategy of the terminals according to the inference results of the inference model;

[0010] Optimize the terminal handover strategy through a multi-objective optimization algorithm, and perform inter-satellite handover according to the optimized handover strategy.

[0011] Optionally, the parameter information for initializing the multi-satellite coverage communication system, determining satellite parameters, terminal distribution, and ground station coverage status includes:

[0012] Initialize satellite parameters, and determine the number of available satellites, satellite orbital altitude, satellite operating speed, satellite coverage radius, orbital period, and inter-satellite link bandwidth within the current time window;

[0013] Initialize terminal parameters, and determine the number of terminals, the initial average density of terminals within the coverage of the communication system, the proportion of high-priority terminals, and the proportion of medium-priority terminals;

[0014] Determine the ground station coverage status according to the signal-to-noise ratio and the response of the ground station.

[0015] Optionally, the receiving the status attributes uploaded by the terminals within the coverage of the communication system, and generating corresponding state vectors according to the uploaded status attributes includes:

[0016] Determine the density distribution characteristics of the terminals within the coverage of the communication system according to the longitude and latitude of the terminals;

[0017] Determine the proportion of terminals with different priorities according to the priority identifiers of the terminals within the coverage of the communication system;

[0018] Obtain the source satellite load, source satellite signaling bandwidth margin, and neighboring satellite load, and determine the network status according to the source satellite load, source satellite signaling bandwidth margin, and neighboring satellite load;

[0019] Generate state vectors according to the density distribution characteristics of the terminals, the proportion of terminals with different priorities, and the network status.

[0020] Optionally, the inputting the generated state vectors into the inference model for inference, and determining the handover strategy of the terminals according to the inference results of the inference model includes:

[0021] Perform inference by calling the corresponding inference model according to the ground station coverage status;

[0022] Adopt an ∈-greedy strategy to select handover actions;

[0023] Determine the inter-satellite handover mode according to the handover action. The inter-satellite handover mode includes: full single-user handover, full group handover, hybrid handover, and hierarchical hybrid handover.

[0024] Optionally, the method of optimizing the terminal handover strategy through a multi-objective optimization algorithm and performing inter-satellite handover according to the optimized handover strategy includes:

[0025] Obtain a list of candidate satellites;

[0026] According to the inter-satellite handover mode, determine the optimization object and the objective function. The optimization object includes: the terminal or terminal group to be optimized, and the objective function includes: handover success rate, handover delay, load balancing, and signaling overhead;

[0027] Use a multi-objective optimization algorithm to perform multi-objective optimization according to the objective function and constraint conditions, and determine the target satellite corresponding to each terminal or terminal group;

[0028] Generate a handover instruction according to the target satellite corresponding to each terminal or terminal group, and perform inter-satellite handover according to the generated handover instruction.

[0029] Optionally, after optimizing the terminal handover strategy through a multi-objective optimization algorithm and performing inter-satellite handover according to the optimized handover strategy, the method further includes:

[0030] Optimize the inference model according to the handover result status, handover delay, target satellite load, and signaling overhead.

[0031] In a second aspect, the present application provides a low-earth orbit satellite communication handover device, and the device includes:

[0032] A first processing module, configured to initialize the parameter information of the multi-satellite coverage communication system, and determine satellite parameters, terminal distribution conditions, and ground station coverage status;

[0033] A second processing module, configured to receive the status attributes uploaded by the terminals within the coverage range of the communication system, and generate corresponding state vectors according to the uploaded status attributes;

[0034] A third processing module, configured to input the generated state vectors into an inference model for inference, and determine the handover strategy of the terminals according to the inference results of the inference model;

[0035] A fourth processing module, configured to optimize the terminal handover strategy through a multi-objective optimization algorithm and perform inter-satellite handover according to the optimized handover strategy.

[0036] In a third aspect, the present application provides a low-earth orbit satellite communication handover device, including:

[0037] A memory;

[0038] A processor;

[0039] Wherein, the memory stores computer-executable instructions;

[0040] The processor executes the computer-executable instructions stored in the memory to implement the low-earth orbit satellite communication handover method as described in the first aspect and various possible implementation manners of the first aspect above.

[0041] In a fourth aspect, the present application provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the low-earth orbit satellite communication handover method as described in the first aspect and various possible implementation manners of the first aspect above.

[0042] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the low-earth orbit satellite communication handover method as described in the first aspect and various possible implementation manners of the first aspect above.

