A satellite system resource scheduling method and device based on multi-user satisfaction degree

Abstract

This invention discloses a satellite system resource scheduling method and apparatus based on multi-user satisfaction, belonging to the field of satellite resource scheduling technology. The method involves: real-time acquisition of satellite system resource data, user service demand data, channel status data, and QoS satisfaction; inputting the acquired data into an adversarial neural network to repeatedly execute resource allocation operations and output a target resource allocation scheme for resource scheduling of each user. The resource allocation operation is as follows: generating a current resource allocation scheme based on the input data and current network parameters; calculating the current user satisfaction level based on the current resource allocation scheme and the QoS satisfaction index range, and then calculating the user satisfaction variance; when the user satisfaction variance is less than a preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance. This invention can solve the problem of low resource utilization efficiency in satellite systems.

Description

Technical Field

[0001] This invention relates to the field of satellite resource scheduling technology, and in particular to a satellite system resource scheduling method and apparatus based on multi-user satisfaction. Background Technology

[0002] Low-Earth orbit (LEO) satellite communication systems are widely used in multi-user concurrent communication scenarios (such as emergency rescue and remote area interconnection) due to their advantages of wide coverage and low transmission latency. Currently, the mainstream technologies for multi-user resource scheduling on LEO satellites mainly include static priority scheduling and fixed QoS optimization. Static priority scheduling refers to allocating resources based on each user's service type, assigning fixed resources to preset service types. Fixed QoS optimization involves using simple machine learning algorithms to optimize the scheduling strategy by fitting historical resource allocation data; the user satisfaction index used in optimizing the scheduling strategy is obtained from the historical resource allocation data.

[0003] However, current mainstream technologies suffer from a lack of fairness and decreased resource utilization efficiency. Allocating resources solely based on business type leads to a bias towards a few fixed business types, impacting user satisfaction with peripheral business needs and resulting in significant differences in service quality among users. Furthermore, user needs are constantly changing; optimizing current scheduling strategies based on historical resource allocation data and user satisfaction levels fails to adapt to current user demands, further hindering resource utilization efficiency. Summary of the Invention

[0004] This invention provides a satellite system resource scheduling method and apparatus based on multi-user satisfaction, which can solve the problem of low resource utilization efficiency in existing satellite systems.

[0005] To address the aforementioned technical problems, this invention provides a satellite system resource scheduling method based on multi-user satisfaction, comprising: Real-time acquisition of satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range; The satellite system resource data, the service demand data of each user, and the channel status data are input into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme. Resource scheduling is performed for each user based on the target resource allocation scheme; The resource allocation operation includes: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, a current resource allocation scheme is generated. For each user, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. Calculate the variance of user satisfaction based on the current user satisfaction level of each user. When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

[0006] As a preferred embodiment, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. This includes: The current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel status data, the service requirement data, and the latency satisfaction index range in QoS satisfaction. Calculate the current bandwidth satisfaction based on the power allocation ratio and time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the service demand data, and the bandwidth satisfaction index range in the QoS satisfaction. Calculate the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel status data, and the bit error rate satisfaction index range in QoS satisfaction. The current user satisfaction is obtained by weighting the current latency satisfaction, the current bandwidth satisfaction, and the current bit error rate satisfaction.

[0007] As a preferred embodiment, the step of calculating the current latency satisfaction based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel state data, the service demand data, and the latency satisfaction index range in QoS satisfaction includes: Determine available spectrum bandwidth based on satellite system resource data; The available spectrum bandwidth is allocated according to the spectrum allocation ratio in the current resource allocation scheme to determine the spectrum bandwidth allocated to the user; The user channel transmission rate is calculated based on the user-allocated spectrum bandwidth and the user signal-to-noise ratio and channel attenuation coefficient in the channel state data. Obtain the business type and latency requirements from the business requirement data, and determine the amount of data for a single business based on the business type; Calculate the time required to transmit the single service data volume at the user channel transmission rate to obtain the data transmission time; The user transmission delay is obtained by superimposing the data transmission time and the preset system fixed delay. When the user's transmission latency is less than or equal to the latency requirement, the current latency satisfaction level is set to 1. When the user transmission latency is greater than the latency requirement, the difference between the user transmission latency and the latency requirement is determined as the consumed latency tolerance resource amount; the difference between the latency requirement and the maximum latency value in the latency satisfaction index range is determined as the total latency tolerance resource amount; the remaining latency tolerance resource percentage is calculated based on the consumed latency tolerance resource amount and the total latency tolerance resource amount, and the remaining latency tolerance resource percentage is determined as the current latency satisfaction level; The user channel transmission rate is calculated using the following formula: In the formula, Let i be the user channel transmission rate; Allocate spectrum bandwidth to user i's users; For user i, the signal-to-noise ratio; Let be the channel attenuation coefficient for user i.

