A multi-dimensional resource allocation method and system for low earth orbit satellite agile beams

Abstract

The application provides a multi-dimensional resource allocation method and system for low-orbit satellite agile beams, and the steps of the method comprise: determining the coverage range of a low-orbit satellite every first time interval, and dividing the coverage range of the low-orbit satellite into a first preset number of sub-areas; constructing an allocation scheme corresponding to the beams and the sub-areas based on the number of beams of the low-orbit satellite and the number of sub-areas, calculating the weight parameters of each allocation scheme, and determining the finally executed allocation scheme based on the weight parameters of the allocation scheme; uniformly dividing the first time interval into a plurality of second time intervals, calculating the power allocated to each user in the sub-area corresponding to the beam every second time interval; and based on the power allocated to each user, using a particle swarm algorithm to calculate the bandwidth allocated to each user. The scheme solves the problem of fine allocation of multi-dimensional resources, changes the optimization target from beam-level resource aggregation service to user-level resource demand guarantee, and realizes the synchronous improvement of the allocation granularity and the service quality.

Description

Technical Field

[0001] This invention relates to the field of satellite internet technology, and in particular to a multi-dimensional resource allocation method and system for low-Earth orbit satellite agile beams. Background Technology

[0002] Low Earth Orbit (LEO) satellite communication, a hot technology in the field of communications today, is reshaping the global communication landscape with unprecedented momentum. LEO satellites typically refer to satellites operating in orbits 300 to 2000 kilometers above the Earth's surface. Compared to traditional high-orbit satellites, LEO satellites are closer to the ground, resulting in shorter signal transmission paths and significantly reduced signal propagation delays. This allows users to experience near-real-time communication, offering significant advantages in real-time interactive applications such as video calls and online games.

[0003] Low Earth orbit (LEO) satellites have an extremely wide coverage area. Although the coverage area of ​​a single LEO satellite is limited, by building a huge satellite constellation, seamless global coverage can be achieved. Whether in remote mountainous areas, vast oceans, or polar regions, stable communication services can be enjoyed, effectively making up for the geographical deficiencies of terrestrial communication networks.

[0004] However, current low-Earth orbit satellite technology often fails to fully consider the link quality in satellite hopping beams, the heterogeneous needs of users, and the complexity of dynamic interference changes. There is insufficient optimization of cross-beam and inter-user resource sharing. Especially in high-density user scenarios, fixed optimization strategies may lead to a decrease in resource utilization. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a multi-dimensional resource allocation method and system for low-Earth orbit satellite agile beams, in order to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a multi-dimensional resource allocation method for low-Earth orbit satellite agile beams, the method comprising the following steps: The coverage area of ​​the low-Earth orbit satellite is determined every preset first time slot, and the coverage area of ​​the low-Earth orbit satellite is divided into a first preset number of sub-regions; Based on the number of beams and sub-regions of the low-Earth orbit satellite, an allocation scheme corresponding to the beams and sub-regions is constructed, the weight parameters of each allocation scheme are calculated, and the final allocation scheme to be executed is determined based on the weight parameters of the allocation scheme. The first time slot is evenly divided into multiple second time slots, and the power allocated to each user in the sub-region corresponding to the beam is calculated every preset second time slot; Every second time slot, based on the power allocated to each user, the bandwidth allocated to each user is calculated using the particle swarm optimization algorithm.

[0007] Compared with traditional approaches, this scheme overcomes the limitations of current research on resource allocation. Firstly, in research on resource scheduling in beam-hopping systems, most scholars focus only on optimizing beam-hopping schemes, neglecting the underlying resource allocation. Secondly, some joint optimization schemes considering beam-hopping systems often treat bandwidth as an evenly distributed or fixed parameter, ignoring its significant potential for improving system capacity as a key degree of freedom. This scheme not only solves the problem of fine-grained multi-dimensional resource allocation but also shifts the optimization objective from beam-level resource aggregation and service to user-level resource demand assurance, achieving simultaneous improvement in allocation granularity and service quality.

[0008] In some embodiments of the present invention, in the step of calculating the power allocated to each user in the sub-region corresponding to the beam, the power allocated to each user is calculated using a weighted geometric water-filling algorithm based on the initially set user weights, the assumed bandwidth allocation results, and the power allocation upper limit.

[0009] In some embodiments of the present invention, in the step of constructing a beam-to-sub-region allocation scheme based on the number of beams and sub-regions of low-orbit satellites, each beam corresponds to one sub-region, and a preset number of allocation schemes are constructed.

[0010] In some embodiments of the present invention, the step of calculating the weight parameters of each allocation scheme and determining the final allocation scheme to be executed based on the weight parameters of the allocation scheme includes: Calculate the penalty weight function value for each sub-region; The weight parameters for each allocation scheme are calculated based on the penalty weight function value for each sub-region; The allocation scheme with the largest weight parameter is selected as the final allocation scheme to be executed.

[0011] In some embodiments of the present invention, the penalty weight function value is calculated using the following formula in the step of calculating the penalty weight function value for each sub-region: ; in, This represents the penalty weight function value for subregion n. Represents users in subregion n The head delay of the queue to be scheduled. This represents the total number of users in subregion n. This represents the calculation parameter corresponding to the maximum allowable delay for the data packet from the terminal to the core network gateway.