[0043] The present application provides a low-earth orbit satellite communication handover method, device, equipment, and storage medium. The method initializes the parameter information of a multi-satellite coverage communication system, determines satellite parameters, terminal distribution conditions, and ground station coverage status; receives the status attributes uploaded by terminals within the coverage of the communication system, and generates corresponding state vectors according to the uploaded status attributes; inputs the generated state vectors into an inference model for inference, and determines the handover strategy of the terminals according to the inference results of the inference model; optimizes the terminal handover strategy through a multi-objective optimization algorithm, and performs inter-satellite handover according to the optimized handover strategy, improving the flexibility and practicality of the handover strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings here are incorporated into the specification and constitute a part of this specification, showing the embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0045] Figure 1 It is a flowchart of a low-earth orbit satellite communication handover method provided by an embodiment of the present application Figure 1 ;

[0046] Figure 2 It is a schematic diagram of a handover scenario provided by an embodiment of the present application;

[0047] Figure 3 It is a flowchart for calculating the normalization parameter D of the present invention max ;

[0048] Figure 4 It is a schematic structural diagram of a low-earth orbit satellite communication handover device provided by an embodiment of the present application;

[0049] Figure 5Schematic diagram of the structure of a low-earth orbit satellite communication switching device provided by an embodiment of the present application.

[0050] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and more detailed descriptions will be given later. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0051] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the drawings in the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0052] In the description and claims of the present invention and the above-mentioned drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein.

[0053] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific manner.

[0054] Figure 1 Flow schematic of a low-earth orbit satellite communication switching method provided by an embodiment of the present application Figure 1 。

[0055] As Figure 1 shown, the low-earth orbit satellite communication switching method provided in this embodiment includes:

[0056] S1: Initialize the parameter information of the multi-satellite coverage communication system, and determine satellite parameters, terminal distribution, and ground station coverage status.

[0057] Specifically, as Figure 2As shown in the figure, in the scenario of a low-earth orbit satellite network, the terminals are unevenly distributed. The multi-satellite cooperative coverage method is adopted to adapt to the distribution differences. Each satellite covers an independent area, and the areas cooperate through inter-satellite links (hereinafter referred to as ISLs). The dashed box represents the coverage area of a single satellite. Satellites S1, S2, S3, and S4 each serve their respective areas. Among them, there is ground-assisted coverage (hereinafter referred to as GAC) in the coverage areas of S3 and S4, and there is no ground-assisted coverage (hereinafter referred to as ASC) in the coverage areas of S1 and S2. To provide a basis for subsequent handover strategy selection, this step initializes the parameter information of the multi-satellite coverage communication system, specifically including:

[0058] S11: Initialize satellite parameters and determine the number of available satellites, satellite orbital altitude, satellite operating speed, satellite coverage radius, orbital period, and inter-satellite link bandwidth within the current time window.

[0059] Sat Params = {N sat , h, v, r, T, ISL}

[0060] Among them, N sat is the number of available satellites within the current time window, h is the satellite orbital altitude, v is the satellite operating speed, r is the satellite coverage radius, T is the orbital period, and ISL is the inter-satellite link bandwidth. The specific values are provided by system configuration or constellation design.

[0061] Specifically, N sat is dynamically determined by orbital parameters and visibility: According to the orbital data of the satellite constellation, calculate the number of satellites covering the target area within the current handover period. Satellites with an elevation angle greater than 10° are considered available. The number of satellites changes with time and is updated once per orbital period, but is considered fixed within a single handover process to simplify the calculation.

[0062] S12: Initialize terminal parameters and determine the number of terminals, the initial average density of terminals within the coverage of the communication system, the proportion of high-priority terminals, and the proportion of medium-priority terminals.

[0063] Term Dist = {N total , D init , P h , P m}

[0064] Among them, N total is the number of terminals, D init is the initial average density of the coverage area, P h is the high-priority proportion, and P m is the medium-priority proportion.

[0065] Specifically, in the GAC scenario, the ground station can estimate the average density of the coverage area through historical data or a geographic information system; in the ASC scenario, the on-board regeneration module (hereinafter referred to as OBR) can initially count the density distribution based on the latitude and longitude data uploaded by the terminal as the initial value. D init This is only for initialization. In subsequent S2, a more accurate density distribution feature will be calculated through the terminal's latitude and longitude.

[0066] S13: Determine the ground station coverage status according to the signal-to-noise ratio and the response of the ground station.

[0067]

[0068] Specifically, when 5 ≤ SNR ≤ 15 dB, the system initially sets G = 1 and attempts to use the GAC mode. However, the system will monitor the response of the ground station: if the ground station response times out or there is a data transmission error, then set G to 0 and keep G = 0 within the current switching period. In the next switching period, the system will re-detect the SNR and update G. This dynamic switching mechanism ensures that the system can flexibly adapt to the availability of the ground station when the SNR is in the intermediate range.

[0069] Furthermore, output the initialization parameters for subsequent state vector generation and switching mode selection.

[0070] S2: Receive the status attributes uploaded by the terminals within the coverage area of the communication system and generate corresponding state vectors according to the uploaded status attributes.