[0008] As a preferred embodiment, the step of calculating the current bandwidth satisfaction based on the power allocation ratio, time slot allocation ratio, satellite system resource data, service demand data, and bandwidth satisfaction index range in the current resource allocation scheme includes: Available power is determined based on satellite system resource data, and bandwidth requirements are obtained from service demand data; The available power is allocated according to the power allocation ratio in the current resource allocation scheme to determine the user's allocated transmission power; Based on the user-allocated transmit power, the nonlinear gain of power with respect to effective bandwidth is calculated, and the transmit power gain correction coefficient is obtained. The user-allocated spectrum bandwidth is corrected based on the transmit power gain correction coefficient to obtain the actual bandwidth occupied by the user; When the actual bandwidth used by the user is greater than or equal to the bandwidth requirement, the current bandwidth satisfaction level is set to 1. When the actual bandwidth used by the user is less than the bandwidth requirement, the difference between the bandwidth requirement and the actual bandwidth used by the user is determined as the amount of bandwidth resources consumed; the total amount of bandwidth resources is calculated based on the maximum and minimum bandwidth values ​​in the bandwidth satisfaction index range; the proportion of remaining bandwidth resources is calculated based on the amount of bandwidth resources consumed and the total amount of bandwidth resources, and the proportion of remaining bandwidth resources is determined as the current bandwidth satisfaction.

[0009] As a preferred embodiment, the step of calculating the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel state data, and the bit error rate satisfaction index range in QoS satisfaction includes: Based on the channel attenuation coefficient in the channel state data, the effective received power of the user-allocated transmit power after channel attenuation is calculated. Based on the natural exponential function, the effective received power is converted into an exponential decay factor for the bit error rate as the effective received power changes. The user's actual bit error rate is obtained by correcting the preset benchmark bit error rate based on the exponential decay factor. Obtain the bit error rate requirement from the business requirements data; When the user's actual bit error rate is less than or equal to the bit error rate requirement, the current bit error rate satisfaction level is set to 1. When the user's actual bit error rate is greater than the bit error rate requirement, the difference between the user's actual bit error rate and the bit error rate requirement is determined as the amount of bit error rate resources consumed; the total amount of all bit error rate resources is calculated based on the maximum and minimum bit error rate values ​​within the bit error rate satisfaction index range; the remaining bit error rate resource ratio is calculated based on the consumed bit error rate resources and the total amount of all bit error rate resources, and the remaining bit error rate resource ratio is determined as the current bit error rate satisfaction level.

[0010] As a preferred embodiment, the resource scheduling for each user based on the target resource allocation scheme includes: In the target resource allocation scheme, the target spectrum allocation ratio, target power allocation ratio, and target time slot allocation ratio for each user are obtained respectively. For each user, the beamforming weights are adjusted according to the target spectrum allocation ratio and the target power allocation ratio. The codec type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio. Determine routing priority based on the latency requirements and target time slot allocation ratio; The target resources for each user are determined based on the target resource allocation scheme, and the target resources are allocated to the corresponding users based on each user's beamforming weight, codec type, and routing priority.

[0011] As a preferred embodiment, beamforming weights are adjusted according to the target spectrum allocation ratio and the target power allocation ratio, including: Determine the user's channel matrix based on channel state data; Based on the channel matrix, the target spectrum allocation ratio, the target power allocation ratio, and the available power, the beamforming weights for each user are calculated using the following formula: In the formula, Beamforming weights for user i; Let be the channel matrix for user i; Let be the transpose of the channel matrix of user i; Preset noise power; The target spectrum allocation ratio for user i; The target power allocation ratio for user i; This represents the available power.