[0012] In some embodiments of the present invention, in the step of calculating the weight parameters of each allocation scheme based on the penalty weight function value of each sub-region, the weight parameters of the allocation scheme are calculated using the following formula: in, This represents the weight parameters of allocation scheme j. This represents the historical average flow rate of subregion n. This indicates whether sub-region n in allocation scheme j is assigned a beam; N represents the total number of sub-regions. This represents the calculation parameters corresponding to the lowest bit rate. This represents the calculation parameter corresponding to the maximum packet loss rate limit under non-congestion conditions.

[0013] In some embodiments of the present invention, in the step of calculating the power allocated to each user in the sub-region corresponding to the beam every preset second time slot, the transmit power allocated to each user is determined by a weighted geometric water-filling algorithm based on a preset transmit power allocation upper limit, the initial bandwidth allocation of each user, the channel gain and weight.

[0014] In some embodiments of the present invention, in the step of determining the transmission power allocated to each user based on a weighted geometric water-filling algorithm using a preset transmission power allocation upper limit, initial bandwidth allocation for each user, channel gain and weight, the initial bandwidth allocation for each user is the allocation result obtained by averaging the total bandwidth.

[0015] In some embodiments of the present invention, in the step of calculating the bandwidth allocated to each user based on the power allocated to each user at every preset second time slot; Obtain the carrier-to-noise-plus-interference ratio under the current power allocation, and map the continuous bandwidth to a discrete set; The throughput is maximized by using discrete sets and the carrier-to-noise-plus-interference ratio, with the total bandwidth used as a constraint to initialize the particle swarm velocity and position. The particle swarm velocity and position are initialized based on fitness updates to obtain the final bandwidth allocation scheme.

[0016] A second aspect of the present invention also provides a multi-dimensional resource allocation system for low-Earth orbit satellite agile beams. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0017] A third aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned multidimensional resource allocation method for low-Earth orbit satellite agile beams.

[0018] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the text, or may be learned by practice of the invention. The objects and other advantages of the invention will become apparent from the description and the accompanying drawings.

[0019] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0020] The accompanying drawings, which are provided to further illustrate the invention and form part of this application, are not intended to limit the scope of the invention.

[0021] Figure 1 This is a schematic diagram illustrating one implementation of the multi-dimensional resource allocation method for low-Earth orbit satellite agile beams in this scheme. Figure 2 This is a schematic diagram of the scheduling framework for the hopping beam system in this scheme; Figure 3 This is a schematic diagram of the multi-timescale scheduling framework of this scheme. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0023] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0024] Introduction to existing technologies: Existing technology 1: Low Earth Orbit satellite agile beam scheduling method The agile beam scheduling method achieves efficient beam scheduling in highly dynamic scenarios by dynamically adjusting beam hopping modes and time slot allocation, combined with user status feedback such as real-time demands, channel gain, and location information. Simultaneously, considering interference from multiple scenarios, it employs various schemes such as time-division multiplexing, clustering, and spatial isolation to avoid co-channel interference. Furthermore, by solving an optimization problem, it obtains the optimal working beam combination, significantly improving the flexibility of resource allocation and adapting to non-uniform user distribution and heterogeneous user traffic demands.

[0025] Specifically, it includes: 1. A dynamic beam-hopping scheduling technology for low-Earth orbit constellations. This technology utilizes real-time feedback of status information (including rate requirements, beam channel gain, and location) from user terminals. The satellite end dynamically generates candidate beam-hopping schemes that meet spatial isolation conditions based on the interference matrix and avoids inter-beam interference through a time-division multiplexing mechanism. The system divides the scheduling period into equal-length time slots. In each time slot, it intelligently selects the optimal working beam by combining channel gain and user rate requirements. It uses an iterative algorithm to dynamically adjust the beam mode and user access strategy, and adaptively controls the maximum number of users accessing a single beam based on the service satisfaction rate.

[0026] 2. A multi-beam coordinated scheduling method and system for low-Earth orbit satellite communication. This method collects the location, resource requirements, and service buffering information of online users through a wide-beam coverage strategy, and maps the user's geographical location to a coordinate plane. Under channel resource constraints, an adaptive clustering algorithm is used to generate hopping beam positions, and the matching between resource requirements and total quantity limits is verified. Based on a deep reinforcement learning algorithm, a neural network model containing state vectors such as buffered service volume and transmission rate is used to periodically optimize the scheduling of hopping beams. Within each scheduling period, by optimizing the hopping beam time frame structure, point beams are used to cover users within the beam positions to achieve efficient service transmission.

[0027] 3. A method for designing hopping beam patterns for low-Earth orbit (LEO) satellites. This method first uses a hybrid clustering algorithm to dynamically cluster ground sub-regions based on sub-region service requirements, ensuring load balancing while making the geographical locations of sub-regions within a cluster compact. Second, it innovatively divides user satisfaction into three levels, guiding the selection, crossover, and mutation operations of a genetic algorithm accordingly. The selection operation uses overall user satisfaction within the hopping beam period to select the optimal individual; the crossover operation performs gene recombination based on cluster-level user satisfaction; and the mutation operation adjusts the individual genes based on user satisfaction in a single sub-region. Finally, it performs power fine-tuning on beams serving the same frequency.