[0071] It can be understood that during the inter-satellite handover process of the user terminal, there will be signaling interactions among the user terminal, the source satellite, and the target satellite. If it is the handover process of multiple single-user terminals, a large amount of signaling overhead will be generated in the signaling interactions among the user terminal, the source satellite, and the target satellite, which may lead to network congestion. If multiple user terminals are divided into several groups for group handover, the repeated parts of the signaling interactions will be significantly reduced, thereby reducing the signaling overhead of the system, but it cannot effectively adapt to scenarios with multiple priorities. Therefore, adopting a handover strategy that combines single-user handover and group handover can reduce the signaling overhead while meeting the quality of service (QoS) requirements of different priorities.

[0072] Terminals with similar network states and service requirements often trigger inter-satellite handover requests within similar time periods. Therefore, for the selection of the handover mode, it is crucial to select appropriate attributes to generate state vectors for intelligent decision-making. To adapt to the dynamic environment of the low-earth orbit satellite network and improve the handover efficiency, the present invention considers from the following aspects:

[0073] Considering the dynamics of terminal distribution, the terminal density within the coverage of low-earth orbit satellites varies with different regions. Therefore, it is necessary to calculate the terminal density and its distribution characteristics in real time to reflect local differences, and the longitude and latitude of user terminals are selected as one of the state attributes.

[0074] In handovers, different priority services have different requirements for latency and reliability. High-priority user terminals tend to select target satellites that meet low latency and high reliability, while ordinary services are more suitable for group handovers to reduce signaling overhead. Therefore, it is necessary to distinguish different priority ratios, and the priority identifier of user terminals is selected as one of the state attributes.

[0075] Considering the real-time nature of the network state, the load and signaling bandwidth margin of low-earth orbit satellites directly affect the handover efficiency. To avoid network congestion and load imbalance, it is necessary to dynamically monitor satellite resources, and the source satellite load, signaling bandwidth margin, and neighboring satellite load are selected as one of the state attributes.

[0076] Based on the above state attributes, a state vector is generated, which provides a basis for subsequent handover mode selection.

[0077] In step S2, in the GAC scenario, the source satellite only collects terminal parameters and finally transmits them to the ground station for calculation and generation of the state vector. In the ASC scenario, the source satellite receives parameters through the OBR and generates the state vector. The specific process of generating the state vector for user terminals within the coverage includes:

[0078] S21: Determine the density distribution characteristics of terminals within the coverage of the communication system according to the longitude and latitude of the terminals.

[0079] (GAC scenario)

[0080] (ASC scenario)

[0081] Among them, in the GAC scenario, the ground station calculates the terminal position through the TDOA algorithm, divides the 1km×1km grid, D i =N i / 1, N i is the number of terminals in the i-th grid, N grids is the total number of grids; in the ASC scenario, the source satellite OBR divides the 5km×5km grid based on the longitude and latitude uploaded by the terminals, D j =N j / 25, N j is the number of terminals in the j-th grid, N grids is the total number of grids, 25km 2 is the area of the 5km×5km grid.

[0082] In addition, Davg is the average density, D var is the density variance, reflecting the degree of dispersion of the density distribution. Sort all grids by density value, and calculate the median density D median , that is, the 50th percentile, and the high-density threshold D high , that is, the 75th percentile, and the maximum density value D max,current to further describe the density distribution characteristics.

[0083] S22: Determine the proportions of terminals with different priorities according to the priority identifiers of the terminals within the coverage of the communication system.

[0084]

[0085] Among them, P h , P m are the proportions of the highest-priority and medium-priority terminals respectively, N h , N m are the numbers of the highest-priority and medium-priority terminals respectively, N total is the total number of terminals within the coverage. The priority identifier is 2 bits, that is, 10 = highest, 01 = medium, 00 = low, and the GAC scenario is counted by the ground station, and the ASC scenario is calculated by the OBR.

[0086] S23: Obtain the source satellite load, the source satellite signaling bandwidth margin, and the neighboring satellite load, and determine the network state according to the source satellite load, the source satellite signaling bandwidth margin, and the neighboring satellite load.

[0087]

[0088] Among them, Load sat represents the load of the source satellite, with a range of [0,1], which is used to measure the resource occupancy of the source satellite and prevent failures due to insufficient resources during the handover process. The calculation method is the average of the CPU usage percentage and the bandwidth usage percentage of the source satellite, which is monitored in real time by the source satellite OBR every second. CPU use represents the CPU usage percentage, and BWuse represents the bandwidth usage percentage;

[0089] B s represents the signaling bandwidth margin of the source satellite, with a range of [0,1], which reflects the bandwidth resources available for handover signaling of the source satellite and prevents signaling congestion. The calculation method is the ratio of the current available signaling bandwidth to the total signaling bandwidth, which is statistically calculated by the source satellite OBR every second in real time. Among them, Avail BW represents the current available signaling bandwidth, and TotalBW represents the total signaling bandwidth;

[0090] Load neiIndicates the average load of neighboring satellites, ranging from [0, 1], used to evaluate the resource status of the target satellite and avoid switching to high-load satellites. The calculation method is to obtain the average value of the loads of K neighboring satellites through ISL, where K is the number of neighboring satellites, and the specific number depends on the satellite topology, Load sat,k Indicates the load condition of the k-th satellite.

