High-efficiency scheduling method and system for large-scale beam hopping of low-orbit satellite based on interference approximation model

Abstract

The invention belongs to the technical field of low earth orbit satellite communication, and particularly relates to a low earth orbit satellite large-scale hopping beam efficient scheduling method and system based on an interference approximation model. According to the method, downlink beam resources of a multi-beam low-orbit satellite are used as research objects, and the problems that in existing large-scale and long-period beam scheduling, the performance evaluation complexity of a beam scheduling strategy is high, and the complexity of a beam scheduling algorithm is rapidly increased along with the increase of variable dimensions are solved. The invention provides low-complexity design and application based on an inter-beam interference approximation model, the inter-beam interference approximation model is introduced from a probability perspective, a rapid performance evaluation method is designed, and a low-complexity scheduling method is designed for two typical low-orbit satellite beam scheduling problems by utilizing an introduced beam activation intensity concept.

Description

Technical Field

[0001] This invention belongs to the field of low-Earth orbit satellite communication technology, specifically relating to an efficient scheduling method and system for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model. Background Technology

[0002] With the evolution of 6G networks, mobile communication networks will face exponentially increasing data demands, requiring support for diverse application scenarios such as high speed, high reliability, low latency, and high-density connections. However, large-scale, dense deployment of terrestrial base stations and backhaul networks incurs high costs for infrastructure, maintenance, and fiber optic installation. Simultaneously, current terrestrial cellular networks struggle to cover remote areas such as oceans, mountains, and the sky at low cost, failing to meet the ubiquitous communication requirements of future space-air-ground networks. Faced with this challenge, low-Earth orbit (LEO) satellites, with their low link loss, low transmission latency, and wide coverage, have become an effective supplement and extension to terrestrial networks, providing strong support for seamless coverage, high-speed data traffic, and massive connectivity services. LEO satellite communication systems, with their high throughput and wide coverage, are considered one of the promising solutions for the 6G era. However, the rapid growth of global communication services presents significant challenges to LEO satellite networks, particularly in terms of high data rates and stringent reliability requirements.

[0003] In low-Earth orbit (LEO) satellite operations, user demands exhibit significant time-varying and bursty characteristics. This necessitates dynamic adjustments to satellite configurations to adapt to changing user needs and improve communication performance. Examples include adjusting beam angles, beam activation states, and beam transmit power. During these adjustments, continuous performance evaluation of the current configuration is required, assessing metrics such as throughput and outage probability. Traditional performance evaluation methods based on beam-by-beam and time-slot signal-to-interference-plus-noise ratio (SIR / NNR) calculations suffer from high computational complexity in highly dynamic, long-cycle scheduling scenarios, directly leading to increased scheduling delays and impacting communication service quality. Therefore, developing low-complexity, rapid performance evaluation methods for large-scale, long-cycle beam scheduling scenarios is of significant application value for achieving real-time optimization of satellite configurations.

[0004] A key technology in modern satellite systems is multi-beam technology, which can deploy hundreds of beams to cover vast areas and serve more users, thereby increasing system capacity. However, as the number of beams increases, beam scheduling and resource management become more complex. Existing beam scheduling solutions are based on two approaches: 1. Determining the optimal beam scheduling for each time slot. Optimization variables (such as power or beam state) often appear simultaneously in both the numerator and denominator of the signal-to-noise ratio (SNR), leading to a non-convex problem. Furthermore, the dimensionality of the variables increases rapidly with scheduling periods, the number of beams, and the number of satellites. In large-scale scheduling problems, high-dimensional variables make the solution process slow and impractical. 2. Determining the allocation time for each beam combination. In this case, each subset of beams is associated with a variable representing the activation time of that subset. However, in large-scale beam systems, the number of beam combinations grows exponentially with the number of beams, making optimization impractical. In existing technologies, low-Earth orbit (LEO) satellite beam resource scheduling methods often exhibit high computational complexity with increasing beam size, leading to decreased optimization performance or even incomputability. Therefore, how to design a low-complexity beam scheduling method to meet the high-efficiency communication requirements of large-scale beam scenarios is an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method and system for efficient scheduling of large-scale beam hopping for low-Earth orbit satellites based on an interference approximation model, in order to solve the technical problems of complex calculations for evaluating the communication performance of low-Earth orbit satellites and high complexity of beam scheduling methods in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: An efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model includes the following steps: Configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites; Based on the downlink communication channel model of multi-beam low-Earth orbit satellites, the activation state of the beams in each time slot is determined, and the beam activation matrix is ​​generated. Introducing beam activation intensity, and based on the beam activation matrix, calculating the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period; Based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam, calculate the achievable rate of the user served by the beam in any time slot. Based on the reachable rate of the users served by the beam in any time slot, the beam activation intensity vector is solved according to the user request rate. Based on the beam activation intensity vector, a beam candidate matrix is ​​generated, resulting in a low-complexity hopping beam scheduling strategy to solve the capacity demand ratio problem. The beam activation intensity vectors are sorted in descending order, and the minimum number of scheduling slots is determined based on the maximum value of the beam activation intensity, thus generating a hopping beam scheduling strategy to solve the minimum time scheduling problem.

[0007] The determination of the activation state of the beam in each time slot specifically involves defining: B Each beam in the time slot t The total user demand covered at that time was And each beam has the same transmission power, that is , Define the total satellite transmit power; define the beam activation matrix. Indicates the period of beam scheduling T The mid-wave beam activation state is indicated by an element of 1 indicating activation and an element of 0 indicating deactivation.