[0028] 4. A low-Earth orbit satellite beam scheduling and user grouping technology, which achieves efficient user access and resource allocation through a three-level beam cooperative working mechanism. First, the target access location of the user is determined by scanning the first-level beam. Then, the user location information is accurately obtained through the intelligent random access mechanism of the second-level beam. Finally, an adaptive clustering algorithm based on distance constraints is used to dynamically group users, and the optimized cluster centroid is used as the beam coverage center to complete beam skipping scheduling.

[0029] Disadvantages of existing technology 1 The hopping beam pattern scheduling method based on non-uniform user demand fails to fully consider the complexity of link quality, heterogeneous user needs, and dynamic interference changes in satellite hopping beams, resulting in poor performance of the optimization algorithm in actual deployment. Secondly, the optimization of cross-beam and inter-user resource sharing is insufficient, especially in high-density user scenarios, where fixed optimization strategies may lead to decreased resource utilization. Finally, and most critically, is the complexity of solving the optimization problem. The scheduling optimization scheme involves solving a mixed-integer nonlinear programming (MINLP) problem, which includes the joint solution of binary and continuous variables. The constraints are non-convex and the parameters are strongly coupled, causing the problem-solving complexity to increase exponentially, failing to meet the millisecond-level response requirements of onboard real-time scheduling.

[0030] Existing technology 2: A hierarchical architecture-based dynamic beam resource allocation scheme Based on a dynamic resource allocation scheme using agile beams, a multi-level scheduling architecture is constructed to achieve hierarchical resource management for low-Earth orbit satellite communication systems. First, the scheduling decision center dynamically plans resource allocation strategies based on scheduling information such as communication link quality, user data cache, and QoS priority of service traffic. Second, the satellite beam scheduling module dynamically allocates time-frequency resources through beam pattern calculation based on the characteristics of beam hopping technology. Finally, resources are allocated to users in the currently served sub-area according to actual needs.

[0031] Specifically, it includes: 1. A low-Earth orbit (LEO) satellite beam-hopping optimization method based on a space-ground integrated architecture, which achieves efficient resource management through a hierarchical optimization framework. First, a stochastic optimization model is constructed that considers onboard resource constraints, power consumption constraints, and ground interference requirements, and is decomposed into beam-level and user-level resource allocation sub-problems. In beam-level optimization, the problem is innovatively modeled as a Markov game, employing a multi-agent reinforcement learning architecture, enabling each agent to make autonomous decisions based on its local state. In user-level optimization, the problem is transformed into a Lagrange dual problem using convex optimization theory. Through the synergistic application of game theory and convex optimization, the fairness of user service processing, system throughput, and onboard resource utilization are guaranteed while ensuring ground interference constraints.

[0032] 2. A wireless resource scheduling method and system based on low-Earth orbit (LEO) constellation satellite communication. This method employs a three-tiered architecture to achieve optimized resource management. The first-tier wireless resource scheduling center integrates global spectrum status, telecommunications policies, and historical data to dynamically plan satellite service areas and generate resource allocation strategies. The second-tier satellite resource scheduling module, based on the scheduling center configuration and combined with user needs and terminal capabilities, achieves precise allocation of beam and time-frequency resources through beam hopping pattern calculation and user clustering technology. The third-tier user service scheduling module, based on service type and QoS requirements, achieves efficient access to multi-service data through priority management, dynamic adjustment of link parameters, and encrypted data storage.

[0033] 3. A two-layer time-frequency resource allocation method based on a genetic algorithm. First, during each beam-hopping time slot, ground users are allocated to either the satellite network or the terrestrial network based on their location, generating a user allocation method and pairing it with transport blocks to form a pairing matrix. Constraints are introduced to establish a user scheduling and resource allocation model. Second, preliminary resource block allocation is performed based on the worst-case interference scenario, calculating the number of resource blocks required by each ground user and allocating the minimum rate resource blocks to obtain the resource allocation matrix. Finally, the resource block rate is calculated based on the actual signal-to-noise ratio, redundant resource blocks are released, and these resource blocks are reallocated using a genetic algorithm. Iterative optimization continues until the termination condition is met.

[0034] 4. A QoE-driven low-Earth orbit satellite phased array beamforming and beam hopping method intelligently generates optimized beam patterns to cover target user terminals by dynamically evaluating the priority of user transmission services. This method first performs adaptive beamforming based on the spatial distribution characteristics of user terminal clusters, achieving precise coverage by combining resource demand and priority; secondly, it constructs a reinforcement learning-based dynamic resource allocation model, iteratively optimizing the real-time resource allocation strategy through historical experience data, and continuously updating model parameters using a probabilistic sampling mechanism.

[0035] Disadvantages of existing technology 2 Existing technology two achieves dynamic resource allocation for beam hopping systems through a hierarchical scheduling architecture. However, it suffers from insufficient scheduling coordination across multiple time scales and lacks a systematic optimization framework for long-term time scales, making it difficult to achieve global optimization. Furthermore, scheduling algorithms face a trade-off between real-time and complexity-sensitive beam planning and short-term resource allocation; many optimization algorithms have high computational complexity, making it difficult to meet the rapid response requirements in beam hopping scenarios. While related research has yielded some results, experimental verification has primarily remained at the simulation level, lacking performance evaluation of real-world systems, thus affecting the practicality of the algorithms.