[0091] S24: Generate a state vector based on the density distribution characteristics of terminals, the proportion of terminals with different priorities, and the network status.

[0092] s = [D avg , D var , D mediam , D high , P h , P m , Load sat , B s , G, Load nei

[0093] Among them, D avg 、D var 、D median 、D high are density distribution characteristics; P h 、P m are priority ratios; G is the ground station coverage status.

[0094] Specifically, all parameter components are normalized to [0, 1]:

[0095] Among them, D avg 、D var 、D median 、D high Normalization requires normalizing the parameter D max , and the calculation steps are as Figure 3 shown, where the dynamic adjustment of D max and its upper and lower limits steps are:

[0096] Obtain the maximum grid density D max,current of the current scenario from step S21 and record it in the historical data (time window is the past 1 hour). Statistically analyze the D max,current distribution characteristics within the time window every hour: that is, the 5th percentile D max,5％ , the 95th percentile D max,95％ .

[0097] Calculate D max,min and D max,max , which is expressed as:

[0098] D max,min =(0.1, D max,5％ )​

[0099] D max,max = D max,95％ × 2

[0100] Update D using exponential moving average smoothing, expressed as: max , expressed as:

[0101] D max,t = α · D max,current + (1 - α) · D max,t-1 , α = 0.1

[0102] where D max,current is the maximum grid density of the current handover period, D max,t-1 is the normalization parameter at the previous moment, α is the smoothing factor, and the finally calculated D max,t is the current D max , and the initial value D max,0 = 1000 terminals / km 2 , and limit D max,t ∈ [D max,min , D max,max .

[0103] Finally, the normalized density component:

[0104]

[0105] where D var,max = (D max ) 2 .

[0106] In addition, P h , P m , Load sat , B s , Load nei , G is ensured to be in the range of [0, 1] when defined, and no additional normalization is required. And the state vector generation frequency is 1 second for GAC and 3 seconds for ASC.

[0107] S3: Input the generated state vector into the inference model for inference, and determine the handover strategy of the terminal according to the inference result of the inference model.

[0108] During the inter-satellite handover process in a low-earth orbit satellite network, the selection of the handover mode directly affects the handover efficiency and system performance. If a single single-user handover mode is adopted, the signaling overhead will increase significantly with the increase in the number of terminals, which may lead to network congestion and handover failures. If a full-group handover mode is adopted, although the signaling overhead is reduced, it may not be able to meet the low-latency requirements of high-priority services. Therefore, the present invention dynamically selects the handover mode according to the real-time state vector through the Deep Q-Network (abbreviated as DQN) algorithm to balance the QoS guarantee of different priority services and the optimization of signaling overhead.

[0109] The selection of the handover mode needs to comprehensively consider the terminal distribution, network status, and service requirements. The state vector contains key information such as the terminal density distribution characteristics, priority ratio, and network status, which can provide a comprehensive decision-making basis for DQN. To achieve intelligent inter-satellite handover mode selection, the present invention designs from the following aspects:

[0110] Considering the difference in terminal density distribution, in high-density areas, that is, D high is relatively high, and it is more suitable for group handover to reduce signaling overhead. In low-density areas, that is, D median is relatively low, and it is more suitable for single-user handover to ensure handover flexibility. Therefore, the density characteristics D avg 、D var 、D median 、D high in the state vector are important inputs for DQN.

[0111] Considering the difference in service priorities, the highest-priority and medium-priority services require low latency and high reliability, and are suitable for single-user handover or small-packet handover. Low-priority services are more suitable for large-packet handover. Therefore, the priority ratios P h 、P m in the state vector are the key basis for DQN.

[0112] Considering the dynamic nature of the network status, the source satellite load Load sat 、the signaling bandwidth margin B s and the average load Load nei of neighboring satellites reflect the current network resource status. DQN avoids handover to high-load satellites or failures caused by insufficient signaling through this information.

[0113] Based on the above state vector, DQN infers and outputs the handover action through reasoning, and further provides a basis for the subsequent execution of inter-satellite handover.

[0114] The specific process of inferring and outputting the handover action through DQN according to the state vector in step S3 includes:

[0115] S31: According to the GAC or ASC scenario, call the corresponding DQN model for inference, expressed as:

[0116] Q(s t , a) = DQN(s t ; θ)

[0117] Among them, in the GAC scenario, the ground station calls the complete DQN model, that is, the input layer has 10 dimensions, and the hidden layers have 64 and 32 neurons, and the parameter θ is updated in real time by the ground station.

[0118] In the ASC scenario, the source satellite OBR calls the lightweight DQN model, that is, the input layer has 10 dimensions, the hidden layers have 24 and 12 neurons, INT8 quantization, and the parameter θ is pre-trained.