[0008] The generation of the beam activation matrix specifically involves: first, constructing a weight parameter that reflects the beam priority. W As the basis for randomly selecting the active beam, the weight parameters of each beam are then normalized to obtain the beam. b In the time slot t Activation probability at time P b,t Furthermore, a pseudo-random number generator is used to generate a floating-point number in the interval [0, 1). , indicating time slot t The activation threshold is set when the activation probability is greater than or equal to the activation threshold. beam b In the time slot t Activated, assigned a value Otherwise, it will not be activated and a value will be assigned. When scheduling a time slot is completed, the beam activation state is assigned to the first value of matrix x. t Column, update beam b exist t The average allocation ratio and weight parameters within the scheduling cycle prior to the current time are calculated iteratively. t The beam activation probability of the +1 time slot generates a new activation threshold, resulting in... t The beam activation state of the +1 time slot continues until the cycle is complete. T The beam activation matrix x is obtained by scheduling time slots.

[0009] The introduction of beam activation intensity specifically involves: for large-scale beam scheduling systems, introducing a new variable vector. , Defined as a beam within a long-period large-scale beam scheduling window. bThe frequency at which the beam is activated is called the beam activation intensity.

[0010] The step of calculating the beam activation intensity of the beam activation matrix based on the beam activation matrix specifically involves: for a given beam activation matrix... beam b The beam activation intensity is expressed as .

[0011] The step of calculating the average interference experienced by the beam during the beam scheduling period based on the beam activation matrix specifically involves: within the beam scheduling period, the beam... b 'Beam b Service users j The resulting interference can be approximated as: ,in, Indicates beam The transmission power.

[0012] The calculation of the achievable rate for the users served by the beam in any time slot, based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam, specifically involves: T Within a time slot of the beam scheduling period, for any beam b Service users j The beam is replaced by the beam activation intensity of the beam activation matrix and the average interference experienced by the beam. b In the time slot t The interference term in the signal-to-interference-plus-noise ratio (SIR) formula for the received signal is weighted and summed using beam activation intensity as the weight to obtain the user's signal. j The signal-to-interference-plus-noise ratio is According to the user j Signal-to-interference-to-noise ratio Calculate users The achievable rate in any time slot is expressed as: .

[0013] The beam activation intensity vector is calculated based on the achievable rate of the user served by the beam in any time slot, according to the user's requested rate. Specifically, it is assumed that the user... j exist T The request rate per time slot is Based on the achievable rate of the users served by the beam in any time slot, since it is represented as the user j The achievable rate in any time slot, which is the average achievable rate over the entire scheduling period, therefore, the beam b Need for users j Serve One time slot is enough to meet the user's request rate; at this time, the beam b The beam activation intensity can be expressed as Substituting the achievable rates of the users served by the beam in any time slot, the beam... b The beam activation intensity is further expressed as ,in, beam b Beam activation intensity is related to Related functions and Irrelevant; based on the beam activation strength of beam b, at a given user request rate Number of time slots T ,bandwidth W B Channel gain g Noise power N 0, and transmission power P In this case, the beam activation intensity vector can be solved efficiently and quickly using a linearly convergent fixed-point iterative method. .

[0014] The method of generating a beam candidate matrix based on the beam activation intensity vector and generating a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem is as follows: generating a beam candidate vector according to the beam activation intensity, generating a beam candidate matrix by arranging the beam candidate vectors into randomly scrambled copies, and randomly extracting one column from each beam in turn from its beam candidate matrix to generate a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem.

[0015] The process involves arranging the beam activation intensity vectors in descending order, determining the minimum number of scheduling time slots based on the maximum value of the beam activation intensity, and generating a hopping beam scheduling strategy to solve the minimum time scheduling problem. Specifically, the beam activation intensity vectors are arranged in descending order, and the maximum value in the vector determines the shortest time scheduling required to complete the transmission. That is, the beam activation vectors are adjusted and updated sequentially according to the maximum value. Based on the updated beam activation intensities, a new beam candidate matrix is ​​generated. By randomly selecting a column from the beam candidate matrix, a beam scheduling strategy to solve the minimum time scheduling problem is generated.

[0016] A high-efficiency scheduling system for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model includes a channel modeling module, an initial scheduling generation module, an interference approximation modeling module, a fast performance evaluation module, an activation intensity solving module, and a scheduling strategy generation module. The channel modeling module is used to configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites. The initial scheduling generation module is used to determine the activation state of the beams in each time slot and generate a beam activation matrix based on the multi-beam low-orbit satellite downlink communication channel model. The interference approximation modeling module is used to introduce beam activation intensity, and calculate the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period based on the beam activation matrix. The rapid performance evaluation module is used to calculate the achievable rate of the user served by the beam in any time slot based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam. The activation intensity calculation module is used to calculate the beam activation intensity vector based on the achievable rate of the user served by the beam in any time slot and according to the user request rate. The scheduling strategy generation module is used to generate a beam candidate matrix based on the beam activation intensity vector, generate a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem, sort the beam activation intensity vectors in the beam candidate matrix in descending order, determine the minimum number of scheduling time slots based on the maximum value of the beam activation intensity, and generate a skip beam scheduling strategy to solve the minimum time scheduling problem.