[0036] In summary, this invention addresses the shortcomings in multi-dimensional resource allocation and the insufficient collaborative optimization of long-term beam planning and short-term resource allocation in low-Earth orbit (LEO) satellite hopping beam systems. Through a cross-timescale joint optimization mechanism, it enhances the dynamic adaptability of resource allocation while ensuring beam coverage stability, thereby significantly improving spectral efficiency and service quality. When designing a joint scheduling scheme for LEO satellite orbits, the unevenly distributed user service demands should be considered, along with inter-beam interference in a fully reused beam frequency system. To meet the QoS requirements of different users, a ground resource grid allocation scheme is introduced. Ultimately, this achieves the goal of independently allocating resource grids under different beams while considering inter-beam interference and full frequency reuse, while meeting user service traffic and QoS performance requirements. It unifies and jointly optimizes BHP scheduling, power allocation, and bandwidth allocation.

[0037] (1) Optimization method for time-varying beam hopping scheduling of low-orbit satellites: As a typical resource-constrained system, satellite communication uses beam hopping technology to achieve frequency reuse between beams in order to improve resource utilization. In order to further optimize system performance, it is necessary to break through the limitations of traditional time and frequency resource allocation and establish a multi-dimensional joint optimization mechanism that integrates time-domain scheduling, spatial beam pointing, frequency-domain resource allocation and power control. By finely coordinating the parameter configuration of each dimension, the maximum utilization of satellite's limited resources can be achieved.

[0038] (2) A hierarchical architecture-based dynamic resource allocation scheme for agile beams: In low-Earth orbit satellite beam-hopping systems, different scheduling decisions are made periodically at different time scales. Therefore, a multi-time-scale scheduling architecture is constructed to process system characteristics of different time dimensions in a hierarchical manner. Specifically, long-term processes such as satellite motion and beam hopping are defined at the beam-level time scale, while processes such as differentiated service requirements of users, spatiotemporal non-uniform distribution, and sudden service requests require more refined resource allocation. That is, user-level resource allocation should be carried out at a smaller time scale (such as time slots).

[0039] like Figure 1 As shown, this invention proposes a multi-dimensional resource allocation method for low-Earth orbit satellite agile beams. The steps of this method include: like Figure 2 As shown, in step S100, the coverage area of ​​the low-orbit satellite is determined every preset first time slot, and the coverage area of ​​the low-orbit satellite is divided into a first preset number of sub-regions. In the specific implementation process, a beam skipping scheduling scenario for low-Earth orbit satellites is defined, and a user traffic model is built to simulate the ground-based hierarchical scheduling function in the satellite system. The user queue to be scheduled is maintained at the Medium Access Control (MAC) layer.

[0040] Specifically, a downlink communication scenario for a low-Earth orbit satellite hopping beam system is constructed. The system is defined to operate in a staring beam mode, ensuring that the satellite can continuously and stably cover T predetermined sub-regions within its service area during operation. In a resource allocation system, an important constraint is generally considered: the maximum number of beams that the satellite can activate simultaneously. Typically much smaller than the total number of service sub-regions T (i.e.) Meanwhile, each service sub-region At least one user terminal must be deployed, and the total user scale of the system is defined as the cumulative number of users in all sub-regions. Finally, to ensure service quality, the system needs to manage two levels of user sets simultaneously, including the local user sets in each sub-region and the system-level global user set, and fully consider the differences in service quality (QoS) requirements among different user terminals.

[0041] Step S200: Construct an allocation scheme for beams and sub-regions based on the number of beams and sub-regions of the low-orbit satellite, calculate the weight parameters of each allocation scheme, and determine the final allocation scheme to be executed based on the weight parameters of the allocation scheme. Step S300: Divide the first time slot evenly into multiple second time slots, and calculate the power allocated to each user in the sub-region corresponding to the beam every preset second time slot; In step S400, every preset second time slot, the bandwidth allocated to each user is calculated using the particle swarm optimization algorithm based on the power allocated to each user.

[0042] In the specific implementation process, this scheme constructs a system model of the downlink of the beam-hopping low-orbit satellite communication system. Under the condition of full frequency reuse, it considers the co-channel interference between beams and establishes a joint optimization problem of beam-hopping, power and bandwidth. To simplify the complex optimization problem, a distance threshold is set to avoid co-channel interference between beams, reducing the solution space size of beam-hopping pattern (BHP) scheduling. Based on the spatial isolation between beams, a ground scheduling resource grid configuration is introduced, and resources are allocated independently for each beam. Specifically, a three-layer scheduling architecture for the LEO hopping beam system is designed, and the beam scheduling scheme for the current beam period is determined by a weighted search scheme at the beam level.

[0043] After determining the scheduling beams, power and bandwidth allocation are jointly optimized using a step-by-step optimization method. First, the optimal solution for power allocation is obtained by fixing the bandwidth. Second, the current power allocation result is used as input for bandwidth allocation to obtain the optimal solution for bandwidth allocation. After one iteration of power and bandwidth allocation, the current bandwidth allocation is used as input for the power scheduler for iterative optimization until convergence is achieved. Each optimization problem is performed independently in different beams.

[0044] Considering the forward link model in a satellite communication system, the normalized antenna gain radiation pattern of the satellite in the communication scenario is expressed as: in, Indicates beam The antenna points towards the user. Angle of deviation from direction This indicates the beam. Service users When considering the antenna gain, for a circular aperture antenna, use Indicates the antenna radius. The speed of light can be represented using the first-order Bessel function of the first kind. Describe the antenna radiation pattern characteristics. Simultaneously, the maximum transmit gain of the transmitting antenna within the beam is used... This means that the transmit gain directed at the user can be obtained from this: Satellite and users The channel gain between them is expressed as: Indicates beam Pointing to user The channel experiences various losses, including path loss, cloud and rain attenuation, and so on.