[0119] Q(s t , a) represents the Q value of each action a in state s t where s t represents the state, a represents the action, and the action space includes: a = 0, that is, full single-user handover; a = 1, that is, full group handover; a = 2, that is, hybrid handover: highest and medium-priority single-user handover, low-priority group handover; a = 3, that is, graded hybrid handover: highest-priority single-user handover, medium-priority small-group handover, low-priority large-group handover.

[0120] S32: Adopt the ∈-greedy strategy to select the handover action, expressed as:

[0121]

[0122] where a t represents the handover action, argmax represents returning the input value that makes the function reach the maximum value, the ∈-greedy strategy balances exploration and exploitation, the initial value of ∈ is 1.0, and it decays to 0.01 with the number of training steps, that is, the decay rate is 0.995, and it is updated every step. In the GAC scenario, ∈ is dynamically adjusted by the ground station; in the ASC scenario, ∈ is locally adjusted by the OBR and updated every 100 handovers.

[0123] S33: Determine the inter-satellite handover mode according to the handover action, expressed as:

[0124]

[0125] where Mode represents the handover mode, Full Single-User Handover represents full single-user handover, FullGroup Handover represents full group handover, Hybrid Handover represents hybrid handover, and Graded HybridHandover represents graded hybrid handover.

[0126] It can be understood that if the output at = 0, full single-user handover is adopted, that is, regardless of high, medium, or low priority, all terminals perform handover in a single-user manner. Each terminal separately establishes a connection with the target satellite and executes the handover process; if the output at = 1, full group handover is adopted, that is, regardless of high, medium, or low priority, all terminals are divided into several groups, and each group performs handover as a whole, and the number of terminals within each group is between 50 and 500; if the output at = 2, hybrid handover is adopted, that is, high-priority and medium-priority terminals perform handover in a single-user manner, and low-priority terminals perform handover in a group manner, and the number of terminals within each group is between 50 and 500; if the output at = 3, hierarchical hybrid handover is adopted, that is, high-priority terminals perform handover in a single-user manner, medium-priority terminals perform handover in small groups, and the number of terminals within each group is between 20 and 100, and low-priority terminals perform handover in large groups, and the number of terminals within each group is between 50 and 500 terminals.

[0127] Among them, if hierarchical hybrid handover is adopted, first-level grouping will be performed, and the number of terminals within each group is between 500 and 2000, so as to perform resource pre-allocation and reduce resource overhead, and then second-level grouping will be performed, that is, medium-priority terminals are in small groups (20 - 100 terminals per group), and low-priority terminals are in large groups (50 - 500 terminals per group).

[0128] Specifically, the above grouping all adopts the K-means algorithm with a dynamically adjustable K value.

[0129] The purpose of this step is to output handover actions and modes for subsequent target satellite selection and handover execution.

[0130] The handover actions inferred by the DQN of the present invention can dynamically adapt to network state changes. In high-density areas, that is, when D high is relatively high, it tends to group handover. In low-density areas, that is, when D median is relatively low, it tends to single-user handover. When P h or P m is relatively high, it tends to hybrid or hierarchical hybrid handover.

[0131] S4: Optimize the terminal handover strategy through a multi-objective optimization algorithm, and perform inter-satellite handover according to the optimized handover strategy.

[0132] During the inter-satellite handover process in a low-earth orbit satellite network, the selection of the target satellite directly affects the handover success rate, latency, and system resource utilization. Therefore, the selection of the target satellite needs to comprehensively consider terminal requirements, network status, and handover mode. The handover mode has been determined by DQN in step S3, which can provide a clear optimization direction for the multi-objective optimization algorithm (Non-dominated Sorting Genetic Algorithm II, hereinafter referred to as NSGA-II). To achieve efficient selection and execution of the target satellite for inter-satellite handover, the present invention is designed from the following aspects:

[0133] Considering the importance of the handover success rate, the highest-priority terminals need to ensure the handover success rate first, the medium-priority terminals need a relatively high success rate, and ordinary terminals also need to maintain a basic success rate. Therefore, the handover success rate is taken as the primary optimization goal.

[0134] Considering the differentiated requirements for latency, the highest-priority services require the lowest latency, the medium-priority services require relatively low latency, and ordinary services allow relatively loose latency. Therefore, the handover latency is taken as the key optimization goal.

[0135] Considering the balance of system resources, the load and signaling overhead of the target satellite directly affect the handover efficiency and network stability. Therefore, load balancing and signaling overhead are taken as optimization goals.

[0136] Based on the above optimization goals, NSGA-II outputs the target satellite ID through multi-objective optimization, and further provides a basis for the execution of inter-satellite handover.