[0017] Compared with the prior art, the present invention has the following beneficial effects: This invention proposes a novel approximate model for inter-beam interference (IBI). This model, from a probabilistic perspective, can rapidly calculate communication performance indicators such as signal-to-interference-plus-noise ratio (SIR) and achievable data rate (ADR). Existing performance evaluation methods are based on slot-by-slot and beam-by-beam SIR calculations. Traditional methods are highly complex in large-scale, long-period beam scheduling strategies, making it difficult to meet the low-latency and high-efficiency requirements of satellite beam management. In particular, because user demand for satellite services is time-varying and abrupt, satellites need to continuously adjust beam strategies during service delivery. The high complexity of traditional performance evaluation methods increases strategy output delay, thereby increasing system latency. Therefore, the proposed rapid performance evaluation method avoids slot-level calculations and, from a statistical perspective, can efficiently obtain long-term average performance with low complexity, achieving an accuracy of over 95% compared to traditional methods.

[0018] This invention proposes the concept of beam activation intensity and two low-complexity strategies for solving multi-beam satellite beam scheduling problems. Existing beam scheduling algorithms based on capacity demand ratio problems exhibit exponentially increasing complexity with the number of beams. In large-scale beam scheduling, traditional algorithms suffer from high computational latency, thus losing their timeliness. The proposed low-complexity method based on beam activation intensity can quickly compute scheduling strategies in large-scale beam scheduling, significantly improving the system's response speed, with performance within 6% of the traditional optimal algorithm. Existing beam scheduling algorithms based on minimizing scheduling time problems also experience increasing complexity with the number of beams and scheduling time slots. In large-scale, long-period beam scheduling, the slow computation of traditional algorithms leads to a significant increase in transmission latency. The proposed low-complexity algorithm based on beam activation intensity can quickly and efficiently approximate the performance of traditional algorithms and is suitable for large-scale, long-period beam scheduling scenarios. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating a multi-beam low-orbit satellite downlink communication scenario, specifically implemented according to the present invention. Figure 2 This is a performance comparison chart of the rapid performance evaluation method of the present invention under different beam counts. Figure 3 This is a performance comparison chart of the rapid performance evaluation method of the present invention under three demand scenarios. Figure 4 This is a performance comparison chart of a low-complexity method based on the capacity-demand ratio problem in a specific implementation of the present invention. Figure 5 This is a performance comparison chart of a low-complexity method based on the minimum time scheduling problem in a specific implementation of the present invention. Detailed Implementation

[0020] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] Example 1 This embodiment proposes an efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model, including the following steps: Configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites; Based on the downlink communication channel model of multi-beam low-Earth orbit satellites, the activation state of the beams in each time slot is determined, and the beam activation matrix is ​​generated. Introducing beam activation intensity, and based on the beam activation matrix, calculating the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period; Based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam, calculate the achievable rate of the user served by the beam in any time slot. Based on the reachable rate of the users served by the beam in any time slot, the beam activation intensity vector is solved according to the user request rate. Based on the beam activation intensity vector, a beam candidate matrix is ​​generated, resulting in a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem. The beam activation intensity vectors in the beam candidate matrix are sorted in descending order, and a skip beam scheduling strategy to solve the minimum time scheduling problem is generated based on the maximum value of the beam activation intensity.

[0023] Based on the above steps, this embodiment provides a detailed description, and the specific implementation method is as follows: Step 1: Assume a multi-beam low-Earth orbit satellite is in Ku The frequency band provides downlink communication services to ground users. The satellite carries multiple phased array antennas, generating multiple directional beams, each covering a specific area on the ground. Users within the beam coverage area are equipped with a unidirectional antenna. Parameters for the low-Earth orbit (LEO) satellite are configured, such as orbital altitude, number of beams, transmit antenna gain, angle of 3dB coverage, total number of scheduling time slots, total power, carrier frequency, and system bandwidth. The LEO satellite achieves beam hopping technology by scheduling different beam activations in each time slot. The multi-beam set is defined as... The set of discrete scheduling time slots is The user set is Beam b and users j The channel gain between can be expressed as (1) The parameters are defined as follows: Beam gain: For beam b Users within the coverage area j Beam gain By user j Location and service beam b Pitch angle at center point The calculation yields the following result, which is expressed as: (2) in G b and G j These represent the satellite's transmitting antenna gain and the user's receiving antenna gain, respectively. (Function) and These represent the first-order and third-order Bessel functions of the first kind, respectively. Parameters , For beam b Coverage of 3dB angle.

[0024] Free space loss: The free space loss of the satellite-to-ground channel is expressed as... (3) in c At the speed of light, f c For carrier frequency, For beam b Transmitting antenna to user j The spatial distance between receiving antennas.

[0025] Small-scale fading: Assuming the link between the satellite and the user follows an independent and identically distributed shadowed Ricean fading distribution, the fading gain... The probability density function can be modeled as (4) Where parameters , , ,function Represents the merging hypergeometry function. Parameters , and They are respectively represented as beams b To users j The average power of the line-of-sight path component, the average power of the non-line-of-sight path component, and the Nakagami distribution parameters are obtained. Using Equation (1), the average power of the line-of-sight path component in any time slot can be calculated. t In the middle, arbitrary beam b For any user j Channel gain between.

[0026] Step 2: Further, based on the completion of the basic satellite parameter configuration, this step aims to generate a multi-beam random scheduling strategy based on the current system status and preset objectives, in order to achieve resource allocation among multiple satellite beams. For any user... j It continuously monitors and reports the quality of the received signal to the control network, ensuring it remains connected to the beam providing the highest signal quality. Due to the long-distance signal transmission between satellite and ground, it is assumed that the satellite-to-ground channel is a slowly varying channel within a scheduling cycle. Definition B Each beam in the time slot t The total user demand covered at that time was And each beam has the same transmission power, that is , P totThis represents the total transmit power of the satellite. Define the beam activation matrix. Indicates the period of beam scheduling T The mid-wave beam activation state is indicated by an element of 1 indicating activation and an element of 0 indicating deactivation.