[0045] In the operation of a hopping beam system, beam scheduling is implemented using Beam Hopping Time Slots (BHTS) as the basic time unit. Within each BHTS period, a single beam can only serve one specific sub-area, and each sub-area can only receive signals from one beam. This design effectively avoids redundant use of beam resources. The system uses a limited number of beam combinations to selectively serve sub-areas within the satellite's coverage area; this dynamic configuration mode is defined as a hopping beam pattern (BHP). During the actual service process, the satellite, based on a preset scheduling algorithm, makes real-time decisions on which BHP scheme to use to provide communication services to users during a specific BHTS period.

[0046] Compared with traditional approaches, this scheme overcomes the limitations of current research on resource allocation. Firstly, in research on resource scheduling in beam-hopping systems, most scholars focus only on optimizing beam-hopping schemes, neglecting the underlying resource allocation. Secondly, some joint optimization schemes considering beam-hopping systems often treat bandwidth as an evenly distributed or fixed parameter, ignoring its significant potential for improving system capacity as a key degree of freedom. This scheme not only solves the problem of fine-grained multi-dimensional resource allocation but also shifts the optimization objective from beam-level resource aggregation and service to user-level resource demand assurance, achieving simultaneous improvement in allocation granularity and service quality.

[0047] In some embodiments of the present invention, in the step of calculating the power allocated to each user in the sub-region corresponding to the beam, the power allocated to each user is calculated using a weighted geometric water-filling algorithm based on the initially set user weights, the assumed bandwidth allocation results, and the power allocation upper limit.

[0048] In some embodiments of the present invention, in the step of constructing a beam-to-sub-region allocation scheme based on the number of beams and sub-regions of low-orbit satellites, each beam corresponds to one sub-region, and a preset number of allocation schemes are constructed.

[0049] In some embodiments of the present invention, the step of calculating the weight parameters of each allocation scheme and determining the final allocation scheme to be executed based on the weight parameters of the allocation scheme includes: Calculate the penalty weight function value for each sub-region; The weight parameters for each allocation scheme are calculated based on the penalty weight function value for each sub-region; The allocation scheme with the largest weight parameter is selected as the final allocation scheme to be executed.

[0050] In some embodiments of the present invention, the penalty weight function value is calculated using the following formula in the step of calculating the penalty weight function value for each sub-region: ; in, This represents the penalty weight function value for subregion n. Represents users in subregion n The head delay of the queue to be scheduled. This represents the total number of users in subregion n. This represents the calculation parameter corresponding to the maximum allowable delay for the data packet from the terminal to the core network gateway.

[0051] In some embodiments of the present invention, in the step of calculating the weight parameters of each allocation scheme based on the penalty weight function value of each sub-region, the weight parameters of the allocation scheme are calculated using the following formula: in, This represents the weight parameters of allocation scheme j. This represents the historical average flow rate of subregion n. This indicates whether sub-region n in allocation scheme j is assigned a beam; N represents the total number of sub-regions. This represents the calculation parameters corresponding to the lowest bit rate. This represents the calculation parameter corresponding to the maximum packet loss rate limit under non-congestion conditions.

[0052] Specifically, the feasible set of BHPs is as follows: ,in The number of feasible solutions, specifically in BHTS Select BHP It is represented as: in It is used to indicate the sub-region in BHP A binary variable indicating whether the light is on or off. Representative at BHTS Middle Individual sub-regions are served, on the contrary This means in BHTS A beam is used to provide services to users within this sub-region.

[0053] In each BHTS Within the satellite, power and bandwidth are allocated to users awaiting scheduling, with the allocated power set being: in In order to be in The set of users waiting to be scheduled, i.e. The set of users within the covered sub-region, thus obtaining the users. The transmit power expression is obtained. Based on the channel gain formula, the user's transmit power expression is obtained. Signal receiving power .

[0054] Simultaneously considering a full-frequency reuse hopping beam communication system, where each beam occupies the entire bandwidth of the satellite, the bandwidth allocated in sub-region n is expressed as: User terminal With service beam The overall signal quality can be determined by the carrier-to-interference-plus-noise ratio. The representation, its expression is: in and These represent the carrier-to-interference ratio and the carrier-to-noise ratio, respectively. The carrier-to-noise ratio can be obtained from the allocated power and bandwidth. The calculation formula is: in Indicates the quality factor at the receiving end; This is the Boltzmann constant. When the system employs frequency reuse technology to improve spectral efficiency, co-channel interference (CCI) between adjacent beams becomes a critical limiting factor. When users Beam During service, assume the interfering beam is The total interference from the other K-1 beams is: The corresponding load-to-dry ratio can be expressed as: Based on the above analysis, the allocated capacity of this satellite system can be obtained as follows: The goal of optimization is to select a suitable scheduling scheme and allocate beam, power, and bandwidth resources to maximize the objective function. The specific optimization problem... The model is as follows: Design optimization algorithms to maximize system throughput while meeting the constraints of Quality of Service (QoS) requirements. middle, This indicates that the allocated power cannot exceed the satellite's maximum power limit; The beam allocation scheme within the hopping beam period is represented by a binary variable. This indicates that the allocated bandwidth cannot exceed the satellite's maximum bandwidth limit; limiting conditions. This indicates that users must meet specified QoS requirements, and users are divided into two categories: Guaranteed Bit Rate (GBR) and non-GBR. Therefore, This problem involves binary variables and set operations, making it NP-hard. Furthermore, the objective function and constraints contain multiple variables with strong coupling between parameters, resulting in high complexity for direct solutions and failing to meet the requirements for rapid decision-making in scheduling schemes.