[0137] Specifically, it includes the following steps:

[0138] S41: Obtain the candidate satellite list, expressed as:

[0139] Sat List = {S1, S2,..., S K}

[0140] Among them, the candidate satellite list Sat List is calculated by the source satellite OBR and includes all visible satellites with an elevation angle greater than 10°. K is the number of candidate satellites, and the specific number depends on the satellite topology.

[0141] S42: According to the inter-satellite handover mode, determine the optimization object and the objective function. The optimization object includes: the terminal or group of terminals to be optimized, and the objective function includes: handover success rate, handover latency, load balancing, and signaling overhead, expressed as:

[0142] Targets = {O1, O2,..., O M}

[0143] Goals = {S h , Sm , S l , T h , T m , T l , B, C}

[0144] Among them, Targets are the terminals or groups to be optimized, M is the number of optimization objects, Goals represents the objective function, S h represents the handover success rate of the highest-priority terminals, S m represents the handover success rate of the medium-priority terminals, S l represents the handover success rate of the low-priority terminals, T h represents the handover latency of the highest-priority terminals, T m represents the handover latency of the medium-priority terminals, T l represents the handover latency of the low-priority terminals, B represents the load balancing B, and C represents the signaling overhead.

[0145] In addition, the optimization object for full single-user handover is each terminal; the optimization object for full group handover is the large group; for hybrid handover, the highest and medium-priority terminals are optimized separately, and the low-priority terminals are optimized in large groups; for hierarchical hybrid handover, the highest-priority terminals are optimized separately, the medium-priority terminals are optimized in small groups, and the low-priority terminals are optimized in large groups.

[0146] The objective function of NSGA-II comprehensively considers multiple optimization objectives, including the handover success rate. Among them, the handover success rate of the highest-priority terminals needs to be maximized to ensure the reliability of emergency communication, the handover success rate of the medium-priority terminals needs to be maximized to meet the quality of service requirements of paid services, and the handover success rate of the low-priority terminals also needs to be maximized to maintain the basic quality of service. In terms of handover latency, the latency of the highest-priority terminals needs to be minimized to achieve the lowest latency, the latency of the medium-priority terminals needs to be minimized to achieve a lower latency, and the latency of the low-priority terminals needs to be minimized but allows a looser latency. The load balancing is measured by the variance of the target satellite load and needs to be minimized to avoid over-concentration of target satellite resources. The signaling overhead is measured by the number of signaling messages and needs to be minimized to reduce the network burden. The signaling messages adopt a compression algorithm.

[0147] S43: Use a multi-objective optimization algorithm to perform multi-objective optimization according to the objective function and constraint conditions, and determine the target satellite corresponding to each terminal or terminal group, expressed as:

[0148] Sat ID = NSGA-II{S h , S m , S l , T h , T m , T l , B, C; Constraints}

[0149] Among them, Sat ID is the ID of each target satellite; the optimization objective is to maximize S h 、S m 、S l ,and minimize T h 、T m 、T l 、B, C; the constraint condition Constraints is that the bandwidth utilization rate of the target satellite is < 90%; and finally output the target satellite ID of each terminal or group.

[0150] Specifically, NSGA-II is executed by the ground station in the GAC scenario and by the OBR in the ASC scenario. Part of the calculation is distributed to neighboring satellites through ISL, and distributed computing is used to reduce the load. The population size is 50, and the number of iterations is 20. In the GAC scenario, the ground station directly executes NSGA-II, initializes the population (50 solutions), evaluates the objective function, performs non-dominated sorting, crowding degree calculation, selection, crossover, and mutation, and outputs the optimal solution after 20 iterations. In the ASC scenario, the OBR distributes the objectives to neighboring satellites. Each satellite executes the NSGA-II subtask. After aggregating the results through ISL, the OBR continues to iterate 20 times on the combined population and outputs the satellite ID.

[0151] S44: Generate a handover instruction according to the target satellite corresponding to each terminal or terminal group, and perform inter-satellite handover according to the generated handover instruction, which is expressed as:

[0152] Cmd = {ID, Sat ID, BW, Priority}

[0153] Among them, ID is the ID of a single user or a group user, Sat ID is the target satellite ID of the corresponding user, BW represents the allocated bandwidth, and Priority represents the priority identifier.

[0154] Specifically, generate a handover instruction according to the satellite ID. Bandwidth allocation is based on priority, that is, high-priority terminals reserve 15% of the bandwidth, medium-priority terminals reserve 10% of the bandwidth, and low-priority terminals are dynamically allocated the remaining bandwidth. If the signaling bandwidth margin is tight, that is, Bs < 0.3, then dynamically increase the large packet limit to reduce the number of signaling messages. Among them, in the GAC scenario, the ground station generates the instruction, the source satellite executes it, and the target satellite confirms it; in the ASC scenario, the source satellite OBR generates the instruction and executes it after negotiating with the target satellite through ISL. And the handover supports a retry mechanism: 5 times for high priority, 3 times for medium priority, and 1 time for low priority.