[0012] As a preferred embodiment, the encoding / decoding type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio, including: When the business type is real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Turbo code; When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset real-time power allocation ratio, the code type is determined to be Turbo code. When the business type is non-real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Polar code; When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset non-real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset non-real-time power allocation ratio, the encoding and decoding type is determined to be Polar code.

[0013] As a preferred embodiment, the routing priority is determined based on the latency requirements and the target time slot allocation ratio, including: When the target time slot allocation ratio is not less than the first time slot allocation ratio threshold and the latency requirement is less than the first latency threshold, the routing priority is determined as the first priority. When the target time slot allocation ratio meets the preset time slot allocation ratio condition and the latency requirement meets the preset latency condition, the routing priority is determined to be the second priority; wherein, the preset time slot allocation ratio condition is that the target time slot allocation ratio is not less than the second time slot allocation ratio threshold and is less than the first time slot allocation ratio threshold, and the preset latency condition is that the latency requirement is not less than the first latency threshold and is not greater than the second latency threshold. When the target time slot allocation ratio is less than the second time slot allocation ratio threshold and the latency requirement is greater than the second latency threshold, the routing priority is determined to be the third priority. Among them, the second time slot allocation ratio threshold is less than the first time slot allocation ratio threshold; the first delay threshold is less than the second delay threshold; the second priority is higher than the first priority and lower than the third priority.

[0014] Accordingly, the present invention provides a satellite system resource scheduling device based on multi-user satisfaction, comprising: a data acquisition module, a resource allocation scheme generation module, and a resource scheduling module; The data acquisition module is used to collect satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range in real time. The resource allocation scheme generation module is used to input the satellite system resource data, the service demand data of each user and the channel status data into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme. The resource scheduling module is used to schedule resources for each user based on the target resource allocation scheme. The resource allocation operation includes: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, a current resource allocation scheme is generated. For each user, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. Calculate the variance of user satisfaction based on the current user satisfaction level of each user. When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

[0015] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention provides a satellite system resource scheduling method based on multi-user satisfaction. It collects real-time service demand data, channel state data, and QoS satisfaction data from each user, as well as satellite system resource data for scheduling. The collected data is input into an adversarial neural network (ANN), which repeatedly executes resource allocation operations to output a target resource allocation scheme for resource scheduling of each user. The resource allocation operation involves: generating a current resource allocation scheme based on the input data of the ANN and current network parameters; for each user, calculating current user satisfaction based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range; calculating the user satisfaction variance based on the current user satisfaction of each user; when the user satisfaction variance is less than a preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the ANN are adjusted according to the user satisfaction variance. This invention utilizes adversarial neural networks to generate target resource allocation schemes for each user, avoiding the bias of system resources towards fixed service types. Furthermore, when generating target resource allocation schemes, the adversarial neural networks optimize and adjust the schemes using the QoS satisfaction index range of each user collected in real time, which can better adapt to the needs of each user. Executing the target resource allocation schemes can improve the resource utilization efficiency of the satellite system. Attached Figure Description

[0016] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating an embodiment of the satellite system resource scheduling method based on multi-user satisfaction provided by the present invention; Figure 2 A flowchart illustrating one embodiment of the resource allocation operation provided by the present invention; Figure 3 This is a schematic diagram of an embodiment of the satellite system resource scheduling device based on multi-user satisfaction provided by the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0020] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0023] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0024] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0025] See Figure 1 To address the problem of low resource utilization efficiency in existing satellite systems, an embodiment of the present invention provides a satellite system resource scheduling method based on multi-user satisfaction. This method includes steps 101 to 103, each step of which is detailed below: Step 101: Real-time collection of satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range.