[0027] First, a weighting parameter reflecting beam priority is constructed. W As the basis for randomly selecting the active beam, it is represented as (5) in C b,t Indicates in time slot t Time Beam b Capacity, parameters and As a weighting factor, Indicates beam b exist t The average allocation ratio within the scheduling period prior to time is expressed as: (6) in x b,t For matrix x, the first... b Line number t The column elements. This parameter takes into account the fairness of random scheduling, avoiding some beams being frequently scheduled while others are never scheduled. Secondly, the weight parameters of each beam are normalized to obtain the beam. b In the time slot t Activation probability at time P b,t , which is represented as (7) Furthermore, a pseudo-random number generator is used to generate a floating-point number in the interval [0, 1). , indicating time slot t The activation threshold. Specifically, when the activation probability is greater than or equal to the activation threshold, i.e. beam b In the time slot t Activated, assigned a value Otherwise, it will not be activated and a value will be assigned. When scheduling a time slot is completed, the beam activation state is assigned to the first value of matrix x. t The parameters of equations (6) and (5) are updated, and the calculation is repeated. t The beam activation probability of the +1 time slot generates a new activation threshold, resulting in... t The beam activation state of the +1 time slot continues until the cycle is complete. T The beam activation matrix x is obtained by scheduling time slots.

[0028] Step 3: Based on the beam activation matrix x obtained above, the traditional performance evaluation method calculates the signal-to-interference-plus-noise ratio (SINR) of the beam on a time-slot-by-beam basis to obtain the total transmission rate, which can be expressed as: (8) Where WB is the signal transmission bandwidth, SINRb,t is the signal-to-interference-plus-noise ratio (SINR) of the signal received by beam b in time slot t, expressed as: (9) Where N0 represents noise power, and the first term in the denominator represents the signal interference generated by B-1 beams on the current beam. It can be seen that the calculation of this ergonomic form of equation (8) has high complexity, especially when the beam activation state is frequently adjusted in large-scale beam scheduling, the complexity of this evaluation method may lead to increased scheduling delay. Therefore, this step takes the inter-beam interference term as the starting point and considers a low-complexity performance evaluation method from a probabilistic perspective.

[0029] For large-scale beam scheduling systems, a new variable vector is introduced. , , defined as the frequency at which beam b is activated within a long-period, large-scale beam scheduling window, is called beam activation intensity. For a given beam activation matrix x, the beam activation intensity of beam b is expressed as: (10) Furthermore, within the beam scheduling period, the interference caused by beam b' to user j served by beam b can be approximated as follows: (11) in Indicates beam The transmit power. This inter-beam interference approximation model can be understood through two extreme examples, when... At that time, beam It remains in an active state, thus causing continuous interference to beam b; when At that time, no interference was generated. Therefore, the beam activation intensity It can be represented as a beam. The probability or degree of interference to beam b is thus represented as a scaling factor in equation (11). Specifically, the inter-beam interference approximation model is independent of time slots. Therefore, unlike traditional time-slot-by-time calculations, the inter-beam interference approximation model can quickly assess the interference experienced by beam b from beams within a scheduling period. The average interference is reduced, thereby improving the computational speed of performance evaluation.

[0030] Step 4: Further, based on the inter-beam interference approximation model constructed above, a rapid performance evaluation method is further developed.T Within a time slot of the beam scheduling period, for any beam b Service users j The interference term in equation (9) is replaced with an inter-beam interference approximation model, and the interference term is weighted and summed using beam activation intensity as the weight to obtain the user's result. j The signal-to-interference-plus-noise ratio is (12) The interference term reveals the average effect of beam interference. It is evident that... Size depends , But not dependent on In large-scale, long-period beam scheduling, Equation (12) can provide a good approximate signal-to-interference-plus-noise ratio compared to Equation (9).

[0031] Furthermore, it is possible to calculate users The achievable rate in any time slot is expressed as: (13) for T The total rate provided by the multi-beam satellite during the scheduling cycle of each time slot is expressed as: (14) It is evident that the computational complexity of using Equation (13) is lower than that of using Equation (8) for evaluating the overall rate performance of the beam activation matrix x obtained in step 2. This fast performance evaluation method, from a probabilistic perspective, rapidly calculates the interference term in the signal-to-interference-plus-noise ratio (SINR) using an inter-beam interference approximation model, regardless of the specific time slot. This low-complexity method demonstrates significantly faster evaluation speed in large-scale, long-period beam scheduling performance evaluation, improving the speed of the evaluation stage in beam scheduling optimization and thus reducing system latency.

[0032] Step 5: Calculate the corresponding beam activation intensity based on the request rates of different users. Based on the results obtained above... achievable rate The inter-beam approximate interference model can be used to design low-complexity beam scheduling methods. Assume the user... j exist T The request rate per time slot is Since equation (13) represents the user j The achievable rate in any time slot is the average achievable rate over the entire scheduling cycle. Therefore, the beam... b Need for users j Serve One time slot is enough to meet the user's request rate. At this time, the beam b The beam activation intensity can be expressed as Substituting equation (13) into the equation, the beam... b The beam activation intensity is further expressed as (15) in beam b Beam activation intensity is related to Related functions and It is irrelevant. It should be noted that the beam activation strength calculated by equation (15) differs from that calculated by equation (10) in step 3. The former is calculated based on the beam activation matrix x and is used for rapid performance evaluation, while the latter is based on the user request rate. Calculate the beam activation intensity vector It is used to design low-complexity beam scheduling strategies.