[0155] In an optional embodiment, after optimizing the terminal handover strategy through a multi-objective optimization algorithm and performing inter-satellite handover according to the optimized handover strategy, the method further includes: optimizing the inference model according to the handover result status, handover delay, target satellite load, and signaling overhead, expressed as:

[0156] Record={ID,Sat ID,Status,Delay,Load,Cost}

[0157] Where Record represents a record, Status represents the handover result status, Delay represents the handover delay, Load represents the target satellite load, and Cost represents the signaling overhead.

[0158] Specifically, record the handover status, delay, satellite load, and signaling overhead for use in DQN optimization, and the DQN reward function is:

[0159] R=w1·S h +w2·S m +w3·S l -w4·T h -w5·T m -w6·T l -w7·B-w8·C

[0160] Where w1, w2, w3, w4, w5, w6, w7, w8 represent weight coefficients, and w1 + w2 + w3 + w4 + w5 + w6 + w7 + w8 = 1.

[0161] The low-earth-orbit satellite communication handover method provided by the embodiments of the present application focuses on the dynamically changing satellite coverage environment, uses the on-board regeneration module and the ground station to work together, and dynamically selects the handover mode through deep reinforcement learning to ensure that the differentiated requirements of high-priority services and ordinary services are met. To improve the adaptability of the handover strategy, one of the state attributes selects the terminal density to achieve the accuracy of mode selection by calculating the terminal distribution characteristics in real time; in addition, in high-priority single-user handover, the terminal preferentially selects the target satellite that meets low latency and high reliability, while ordinary services tend to group handovers to reduce signaling overhead, so the second state attribute selects the service priority; to cope with the dynamic changes in network load, the last state attribute selects the satellite load and signaling bandwidth margin. Terminal density, service priority, and network load are key factors in low-earth-orbit satellite communication. Compared with traditional handover methods with fixed thresholds or single modes, the selection of state attributes in the present invention is more comprehensive and reasonable, and can effectively adapt to both scenarios with ground station coverage and without ground station coverage, improving the flexibility and practicality of the handover strategy.

[0162] Under the framework of multi-objective optimization, the present invention comprehensively considers four factors: handover success rate, handover latency, satellite load balancing, and signaling overhead. It establishes handover optimization objectives for each terminal or group through the multi-objective optimization algorithm NSGA-II, then schedules high-priority single-user handovers and ordinary group handovers according to the optimization results, and finally completes the execution of hybrid handovers. Scheduling based on real-time optimization objective values can maximize the overall performance of the system, significantly reduce signaling overhead, improve the quality of service guarantee for high-priority services, and optimize the load balancing of the satellite network.

[0163] Figure 4 It is a schematic structural diagram of a low-earth orbit satellite communication handover device provided by an embodiment of the present application. As Figure 4 shown, the low-earth orbit satellite communication handover device 300 provided in this embodiment includes:

[0164] The first processing module 301 is used to initialize the parameter information of the multi-satellite coverage communication system, and determine satellite parameters, terminal distribution, and ground station coverage status;

[0165] The second processing module 302 is used to receive the status attributes uploaded by the terminals within the coverage of the communication system, and generate corresponding state vectors according to the uploaded status attributes;

[0166] The third processing module 303 is used to input the generated state vectors into the inference model for inference, and determine the handover strategy of the terminal according to the inference results of the inference model;

[0167] The fourth processing module 304 is used to optimize the terminal handover strategy through a multi-objective optimization algorithm, and perform inter-satellite handovers according to the optimized handover strategy.

[0168] The low-earth orbit satellite communication handover device provided in this embodiment can execute the low-earth orbit satellite communication handover method provided in the above method embodiment. The implementation principle and technical effects are similar, and will not be elaborated here.

[0169] Figure 5 It is a schematic structural diagram of a low-earth orbit satellite communication handover device provided by an embodiment of the present application. As Figure 5 shown, the low-earth orbit satellite communication handover device provided in this embodiment of the present application, the low-earth orbit satellite communication handover device 400 includes: a receiver 401, a transmitter 402, a processor 403, and a memory 404.

[0170] The receiver 401 is used to receive instructions and data;

[0171] The transmitter 402 is used to send instructions and data;

[0172] The memory 404 is used to store computer execution instructions;

[0173] A processor 403 is configured to execute computer-executable instructions stored in a memory 404 to implement the respective steps performed by the low-Earth orbit satellite communication handover method in the above embodiments. For details, reference may be made to the relevant descriptions in the foregoing embodiments of the low-Earth orbit satellite communication handover method.

[0174] Optionally, the above-mentioned memory 404 may be either independent or integrated with the processor 403.

[0175] When the memory 404 is independently provided, the electronic device further includes a bus for connecting the memory 404 and the processor 403.

[0176] An embodiment of the present application further provides a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the low-Earth orbit satellite communication handover method performed by the above-mentioned low-Earth orbit satellite communication handover device.