[0026] In this embodiment of the invention, satellite system resource scheduling is performed by first acquiring satellite system resource data, including available spectrum bandwidth, available power, and available time slots. Simultaneously, service requirement data, channel state data, and QoS satisfaction index ranges for each user to be scheduled are acquired. Service requirement data includes service type, latency requirements, bandwidth requirements, and bit error rate requirements. Service types include real-time data services and non-real-time data services. Channel state data includes user signal-to-noise ratio, Doppler shift, and channel attenuation coefficient. QoS satisfaction index ranges include latency satisfaction index ranges, bandwidth satisfaction index ranges, and bit error rate satisfaction index ranges.

[0027] Step 102: Input the satellite system resource data, the service demand data of each user, and the channel state data into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme.

[0028] In this embodiment of the invention, the adversarial neural network includes a generator and a discriminator. Satellite system resource data, user service demand data, and channel state data are input into the adversarial neural network. The generator generates a resource allocation scheme based on the input data and calculates the current user satisfaction for each user corresponding to the current resource allocation scheme. The discriminator is input to the resource allocation scheme generated by the generator and the current user satisfaction for each user. The discriminator determines whether the current resource allocation scheme is the optimal one by calculating the variance of user satisfaction. When the discriminator determines that the current resource allocation scheme is the optimal one, it directly identifies the current resource allocation scheme as the target resource allocation scheme; when the discriminator determines that the current resource allocation scheme is not the target resource allocation scheme, it adjusts the network parameters of the generator based on the variance of user satisfaction. The generator of the adversarial neural network is trained based on satellite system resource samples, user service demand samples, channel state samples, and real resource allocation scheme samples. The discriminator of the adversarial neural network is trained based on training resource allocation schemes generated by the generator and real resource allocation scheme samples.

[0029] See Figure 2 This is a flowchart illustrating an embodiment of the resource allocation operation provided by the present invention. The resource allocation operation includes the following steps 201 to 204, each of which is detailed below: Step 201: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, generate the current resource allocation scheme.

[0030] In this embodiment of the invention, the adversarial neural network is pre-configured with network parameters. When satellite system resource data, service demand data of each user, and channel state data are input into the adversarial neural network, intelligent calculations and optimization decisions are performed based on the current network parameters of the adversarial neural network to generate a current resource allocation scheme. The current resource allocation scheme includes the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio for each user.

[0031] Step 202: For each user, calculate the current user satisfaction based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range.

[0032] In this embodiment of the invention, after generating the current resource allocation scheme, the scheme is evaluated to determine whether it is a balanced resource allocation scheme that meets the needs of all users, taking into account the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range within the QoS satisfaction index range. Specifically, the current user satisfaction is calculated as the evaluation standard, and each user's current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range within the QoS satisfaction index range.

[0033] As a preferred embodiment, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range, including: The current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel status data, the service requirement data, and the latency satisfaction index range in QoS satisfaction. Calculate the current bandwidth satisfaction based on the power allocation ratio and time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the service demand data, and the bandwidth satisfaction index range in the QoS satisfaction. Calculate the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel status data, and the bit error rate satisfaction index range in QoS satisfaction. The current user satisfaction is obtained by weighting the current latency satisfaction, the current bandwidth satisfaction, and the current bit error rate satisfaction.

[0034] In this embodiment of the invention, the current user satisfaction consists of three parts: current latency satisfaction, current bandwidth satisfaction, and current bit error rate satisfaction. Specifically, the current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, satellite system resource data, channel state data, service demand data, and the latency satisfaction index range in QoS satisfaction. The current bandwidth satisfaction is calculated based on the power allocation ratio in the current resource allocation scheme, time slot allocation ratio, satellite system resource data, service demand data, and the bandwidth satisfaction index range in QoS satisfaction. The current bit error rate satisfaction is calculated based on the power allocation ratio in the current resource allocation scheme, service demand data, and the bit error rate satisfaction index range in QoS satisfaction. After calculating the current latency satisfaction, current bandwidth satisfaction, and current bit error rate satisfaction, they are weighted and summed using preset weights to calculate the current user satisfaction.

[0035] The current user satisfaction level can be calculated using the following formula: In the formula, The current user satisfaction level of user i; The current latency satisfaction of user i; The current bandwidth satisfaction of user i; The current bit error rate satisfaction of user i; b and c are preset weights. The value of can be 0.4, the value of b can be 0.3, and the value of c can be 0.3.