[0033] As can be seen from equation (15), considering the average interference, the inter-beam interference approximation model can map the user's request rate to the beam activation intensity. Therefore, given a user request rate... Number of time slots T ,bandwidth W B Channel gain g Noise power N 0, and transmission power P In this case, the beam activation intensity vector can be solved efficiently and quickly using a linearly convergent fixed-point iterative method. The specific process is as follows: Step 5.1: Initialize system parameters and initialize beam activation intensity vector. All initial values ​​are set to 1, and the iteration termination condition parameter is set. and U ; Step 5.2, for the first k In the next iteration, each element is calculated and updated according to equation (15), denoted as: (16) Step 5.3: Stop iteration when one of the following conditions is met, i.e., when... Or the maximum number of iterations equals U The iteration terminates at time 1. Output the beam activation intensity vector of the last iteration. *

[0034] For the beam activation intensity vector, there exists a unique solution set. * Satisfies equation (15). In practical applications, especially for large-scale beam scheduling scenarios, solving equation (15) is very fast. Furthermore, the magnitude of the beam activation intensity characterizes the activation frequency required for each beam. For example, for any beam... bIf the value of beam activation intensity satisfies This means that the user request rate can be [missing information]. T The beam activation intensity is satisfied within each time slot. Based on this idea, the obtained beam activation intensity can be further applied to design low-complexity beam scheduling methods.

[0035] Step 6: Based on the beam activation intensities obtained above, design a low-complexity beam scheduling strategy based on the capacity-demand ratio optimization problem. The main task of beam hopping is to determine which beams to activate and for how long, so as to achieve the best match between the capacity provided by the satellite beams and user demand. For a given... B A beam exists Possible beam combinations, using sets This represents all possible beam combinations. The beams are then optimized for allocation to these combinations. Number of time slots This allows for an optimal match between the capacity provided by the satellite beam and the user's requested rate. The optimization problem is expressed as follows: (17a) (17b) (17c) (17d) in T s This represents a unit time slot. Constraint (17b) states that the sum of the times occupied by each beam combination should equal the total hopping beam period. Constraint (17c) defines the beam... b The provided capacity. Constraint (17d) stipulates that the number of time slots allocated to each beam combination must be an integer. Therefore, this problem is a mixed-integer linear programming problem. Clearly, as the number of beams increases, the number of beam combinations... The exponential growth makes the problem difficult to solve.

[0036] To address the capacity-to-demand ratio optimization problem, starting from the idea of ​​an approximate model of inter-beam interference, a low-complexity beam-hopping method can be designed using beam activation intensity to achieve near-optimal performance. This method is based on the user request rate for each beam. Using the beam activation intensity vector obtained above The specific steps for designing a beam scheduling strategy are as follows: Step 6.1, beam activation intensity is defined as the activation frequency of the beam throughout the entire hopping beam cycle. Based on the beam activation intensity vector... Generate beam candidate vectors for each beam. Specifically, for the first b Beam activation intensity of each beam , producing T A one-dimensional vector of all zeros with elements. Output using a pseudo-random number generator [1, T Use unique random numbers between [] as position indices. Then use the vector The element at the corresponding position is assigned the value 1; Step 6.2, the beam candidate vector obtained in step 6.1 The transpose of is used as a beam candidate matrix The first column sequentially lists the beam candidate vectors. After randomly scrambling and transposing the elements, assign them to the 2nd to 3rd columns of the beam candidate matrix. Column, among which T ; Step 6.3, using a random number generator in the interval A positive integer is randomly generated above. Selecting a candidate beam matrix The Column as T Beam in each time slot b activation state vector ; Step 6.4, repeat steps 6.1 to 6.3 until a result is generated. B One beam in T Beam activation status in each time slot Output a low-complexity beam scheduling strategy ; This beam-hopping method can approximate a good match between capacity and demand while significantly reducing computational complexity. Especially in large-scale beam scheduling scenarios, it can also quickly solve for high-performance beam scheduling strategies.

[0037] Step 7: Based on the obtained beam activation intensity and beam scheduling strategy, design a low-complexity beam scheduling strategy based on the minimum time scheduling problem. For the minimum time scheduling problem, the objective is to ensure that the low-Earth orbit satellite completes the user's transmission task in the shortest possible time, i.e., minimizing the number of time slots required to meet the demand. The optimization model of the minimum time scheduling problem is expressed as follows: (18a)

[0038] (18b) (18c) Its goal is to minimize the overall transmission time of multi-beam satellites, a binary variable. Indicates waveb In the time slot t Internally activated, Indicates beam b In the time slot t Inactive. Similarly, binary variables. Indicates time slot t Used, and Indicates time slot t It was not used. Furthermore, constraint (18b) constrains the variable. x With variables y The relationship between them. Constraint (18c) ensures that the user's request rate is satisfied. It can be seen that the time minimization problem shown in equation (18) is a strongly NP-hard problem. Its computational complexity becomes slow and loses timeliness as the number of variables increases.

[0039] To address the minimum time scheduling problem, starting from the idea of ​​an approximate model of inter-beam interference, a low-complexity beam-hopping method can be designed using beam activation intensity to achieve near-optimal performance. First, for the beam activation intensity vector obtained in step 5... Sort them in descending order. Take the first element and define it as... ,if = 1 means that all given values ​​are required. T Each time slot is used for beam scheduling, and the output of the minimum time scheduling is: T .if If the number of time slots is less than 1, it means that the number of time slots required to meet the user request rate is less than 1. T At this point, the beam corresponding to the maximum beam activation intensity needs to... T The user request rate can be completed in one time slot, and the output result of the minimum time scheduling is: T Beam activation intensity vector Updated to (19) The updated scheduling cycle is T Finally, according to step 6.4, a beam scheduling strategy is generated for... T Beam scheduling for each time slot.