[0177] An embodiment of the present application further provides a computer program product including a computer program, which, when executed by a processor, implements the above-mentioned low-Earth orbit satellite communication handover method.

[0178] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations. In the hardware implementation, the division between the functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, one physical component can have multiple functions, or one function or step can be executed by several physical components in cooperation. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that a communication medium typically contains computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0179] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present application. The present application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include the common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0180] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A low-orbit satellite communication switching method, characterized in that: The method comprises: Initialize the parameter information of the multi-satellite coverage communication system, determine the satellite parameters, terminal distribution and ground station coverage status; Receiving state attributes uploaded by a terminal within the coverage area of ​​the communication system, and generating a corresponding state vector according to the uploaded state attributes; The generated state vector is input into the inference model for inference, and the switching strategy of the terminal is determined according to the inference result of the inference model; The terminal switching strategy is optimized through a multi-objective optimization algorithm, and inter-satellite switching is performed according to the optimized switching strategy.

2. The method according to claim 1, characterized in that: The initialization of parameter information of the multi-satellite coverage communication system, determining satellite parameters, terminal distribution and ground station coverage status, includes: Initialize satellite parameters and determine the number of available satellites, satellite orbit altitude, satellite speed, satellite coverage radius, orbit period and inter-satellite link bandwidth within the current time window; Initialize terminal parameters and determine the number of terminals, the initial average density of terminals within the coverage area of ​​the communication system, the proportion of high-priority terminals, and the proportion of medium-priority terminals; The ground station coverage status is determined based on the signal-to-noise ratio and the ground station's response.

3. The method according to claim 1, characterized in that The receiving state attributes uploaded by a terminal within the coverage area of ​​the communication system and generating a corresponding state vector according to the uploaded state attributes includes: Determine the density distribution characteristics of the terminals within the coverage area of ​​the communication system according to the longitude and latitude of the terminals; Determine the proportion of terminals with different priorities according to the priority identifiers of the terminals within the coverage area of ​​the communication system; Obtaining a source satellite load, a source satellite signaling bandwidth margin, and an adjacent satellite load, and determining a network state according to the source satellite load, the source satellite signaling bandwidth margin, and the adjacent satellite load; A state vector is generated according to the density distribution characteristics of the terminals, the proportion of terminals with different priorities and the network status.

4. The method according to claim 1, characterized in that: The step of inputting the generated state vector into the inference model for inference, and determining the switching strategy of the terminal according to the inference result of the inference model includes: Call the corresponding inference model to perform inference according to the ground station coverage status; Adopt ∈-greedy strategy to select switching actions; The inter-satellite switching mode is determined according to the switching action, and the inter-satellite switching modes include: full single-user switching, full group switching, mixed switching and hierarchical mixed switching.

5. The method according to claim 1, characterized in that The method of optimizing the terminal switching strategy by a multi-objective optimization algorithm and performing inter-satellite switching according to the optimized switching strategy includes: Get the candidate satellite list; According to the inter-satellite handover mode, the optimization object and objective function are determined. The optimization object includes: the terminal or terminal group to be optimized, and the objective function includes: handover success rate, handover delay, load balancing and signaling overhead; A multi-objective optimization algorithm is used to perform multi-objective optimization according to the objective function and constraint conditions to determine the target satellite corresponding to each terminal or terminal group; A switching instruction is generated according to the target satellite corresponding to each terminal or terminal group, and inter-satellite switching is performed according to the generated switching instruction.

6. The method according to claim 5, characterized in that After optimizing the terminal switching strategy by a multi-objective optimization algorithm and performing inter-satellite switching according to the optimized switching strategy, the method further includes: The inference model is optimized according to the handover result status, handover delay, target satellite load and signaling overhead.

7. A low-orbit satellite communication switching device, characterized in that: The device comprises: The first processing module is used to initialize parameter information of the multi-satellite coverage communication system, determine satellite parameters, terminal distribution and ground station coverage status; A second processing module is used to receive state attributes uploaded by a terminal within the coverage area of ​​the communication system, and generate a corresponding state vector according to the uploaded state attributes; The third processing module is used to input the generated state vector into the inference model for inference, and determine the switching strategy of the terminal according to the inference result of the inference model; The fourth processing module is used to optimize the terminal switching strategy through a multi-objective optimization algorithm, and perform inter-satellite switching according to the optimized switching strategy.

8. A low-orbit satellite communication switching device, characterized in that: The device comprises: Memory; processor; Wherein, the memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the low-orbit satellite communication switching method as described in any one of claims 1-6.

9. A computer storage medium, characterized in that The computer storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the low-orbit satellite communication switching method as described in any one of claims 1-6.

10. A computer program product, characterized in that It includes a computer program, which, when executed by a processor, is used to implement the low-orbit satellite communication switching method as described in any one of claims 1 to 6.

Images