[0036] As a preferred embodiment, the current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel state data, the service demand data, and the latency satisfaction index range in QoS satisfaction, including: Determine available spectrum bandwidth based on satellite system resource data; The available spectrum bandwidth is allocated according to the spectrum allocation ratio in the current resource allocation scheme to determine the spectrum bandwidth allocated to the user; The user channel transmission rate is calculated based on the user-allocated spectrum bandwidth and the user signal-to-noise ratio and channel attenuation coefficient in the channel state data. Obtain the business type and latency requirements from the business requirement data, and determine the amount of data for a single business based on the business type; Calculate the time required to transmit the single service data volume at the user channel transmission rate to obtain the data transmission time; The user transmission delay is obtained by superimposing the data transmission time and the preset system fixed delay. When the user's transmission latency is less than or equal to the latency requirement, the current latency satisfaction level is set to 1. When the user transmission latency is greater than the latency requirement, the difference between the user transmission latency and the latency requirement is determined as the consumed latency tolerance resource amount; the difference between the latency requirement and the maximum latency value in the latency satisfaction index range is determined as the total latency tolerance resource amount; the remaining latency tolerance resource percentage is calculated based on the consumed latency tolerance resource amount and the total latency tolerance resource amount, and the remaining latency tolerance resource percentage is determined as the current latency satisfaction level; The user channel transmission rate is calculated using the following formula: In the formula, Let i be the user channel transmission rate; Allocate spectrum bandwidth to user i's users; For user i, the signal-to-noise ratio; Let be the channel attenuation coefficient for user i.

[0037] In this embodiment of the invention, the current latency satisfaction is calculated by first calculating the user-allocated spectrum bandwidth based on the spectrum allocation ratio in the current resource allocation scheme and the available spectrum bandwidth in the satellite system resource data: In the formula, Allocate spectrum bandwidth to user i's users; The spectrum allocation ratio for user i; The available spectrum bandwidth for user i.

[0038] In this embodiment of the invention, calculating the current latency satisfaction requires calculating the user transmission latency caused by transmitting service data based on the user channel transmission rate. Different service types correspond to different amounts of single-service data, resulting in different calculated user transmission latency.

[0039] Therefore, to calculate user transmission delay, we first need to determine the type of service the user is performing, and then determine the amount of data per service based on the service type. Based on the user channel transmission rate and the amount of data per service, the user transmission delay is calculated using the following formula: In the formula, For user i, the user transmission delay; Let i be the user channel transmission rate; This refers to the data volume of a single business stream. Set a fixed system delay. This represents the data transmission time. For non-real-time data services, the data volume per single service stream can be 10. Real-time data services include real-time voice services and real-time video services. The data volume per single service stream for real-time voice services can be 0.01, and the data volume per single service stream for real-time video services can be 1.

[0040] In this embodiment of the invention, by comparing the calculated user transmission latency with the latency requirements in the service demand data, the current latency satisfaction can be further calculated. The current latency satisfaction can be calculated using the following formula: In the formula, The current latency satisfaction of user i; For user i, the user transmission delay; For user i's latency requirements; This represents the maximum value of the latency indicator; This represents the amount of latency tolerance resources already consumed. This represents the total amount of latency-tolerant resources. This represents the percentage of remaining latency tolerance resources.

[0041] Based on the above formula, when the user's transmission delay is less than or equal to the delay requirement, the current delay satisfaction is 1; when the user's transmission delay is greater than the delay requirement, the current delay satisfaction is calculated based on the maximum delay value in the range of user transmission delay, delay requirement, and delay satisfaction index.