[0040] Based on the aforementioned efficient scheduling method for large-scale beam hopping of low-Earth orbit (LEO) satellites using an interference approximation model, this invention provides applications of this method in different scenarios to achieve rapid performance evaluation and low-complexity scheduling under large-scale, long-period LEO satellite beam scheduling. The specific implementation of this invention is described in detail below with reference to practical application scenarios, and the technical details are further illustrated with accompanying drawings: The following is a schematic diagram illustrating the application scenario of the efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on the interference approximation model of this invention. Figure 1 As shown. First, multi-beam low-Earth orbit satellite services. J For each user, the satellite generates multiple beams covering the same frequency band, and each receiving user is equipped with a monodirectional antenna to receive the satellite signal. Due to the propagation characteristics of wireless signals, the service... j Beam for individual users b Will affect others J -1 user causes interference, i.e., inter-beam interference. Therefore, user j The received signal consists of the desired signal, interference signals, and noise. Next, user status information, such as channel conditions, request rate, and user location, is fed back to the gateway via the return link. The resource manager uses this information to execute resource optimization algorithms, generate beam scheduling policies, and evaluate the performance of the scheduling policies. Further, the gateway transmits the decision to the satellite. Finally, the satellite generates beam coverage based on the received control information and transmits data to the target user.

[0041] In the specific embodiment of step 1, the parameters of the low-Earth orbit satellite are configured as follows: orbital altitude 800km, number of beams 120, transmit antenna gain 30dB, user receive antenna gain 0.5dB, 3dB coverage angle of 0.5°, total scheduling time slots 125, transmit power 40dBW / MHz, carrier frequency 15GHz, system bandwidth 20MHz, and noise power of [missing information]. W. In addition, among the small-scale fading parameters of the satellite-to-ground channel, the average power parameter of the line-of-sight path is 0.25, the average power parameter of the non-line-of-sight path is 0.28, and the Nakagami parameter is 5.

[0042] Step 1 completes the satellite parameter configuration and establishes the satellite-to-ground channel model. Step 2 aims to generate a multi-beam random scheduling strategy based on the system state and preset objectives to achieve resource allocation among multiple satellite beams. In a specific embodiment of Step 2, a weight parameter reflecting beam priority is first constructed. W As the basis for randomly selecting the activation beam, the weighting factor parameters =0.6, =0.4. Next, the weight parameters of each beam are normalized to obtain the beam... b In the time slot t Activation probability at time P b,t Furthermore, a random floating-point number is generated for the activation threshold of time slot t. = 0.73. Finally, update cyclically. W and beam activation probability P Compare with the activation threshold, and cycle. T =125 time slots, to obtain the beam activation matrix x. Based on the beam activation matrix x obtained in step 2, use step 3 to calculate B =Approximate inter-beam interference values ​​for 20 beams, and then substitute them into step 3 to calculate the beam achievable rate based on the inter-beam interference approximation model.

[0043] A comparison of the achievable rates calculated per slot and per beam with those calculated by the proposed fast performance evaluation method is shown in the appendix. Figure 2 The figure shows the results. A total of 1000 simulations were performed, and the average values ​​of three different beam activation intensities were extracted for comparison. These values ​​correspond to the three dashed lines from top to bottom in the figure, representing 0.8, 0.5, and 0.2 respectively. The three discrete points represent the achievable rate calculated per time slot and per beam. It can be seen that when the beam activation intensity is low and the number of beams is small, the approximate value calculated based on the inter-beam interference approximation model deviates significantly from the comparison method. This is because when the beam activation intensity and the number of beams are low, beam scheduling approximates a time-division multiple access scenario. However, the inter-beam interference approximation model is not suitable for communication scenarios with approximate time-division multiple access. Specifically, in applying the inter-beam interference approximation model, interference always exists even with low beam activation intensity values; while in applying the per-time slot and per-beam achievable rate calculation, a beam transmitting in one time slot introduces no interference. Therefore, if the beam activation intensity is low, the interference in one time slot is small or even zero. As the number of beams or the beam activation intensity increases, the difference between the two evaluation methods decreases rapidly. B When the beam activation intensity is around 0.8 or 60, the estimation errors of the two methods rapidly approach a few percentage points. This demonstrates that the low-complexity performance evaluation method based on the inter-beam interference approximation model exhibits good approximation performance in scenarios with high user demands and large-scale beam scheduling.

[0044] A comparison of the sum rate accuracy of multibeam low-Earth orbit satellites under three different user demand scenarios (low, medium, and high) is shown in the appendix. Figure 3 As shown, the three demand scenarios correspond to the Industrial Internet of Things (IIoT), traditional voice services, and ultra-high-definition video applications, respectively. T =125, number of beams is B= 120. Since user needs differ across the three scenarios, beam activation intensity is generally lower in low-demand scenarios, higher in high-demand scenarios, and falls between low and high demand scenarios. It is evident that as the number of beams and user demand increase, the gap between the sum rate calculated based on the inter-beam interference approximation model and the achievable rate calculated per time slot per beam rapidly narrows. Specifically, when the number of beams exceeds 60, the difference between the two methods remains within 5% for medium and high user demand scenarios. This further verifies the accuracy of the fast performance evaluation method based on the inter-beam interference approximation model in large-scale beam systems, while maintaining low computational complexity.

[0045] Using step 5, the beam activation intensity of each beam is calculated based on the request rates of different users. The beam activation intensity vector is then solved using equation (15) via a linearly convergent fixed-point iterative method. Set the iteration termination condition parameter. = 0.001, U = 300.