[0055] To avoid inter-beam interference for users, and to fully utilize the beam-hopping advantage of multi-beam satellite communication systems, spatial isolation is established to avoid co-channel interference between beams. Assume the cell radius covered by the satellite is... When the center distance between two cells is 3R or more, the CCI between the two beams covering these two cells is negligible. Therefore, the number of candidate Beam Hopping Time Slots (BHPs) is reduced by setting a distance threshold. Thus, a scheduling scheme without inter-beam interference is obtained, and the final effective set of BHPs is: Secondly, to fully integrate 5G scheduling strategies, a terrestrial network resource grid scheduling scheme is introduced, and a scheduling architecture suitable for LEO beam-hopping systems is designed. This scheduling architecture includes three stages: BHP scheduling, power allocation, and RB scheduling. In terrestrial mobile networks, radio resources only have two dimensions: time domain and frequency domain. User data in the downlink cannot be transmitted simultaneously on the same frequency, as this would result in data not being correctly decoded by the user. However, in LEO beam-hopping systems, since the beams illuminated by satellites are not adjacent to each other, and the frequency reuse method between beams is full-band reuse, the same resource grid can be replicated to different beams for deployment. Therefore, power and bandwidth allocation are divided into time slots.

[0056] During the BHP cycle Within, the power allocation set for the i-th time slot: Therefore, the power allocation constraints are limited to the time slot scale, ensuring that the power constraint requirements are met in each time slot throughout the entire BHTS cycle. Let Let be the total number of time slots in the entire BHTS cycle. Therefore, the power allocation within the entire cycle is as follows: Following the resource grid partitioning method, RB allocation is performed in the frequency domain. After determining the total bandwidth and subcarrier spacing, the number of RBs available in a time slot is also fixed. Users transmitting data using different RBs will not experience interference. The Within each time slot, the user The RB allocation is as follows: Users in the same cell share the same resource grid, therefore in cell n: An RB belonging to the same resource grid can only be allocated to one user in a time slot, therefore: Therefore, within the entire BHTS, the resource grid allocation for each cell is independent and unrelated.

[0057] Therefore, under the conditions of introducing spatial isolation and resource grids, Updated to: exist In the middle, restrictions Updated to any time slot In this process, the allocated power all meets the satellite's maximum power limit; limiting conditions and Remain unchanged; Update the bandwidth limit to a binary variable. Constraints. In step four, a three-layer scheduling architecture for the LEO hopping beam system is designed to complete the BHP decision for the current cycle.

[0058] This design scheme divides beam allocation on the BHP time scale, while power and bandwidth allocation are divided on the time slots; the beam scheduling period is defined as... The resource grid time slots for power and bandwidth allocation are Then there is Therefore, the above In the middle, restrictions and What happened On a time scale; What happened In other words, once the beam scheduling scheme is determined within the BHP, it will not affect subsequent [processes / procedures]. and Allocation. Secondly, regarding the conditions... If the requirement is met throughout the entire BHP time period, then as a global constraint, it is also satisfied in every time slot within the BHP. Therefore, it is defined in The solution is as follows: The specific scheduling framework model is as follows Figure 3 As shown.

[0059] Determine the weights for beam scheduling to select the highest priority BHP; After determining the scheduling beam, step five involves joint optimization of power and bandwidth allocation. At the aforementioned beam scheduling scale, the selection of scheduling cells is completed; at the time slot scale, the weights for user scheduling are determined to allocate power and frequency resources. Therefore, the power and bandwidth optimization problem simplifies to: With the goal of maximizing the system capacity of the channel, the utility function is expressed as: Among users bandwidth used express, ,Depend on The calculation formula for noise power can be obtained from the following formula: First, constraint (a) ensures that the allocated power does not exceed the maximum value that the satellite system can provide, and that the power allocation is continuous. Constraint (b) ensures that the allocated RBs do not exceed the upper limit of the bandwidth that the satellite system can provide, and The optimization variable is an integer variable. Constraint (b) guarantees that the power allocated to each user is non-negative and does not exceed the upper limit. The objective function, which aims to maximize channel capacity, is a nonlinear function of power and RB allocation. Therefore, this optimization problem is a mixed integer nonlinear optimization problem (MINLP) and also a nonconvex problem.

[0060] The optimization problem described above is a non-convex problem with constrained coupling between variables, making it difficult to solve directly. Generally, this problem is NP-hard, especially when there are multiple users and interfering factors, where the computational complexity of searching for the optimal solution increases exponentially with the number of users. Therefore, we decouple the original optimization problem into two sub-problems.

[0061] First, fixed bandwidth allocation ,optimization Substituting constraints (a) and (c), construct the Lagrange function: .