[0042] As a preferred embodiment, the current bandwidth satisfaction is calculated based on the power allocation ratio, time slot allocation ratio, satellite system resource data, service demand data, and bandwidth satisfaction index range in the current resource allocation scheme, including: Available power is determined based on satellite system resource data, and bandwidth requirements are obtained from service demand data; The available power is allocated according to the power allocation ratio in the current resource allocation scheme to determine the user's allocated transmission power; Based on the user-allocated transmit power, the nonlinear gain of power with respect to effective bandwidth is calculated, and the transmit power gain correction coefficient is obtained. The user-allocated spectrum bandwidth is corrected based on the transmit power gain correction coefficient to obtain the actual bandwidth occupied by the user; When the actual bandwidth used by the user is greater than or equal to the bandwidth requirement, the current bandwidth satisfaction level is set to 1. When the actual bandwidth used by the user is less than the bandwidth requirement, the difference between the bandwidth requirement and the actual bandwidth used by the user is determined as the amount of bandwidth resources consumed; the total amount of bandwidth resources is calculated based on the maximum and minimum bandwidth values ​​in the bandwidth satisfaction index range; the proportion of remaining bandwidth resources is calculated based on the amount of bandwidth resources consumed and the total amount of bandwidth resources, and the proportion of remaining bandwidth resources is determined as the current bandwidth satisfaction.

[0043] In this embodiment of the invention, the current bandwidth satisfaction is calculated by first calculating the user-allocated transmit power based on the power allocation ratio in the current resource allocation scheme and the available power in the satellite system resource data: In the formula, Assign transmit power to user i's user; The power allocation ratio for user i; The available power for user i.

[0044] In this embodiment of the invention, calculating the current bandwidth satisfaction requires calculating the actual bandwidth occupied by the user, which can be obtained by correcting the user's allocated spectrum bandwidth.

[0045] The formula for calculating the actual bandwidth used by the user is: In the formula, The actual bandwidth used by user i; Allocate spectrum bandwidth to user i's users; Assign transmit power to user i's user; This serves as the reference power standard. Wherein, This is the transmit power gain correction factor.

[0046] In this embodiment of the invention, by comparing the calculated actual bandwidth occupied by the user with the bandwidth demand in the service demand data, the current bandwidth satisfaction can be further calculated. The current bandwidth satisfaction can be calculated using the following formula: In the formula, For bandwidth satisfaction; The actual bandwidth used by user i; For user i's bandwidth requirements; This represents the maximum value of the bandwidth indicator. This represents the minimum bandwidth specification. This represents the amount of bandwidth resources already consumed. This refers to the total amount of bandwidth resources. This represents the percentage of remaining bandwidth resources.

[0047] Based on the above formula, when the user's actual bandwidth usage is greater than or equal to the bandwidth demand, the current bandwidth satisfaction is 1; when the user's actual bandwidth usage is less than the bandwidth demand, the current bandwidth satisfaction is calculated based on the user's actual bandwidth usage, bandwidth demand, maximum bandwidth value, and minimum bandwidth value.

[0048] As a preferred embodiment, the current bit error rate satisfaction is calculated based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel state data, and the bit error rate satisfaction index range in QoS satisfaction, including: Based on the channel attenuation coefficient in the channel state data, the effective received power of the user-allocated transmit power after channel attenuation is calculated. Based on the natural exponential function, the effective received power is converted into an exponential decay factor for the bit error rate as the effective received power changes. The user's actual bit error rate is obtained by correcting the preset benchmark bit error rate based on the exponential decay factor. Obtain the bit error rate requirement from the business requirements data; When the user's actual bit error rate is less than or equal to the bit error rate requirement, the current bit error rate satisfaction level is set to 1. When the user's actual bit error rate is greater than the bit error rate requirement, the difference between the user's actual bit error rate and the bit error rate requirement is determined as the amount of bit error rate resources consumed; the total amount of all bit error rate resources is calculated based on the maximum and minimum bit error rate values ​​within the bit error rate satisfaction index range; the remaining bit error rate resource ratio is calculated based on the consumed bit error rate resources and the total amount of all bit error rate resources, and the remaining bit error rate resource ratio is determined as the current bit error rate satisfaction level.

[0049] In this embodiment of the invention, calculating the current bandwidth satisfaction requires calculating the user's actual bit error rate, which can be obtained by correcting the preset benchmark bit error rate.