[0046] Further, set parameters =300, execute steps 6.1 to 6.4, and output a low-complexity beam scheduling strategy. .

[0047] The capacity demand ratio of implementing a low-complexity beam scheduling strategy will be compared with that of the optimal method solved by equation (17), as shown in the appendix. Figure 4 As shown. The dashed line with circles represents the performance curve of the traditional beam-hopping method, which specifically involves randomly selecting half of the beams for activation within a scheduling time slot. Discrete points represent the performance of the low-complexity method, while the solid line with triangles represents the performance of the optimal method. Simulation parameters, number of beams. B = 30, T = 125. It is evident that the optimal method achieves a perfect match between capacity and demand, with a constant performance of 1, as it is theoretically optimal. The proposed method has only a 6% performance gap compared to this optimal benchmark, while traditional beam-hopping methods show a significantly larger deviation. In particular, the computational complexity of the optimal method increases exponentially with the number of beams, making it unsuitable for meeting latency requirements or even computationally impossible in large-scale beam scheduling. In contrast, the proposed low-complexity method maintains real-time execution speed regardless of the number of beams, with an accuracy decrease of less than 6%. Therefore, the low-complexity beam scheduling method based on beam activation intensity exhibits superior performance while simultaneously considering communication efficiency and timeliness.

[0048] Using step 7, a low-complexity beam scheduling strategy based on the minimum time scheduling problem is designed. Specifically, the beam activation intensity vector obtained in step 5 is... Sort in descending order, then take the first element, which is the element with the largest value. If the value of the largest element is less than... T The beam activation intensity is updated. Using the updated beam activation intensity vector, a beam scheduling strategy is generated using steps 6.1 to 6.4.

[0049] Performance comparisons of low-complexity methods based on the minimum time scheduling problem under different user requirements are shown in the appendix. Figure 5 As shown. Among them The minimum number of time slots is obtained using the binary search method, with its maximum value as close to 1 as possible. The solid line with squares represents the number of time slots for completed transmission obtained from the updated beam activation intensity, i.e., the minimum number of time slots, which is usually a decimal. The dashed line with triangles is the optimal number of time slots calculated using Equation (18). The dashed line with circles represents the number of completed transmissions for the traditional beam hopping method. The simulation was run 1000 times and the average value was taken. It can be seen that the minimum number of time slots for data transmission increases with the increase of user demand. In particular, the time slots obtained from the low-complexity method exhibit performance almost the same as the optimal number of time slots. In addition, the performance of the low-complexity method is about 20% higher than that of the traditional beam hopping method, or even more. Therefore, the proposed low-complexity beam scheduling method based on beam activation intensity has good performance in solving the minimum time problem while maintaining low computational complexity.

[0050] Example 2 Based on the efficient scheduling method for large-scale beam hopping of low-Earth orbit (LEO) satellites based on an interference approximation model proposed in Example 1, this example proposes an efficient scheduling system for large-scale beam hopping of LEO satellites based on an interference approximation model to solve the problems of high complexity and slow performance evaluation in large-scale beam scheduling in LEO satellite communication. This system achieves low complexity and near-optimal performance in beam scheduling through modular design, specifically including a channel modeling module, an initial scheduling generation module, an interference approximation modeling module, a fast performance evaluation module, an activation intensity solution module, and a scheduling strategy generation module. The channel modeling module is used to configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites. The initial scheduling generation module is used to determine the activation state of the beams in each time slot and generate a beam activation matrix based on the multi-beam low-orbit satellite downlink communication channel model. The interference approximation modeling module is used to introduce beam activation intensity, and calculate the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period based on the beam activation matrix. The rapid performance evaluation module is used to calculate the achievable rate of the user served by the beam in any time slot based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam. The activation intensity calculation module is used to calculate the beam activation intensity vector based on the achievable rate of the user served by the beam in any time slot and according to the user request rate. The scheduling strategy generation module is used to generate a beam candidate matrix based on the beam activation intensity vector, generate a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem, sort the beam activation intensity vectors in the beam candidate matrix in descending order, determine the minimum number of scheduling time slots based on the maximum value of the beam activation intensity, and generate a skip beam scheduling strategy to solve the minimum time scheduling problem.

[0051] This system works in concert with the above modules, which can reduce the complexity of performance evaluation and scheduling through interference approximation models, while ensuring that the scheduling strategy meets user needs. It is especially suitable for low-Earth orbit satellite communication scenarios with large-scale beams.

[0052] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for efficient scheduling of large-scale beam hopping for low-Earth orbit satellites based on an interference approximation model, characterized in that, Includes the following steps: Configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites; Based on the downlink communication channel model of multi-beam low-Earth orbit satellites, the activation state of the beams in each time slot is determined, and the beam activation matrix is ​​generated. Introducing beam activation intensity, and based on the beam activation matrix, calculating the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period; Based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam, calculate the achievable rate of the user served by the beam in any time slot. Based on the reachable rate of the users served by the beam in any time slot, the beam activation intensity vector is solved according to the user request rate. Based on the beam activation intensity vector, a beam candidate matrix is ​​generated, resulting in a low-complexity hopping beam scheduling strategy to solve the capacity demand ratio problem. The beam activation intensity vectors in the beam candidate matrix are sorted in descending order, and the minimum number of scheduling slots is determined based on the maximum value of the beam activation intensity, thus generating a hopping beam scheduling strategy to solve the minimum time scheduling problem.

2. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The determination of the activation state of the beam in each time slot specifically involves defining: B Each beam in the time slot t The total user demand covered at that time was And each beam has the same transmission power, that is , Define the total satellite transmit power; define the beam activation matrix. Indicates the period of beam scheduling T The mid-wave beam activation state is indicated by an element of 1 indicating activation and an element of 0 indicating deactivation.

3. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The generation of the beam activation matrix specifically involves: first, constructing a weight parameter that reflects the beam priority. W As the basis for randomly selecting the active beam, the weight parameters of each beam are then normalized to obtain the beam. b In the time slot t Activation probability at time P b,t Furthermore, a pseudo-random number generator is used to generate a floating-point number in the interval [0, 1). , indicating time slot t The activation threshold is set when the activation probability is greater than or equal to the activation threshold. Beam b In the time slot t Activated, assigned a value Otherwise, it will not be activated and a value will be assigned. When scheduling a time slot is completed, the beam activation state is assigned to the first value of matrix x. t Column, update beam b exist t The average allocation ratio and weight parameters within the scheduling cycle prior to the current time are calculated iteratively. t The beam activation probability of the +1 time slot generates a new activation threshold, resulting in... t The beam activation state of the +1 time slot continues until the cycle is complete. T The beam activation matrix x is obtained by scheduling time slots.

4. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The introduction of beam activation intensity specifically involves: for large-scale beam scheduling systems, introducing a new variable vector. , Defined as a beam within a long-period large-scale beam scheduling window. b The frequency at which the beam is activated is called the beam activation intensity.

5. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The step of calculating the beam activation intensity of the beam activation matrix based on the beam activation matrix specifically involves: for a given beam activation matrix... Beam b The beam activation intensity is expressed as 。 6. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The step of calculating the average interference experienced by the beam during the beam scheduling period based on the beam activation matrix specifically involves: within the beam scheduling period, the beam... b 'Beam b Service users j The resulting interference can be approximated as: ,in, Indicates beam The transmission power.

7. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The calculation of the achievable rate for the users served by the beam in any time slot, based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam, specifically involves: T Within a time slot of the beam scheduling period, for any beam b Service users j The beam is replaced by the beam activation intensity of the beam activation matrix and the average interference experienced by the beam. b In the time slot t The interference term in the signal-to-interference-plus-noise ratio (SIR) formula for the received signal is weighted and summed using beam activation intensity as the weight to obtain the user's signal. j The signal-to-interference-plus-noise ratio is According to the user j Signal-to-interference-to-noise ratio Calculate users The achievable rate in any time slot is expressed as: .

8. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The beam activation intensity vector is calculated based on the achievable rate of the user served by the beam in any time slot, according to the user's requested rate. Specifically, it is assumed that the user... j exist T The request rate per time slot is Based on the achievable rate of the users served by the beam in any time slot, since it is represented as the user j The achievable rate in any time slot, which is the average achievable rate over the entire scheduling period, therefore, the beam b Need for users j Serve One time slot is enough to meet the user's request rate; at this time, the beam b The beam activation intensity can be expressed as Substituting the achievable rates of the users served by the beam in any time slot, the beam... b The beam activation intensity is further expressed as ,in, Beam b Beam activation intensity is related to Related functions and Irrelevant; based on the beam activation strength of beam b, at a given user request rate Number of time slots T ,bandwidth W B Channel gain g Noise power N 0, and transmission power P In this case, the beam activation intensity vector can be solved efficiently and quickly using a linearly convergent fixed-point iterative method. .

9. The efficient scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model according to claim 1, characterized in that, The process involves generating a beam candidate matrix based on the beam activation intensity vector, sorting the beam activation intensity vectors in the candidate matrix in descending order, and determining the minimum number of scheduling time slots based on the maximum value of the beam activation intensity. Specifically, this involves determining which beams to activate and for how long based on the beam activation intensity vector, to achieve an optimal match between the capacity provided by the satellite beams and user needs. B A beam exists Possible beam combinations, using sets Represents all possible beam combinations; optimizes the allocation to beam combinations. Number of time slots This allows for an optimal match between the capacity provided by the satellite beam and the user's requested rate.

10. A high-efficiency scheduling system for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model, comprising the high-efficiency scheduling method for large-scale beam hopping of low-Earth orbit satellites based on an interference approximation model as described in any one of claims 1 to 9, characterized in that, It includes a channel modeling module, an initial schedule generation module, an interference approximation modeling module, a fast performance evaluation module, an activation strength solving module, and a scheduling policy generation module; The channel modeling module is used to configure satellite parameters and establish a downlink communication channel model for multi-beam low-Earth orbit satellites. The initial scheduling generation module is used to determine the activation state of the beams in each time slot and generate a beam activation matrix based on the multi-beam low-orbit satellite downlink communication channel model. The interference approximation modeling module is used to introduce beam activation intensity, and calculate the beam activation intensity of the beam activation matrix and the average interference experienced by the beam during the beam scheduling period based on the beam activation matrix. The rapid performance evaluation module is used to calculate the achievable rate of the user served by the beam in any time slot based on the beam activation strength of the beam activation matrix and the average interference experienced by the beam. The activation intensity calculation module is used to calculate the beam activation intensity vector based on the achievable rate of the user served by the beam in any time slot and according to the user request rate. The scheduling strategy generation module is used to generate a beam candidate matrix based on the beam activation intensity vector, generate a low-complexity skip beam scheduling strategy to solve the capacity demand ratio problem, sort the beam activation intensity vectors in the beam candidate matrix in descending order, determine the minimum number of scheduling time slots based on the maximum value of the beam activation intensity, and generate a skip beam scheduling strategy to solve the minimum time scheduling problem.

Images