[0062] In practical implementation, to differentiate QoS for different services and maintain the data packets to be transmitted by different users, thus achieving a complete scheduling process, this solution fully integrates the scheduling function of the terrestrial access network protocol stack into the satellite communication system. In the terrestrial network, the MAC layer executes scheduling decisions based on scheduling information fed back from upper and lower layers, such as Channel Quality Indicator (CQI) and buffer information. Therefore, this solution designs an infinitely long queue for each user to be scheduled at the MAC layer of the satellite system. Simultaneously, in accordance with the 5G New Radio (5G NR) specifications, the upper-layer Service Data Adaptation Protocol (SDAP) layer maintains the QoS indicator requirements for different services and transmits the traffic information requested by each user and the corresponding QoS indicators to the MAC layer. Each user requests only one type of traffic service, and for each type of service requested by the user... , ,in For all selectable service types, all service QoS indicators are considered. ,in , , It also takes into account the default priority of each business type.

[0063] For request business type users The QoS indicators it needs to meet are: Secondly, the satellite's MAC layer is for the request Users of similar services Maintain an infinitely long queue to be scheduled, denoted as . Among them, in the queue Indicates a request Users of similar services The queue to be scheduled Data packets, This indicates the queue length. Additionally, the satellite system's MAC layer stores the arrival time for each arriving data packet.

[0064] In some embodiments of the present invention, in the step of calculating the power allocated to each user in the sub-region corresponding to the beam every preset second time slot, the transmit power allocated to each user is determined by a weighted geometric water-filling algorithm based on a preset transmit power allocation upper limit, the initial bandwidth allocation of each user, the channel gain and weight.

[0065] The weighted geometric water injection algorithm is adopted, which is simpler to implement, has lower complexity, and the solution process follows the KKT conditions, thus obtaining the optimal solution for power allocation under the current conditions. The pseudocode of the algorithm is shown below: According to the above algorithm, the water level line ,in: It represents steps. The amount of water above (i.e., power) can be obtained by subtracting the amount of water below the steps from the total power. Find the way to make The highest step that is not zero is thus the water level line.

[0066] Therefore there is .

[0067] The power optimization problem is formulated as a nonlinear optimization problem with inequality constraints. It can be proven that the method satisfies the KKT conditions, and therefore the output power allocation solution is the optimal solution to the problem.

[0068] To obtain a solution for power allocation under the current conditions, and thereby optimize bandwidth. Allocation. Similarly, establish the Lagrange function: In some embodiments of the present invention, in the step of determining the transmission power allocated to each user based on a weighted geometric water-filling algorithm using a preset transmission power allocation upper limit, initial bandwidth allocation for each user, channel gain and weight, the initial bandwidth allocation for each user is the allocation result obtained by averaging the total bandwidth.

[0069] In some embodiments of the present invention, in the step of calculating the bandwidth allocated to each user based on the power allocated to each user at every preset second time slot; Obtain the carrier-to-noise-plus-interference ratio under the current power allocation, and map the continuous bandwidth to a discrete set; The throughput is maximized by using discrete sets and the carrier-to-noise-plus-interference ratio, with the total bandwidth used as a constraint to initialize the particle swarm velocity and position. The particle swarm velocity and position are initialized based on fitness updates to obtain the final bandwidth allocation scheme.

[0070] Specifically, the power scheduler uses resource block (RB) scheduling results as its input, while bandwidth allocation also requires the output of power allocation. These two stages iteratively optimize each other using the scheduling results as their respective inputs until the power allocation result is achieved. And until the RB allocation results converge. Since there are no RB allocation results when entering the power allocation phase for the first time, the RBs are initially allocated evenly as the optimization problem in the first iteration. The input is given, and the algorithm pseudocode is shown below: The beneficial effects of this plan include: 1. This proposal presents a multi-dimensional resource joint optimization method for satellite beam-hopping systems, constructing a collaborative resource allocation framework of "beam-hopping scheduling-power allocation-bandwidth configuration". First, the beam-hopping scheduling scheme is selected at the beam-level time scale. Then, in the physical layer resource allocation stage, power is optimized by fixing the bandwidth, and the power allocation result is used as input for bandwidth optimization. Finally, the bandwidth optimization result is fed back to the power allocation module, forming an iterative closed loop until convergence. Simultaneously, the power and bandwidth allocation schemes also influence the beam-hopping scheme selection for the next BHTS. In terms of modeling, a mixed-integer programming model incorporating binary beam scheduling variables is established, a ground-based hierarchical scheduling mechanism is introduced, satellite MAC layer functions are constructed, and a dynamic resource scheduling database is maintained to support differentiated service QoS requirements. Regarding interference suppression, inter-beam spatial isolation is used to avoid co-channel interference, and ground-assisted time-slot-level resource block scheduling is combined to suppress intra-beam interference, thereby decomposing the original NP-hard joint optimization problem into parallel and efficiently solvable sub-problems.

[0071] 2. This scheme proposes a multi-timescale hierarchical optimization mechanism. At the beam-level timescale, it addresses the cross-beam hopping scheduling problem by employing a weighted search strategy to determine the optimal beam combination for each beam cycle. At the slot-level timescale, it addresses the user-level power and bandwidth joint allocation problem by optimizing power and bandwidth separately through a step-by-step decomposition method, with feedback between them during the iteration process to achieve convergence. This time-domain decoupling strategy breaks down the complex global joint optimization problem into sub-problems with different time dimensions, significantly reducing computational complexity and improving system scheduling efficiency while maintaining optimization accuracy.