[0050] The formula for calculating the actual bit error rate for users is as follows: In the formula, The actual bit error rate for user i; The preset baseline bit error rate can take the following values: ; Assign transmit power to user i's user; Reference power; Let be the channel attenuation coefficient for user i; Allocate the effective received power after channel attenuation to the user's transmit power; It is an exponential decay factor.

[0051] In this embodiment of the invention, by comparing the calculated actual bit error rate of the user with the bit error rate requirement in the business demand data, the current bit error rate satisfaction can be further calculated. The current bit error rate satisfaction can be calculated using the following formula: In the formula, The current bit error rate satisfaction of user i; The actual bit error rate for the user; For bit error rate requirements; This represents the maximum value of the bit error rate indicator. This represents the minimum value of the bit error rate metric. This represents the amount of resources already consumed for the bit error rate. This represents the total amount of resources used for the entire bit error rate; This represents the percentage of remaining error rate resources.

[0052] Based on the above formula, when the user's actual bit error rate is greater than or equal to the bit error rate requirement, the current bit error rate satisfaction is determined to be 1; when the user's actual bit error rate is less than the bit error rate requirement, the current bit error rate satisfaction is calculated based on the user's actual bit error rate, bit error rate requirement, maximum bit error rate index, and minimum bit error rate index.

[0053] Step 203: Calculate the variance of user satisfaction based on the current user satisfaction of each user.

[0054] In this embodiment of the invention, after calculating the user satisfaction of each user, the user satisfaction variance can be calculated. The user satisfaction variance is used to determine whether the current resource allocation scheme balances the requirements of each user, thereby determining whether the current resource allocation scheme is the target resource allocation scheme.

[0055] Step 204: When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

[0056] In this embodiment of the invention, by comparing the user satisfaction variance with a preset variance, it can be determined whether the current resource allocation scheme is the target resource allocation scheme. Specifically, when the user satisfaction variance is less than the preset variance, the current resource allocation scheme is considered to balance the needs of all users, and thus the current resource allocation scheme is determined to be the target resource allocation scheme. However, when the user satisfaction variance is not less than the preset variance, the current resource allocation scheme is considered not the optimal resource allocation scheme. In this case, it is necessary to adjust the network parameters of the adversarial neural network according to the user satisfaction variance, and then use the updated adversarial neural network to regenerate the target resource allocation scheme.

[0057] Step 103: Perform resource scheduling for each user based on the target resource allocation scheme.

[0058] As a preferred embodiment, resource scheduling for each user is performed based on the target resource allocation scheme, including: In the target resource allocation scheme, the target spectrum allocation ratio, target power allocation ratio, and target time slot allocation ratio for each user are obtained respectively. For each user, the beamforming weights are adjusted according to the target spectrum allocation ratio and the target power allocation ratio. The codec type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio. Determine routing priority based on the latency requirements and target time slot allocation ratio; The target resources for each user are determined based on the target resource allocation scheme, and the target resources are allocated to the corresponding users based on each user's beamforming weight, codec type, and routing priority.

[0059] In this embodiment of the invention, resource scheduling for each user based on the target resource allocation scheme requires dynamic adjustment in three layers. At the physical layer, beamforming weights need to be adjusted according to the target spectrum allocation ratio and target power allocation ratio to ensure beam gain matches the spectrum allocation ratio and reduce interference between users. At the link layer, the encoding / decoding type needs to be determined based on the service type, bit error rate requirements, and target power allocation ratio to control encoding / decoding processing latency and redundancy overhead, ensuring transmission error rate performance and improving spectrum resource utilization. At the network layer, routing priorities need to be determined based on latency requirements and the target time slot allocation ratio to plan data forwarding paths, avoid congested links, ensure service transmission latency constraints, and reduce multi-user packet forwarding conflicts.

[0060] As a preferred embodiment, adjusting the beamforming weights according to the target spectrum allocation ratio and the target power allocation ratio includes: Determine the user's channel matrix based on channel state data; Based on the channel matrix, the target spectrum allocation ratio, the target power allocation ratio, and the available power, the beamforming weights for each user are calculated using the following formula: In the formula, Beamforming weights for user i; Let be the channel matrix for user i; Let be the transpose of the channel matrix of user i; Preset noise power; The target spectrum allocation ratio for user i; The target power allocation ratio for user i; This represents the available power.