[0072] 3. This scheme jointly optimizes beam hopping scheduling, power, and bandwidth allocation to achieve efficient resource utilization. In the beam hopping scheduling system, firstly, a user traffic model is built to simulate the hierarchical ground scheduling function to maintain the user queue for scheduling; secondly, a downlink model of the beam hopping low-Earth orbit satellite communication system is constructed, and a joint optimization problem of beam hopping, power, and bandwidth is proposed; to maximize the utilization of onboard resources, a full-frequency reuse scheme is adopted between beams, therefore a distance threshold is set to avoid inter-beam co-channel interference; and inter-user interference within a beam is avoided by introducing a ground scheduling resource grid. In summary, this invention simplifies the optimization problem and reduces the solution space size of BHP scheduling.

[0073] 4. This scheme proposes a multi-timescale joint scheduling approach. BHP selection is performed at the beam-level time scale, and power and frequency allocation for users is performed at the time slot-level time scale. A beam-level time period contains multiple time slots. Based on this technology, a three-layer scheduling architecture is designed. At the beam-level scale, a weighted search scheme determines the current BHP, and at the time slot scale, joint optimization of power and bandwidth is performed. A step-by-step optimization method is adopted, with independent power and bandwidth scheduling for each optimization problem to ensure convergence of the optimization process. This invention, while maximizing system capacity, can maintain a high level of service satisfaction for all users. The multi-layer scheduling architecture achieves optimal power and bandwidth allocation in each iteration, thereby approximating the optimal solution of the joint optimization, while reducing solution complexity and saving computational resources.

[0074] This invention also provides a multi-dimensional resource allocation system for low-Earth orbit satellite agile beams. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0075] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned multi-dimensional resource allocation method for low-Earth orbit satellite agile beams. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0076] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0077] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0078] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0079] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-dimensional resource allocation method for low-Earth orbit satellite agile beams, characterized in that, The steps of this method include: The coverage area of ​​the low-Earth orbit satellite is determined every preset first time slot, and the coverage area of ​​the low-Earth orbit satellite is divided into a first preset number of sub-regions; Based on the number of beams and sub-regions of the low-Earth orbit satellites, a beam-to-sub-region allocation scheme is constructed. The weight parameters for each allocation scheme are calculated. The final allocation scheme to be executed is determined based on these weight parameters. The steps include calculating the penalty weight function value for each sub-region using the following formula: ； in, This represents the penalty weight function value for subregion n. Represents users in subregion n The head delay of the queue to be scheduled. This represents the total number of users in subregion n. This represents the calculation parameter corresponding to the maximum allowable delay of the data packet from the terminal to the core network gateway, where s represents the service type; The weight parameters of each allocation scheme are calculated based on the penalty weight function value of each sub-region; the allocation scheme with the largest weight parameter is selected as the final allocation scheme to be executed. The first time slot is evenly divided into multiple second time slots, and the power allocated to each user in the sub-region corresponding to the beam is calculated every preset second time slot; Every second time slot, based on the power allocated to each user, the bandwidth allocated to each user is calculated using the particle swarm optimization algorithm.

2. The multi-dimensional resource allocation method for low-Earth orbit satellite agile beams according to claim 1, characterized in that, In the step of calculating the power allocated to each user in the sub-region corresponding to the beam, the weighted geometric water-filling algorithm is used to calculate the power allocated to each user based on the initially set user weights, the assumed bandwidth allocation results, and the power allocation upper limit.

3. The multi-dimensional resource allocation method for low-Earth orbit satellite agile beams according to claim 1, characterized in that, In the step of constructing the beam-to-subregion allocation scheme based on the number of beams and subregions of low-Earth orbit satellites, each beam corresponds to one subregion, and a preset number of allocation schemes are constructed.

4. The multi-dimensional resource allocation method for low-Earth orbit satellite agile beams according to claim 1, characterized in that, In the step of calculating the power allocated to each user in the sub-region corresponding to the beam every preset second time slot, the transmit power allocated to each user is determined by a weighted geometric water-filling algorithm based on the preset transmit power allocation upper limit, the initial bandwidth allocation of each user, the channel gain and weight.

5. The multi-dimensional resource allocation method for low-Earth orbit satellite agile beams according to claim 4, characterized in that, In the step of determining the transmit power allocated to each user based on the pre-set transmit power allocation limit, the initial bandwidth allocation for each user, the channel gain and weight using the weighted geometric water-filling algorithm, the initial bandwidth allocation for each user is the allocation result obtained by averaging the total bandwidth.

6. The multi-dimensional resource allocation method for low-Earth orbit satellite agile beams according to claim 1, characterized in that, In the step of calculating the bandwidth allocated to each user based on the power allocated to each user in every preset second time slot; Obtain the carrier-to-noise-plus-interference ratio under the current power allocation, and map the continuous bandwidth to a discrete set; The throughput is maximized by using discrete sets and the carrier-to-noise-plus-interference ratio, with the total bandwidth used as a constraint to initialize the particle swarm velocity and position. The particle swarm velocity and position are initialized based on fitness updates to obtain the final bandwidth allocation scheme.

7. A multi-dimensional resource allocation system for low-Earth orbit satellite agile beams, characterized in that, The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method as described in any one of claims 1 to 6.

Images