[0061] In this embodiment of the invention, adjusting beamforming weights according to the target spectrum allocation ratio and the target power allocation ratio first requires obtaining the user's channel matrix. The channel matrix can be obtained in the following ways: First, noise signals are collected from N consecutive noise sampling points. ,in, Let N be the noise signal at the Nth noise sampling point, where N ≥ 1024. A statistical model of the current background noise is established using these noise signals to clarify the distribution pattern of the noise signal.

[0062] When solving for the channel matrix for each user, pilot sequences are sent to each user's terminal so that each terminal can receive and record the amplitude and phase information of the pilot signals. Based on the sent pilot sequences, the pilot signals received by the users, and the noise components determined based on the noise signal distribution pattern, the following channel input-output relationship can be established: In the formula, Let be the channel matrix for user i; The pilot signal with index p received by the user terminal; The pilot sequence with index p; This refers to the noise signal received by the user terminal when receiving the pilot signal with index p.

[0063] Therefore, after establishing the above channel input-output relationship, by substituting the known pilot signal, pilot sequence, and noise signal, the channel matrix of each user can be calculated.

[0064] After obtaining the channel matrix, the beamforming weights of each user are adjusted by combining the target spectrum allocation ratio, target power allocation ratio in the target resource allocation scheme, and available spectrum bandwidth in the satellite system resource data.

[0065] As a preferred embodiment, determining the encoding / decoding type based on the service type, the bit error rate requirement, and the target power allocation ratio includes: When the business type is real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Turbo code; When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset real-time power allocation ratio, the code type is determined to be Turbo code. When the business type is non-real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Polar code; When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset non-real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset non-real-time power allocation ratio, the encoding and decoding type is determined to be Polar code.

[0066] In this embodiment of the invention, the encoding / decoding type of the link layer is determined based on the service type, bit error rate requirement, and target power allocation ratio. For real-time data services, the encoding / decoding type can be either Turbo code or LDPC code. Specifically, when the bit error rate requirement is high, i.e., the bit error rate requirement is less than or equal to a preset bit error rate threshold, the encoding / decoding type is determined to be Turbo code regardless of the power allocation ratio; the preset bit error rate threshold can be set to 10⁻⁻⁶. 5 When the bit error rate requirement is a normal requirement (i.e., the required bit error rate is greater than the preset bit error rate threshold) and the power allocation is sufficient (i.e., the target power allocation ratio is not less than the preset real-time power allocation ratio), the encoding / decoding type is determined to be LDPC code; the preset real-time power allocation ratio can be set to 0.05. When the bit error rate requirement is a normal requirement (i.e., the required bit error rate is greater than the preset bit error rate threshold) and the power allocation is insufficient (i.e., the target power allocation ratio is less than the preset real-time power allocation ratio), the encoding / decoding type is determined to be Turbo code.

[0067] For non-real-time data services, the encoding / decoding type can be either Polar code or LDPC code. Specifically, when the bit error rate requirement is high, i.e., the bit error rate requirement is less than or equal to a preset bit error rate threshold, the encoding / decoding type is determined to be Polar code regardless of the power allocation ratio; the preset bit error rate threshold can be set to 10⁻⁻⁶. 5 When the bit error rate requirement is a normal requirement (i.e., the required bit error rate is greater than the preset bit error rate threshold) and the power allocation is sufficient (i.e., the target power allocation ratio is not less than the preset non-real-time power allocation ratio), the encoding / decoding type is determined to be LDPC code; the preset non-real-time power allocation ratio can be set to 0.03. When the bit error rate requirement is a normal requirement (i.e., the required bit error rate is greater than the preset bit error rate threshold) and the power allocation is insufficient (i.e., the target power allocation ratio is less than the preset non-real-time power allocation ratio), the encoding / decoding type is determined to be Polar code.

[0068] As a preferred embodiment, determining the routing priority based on the latency requirement and the target time slot allocation ratio includes: When the target time slot allocation ratio is not less than the first time slot allocation ratio threshold and the latency requirement is less than the first latency threshold, the routing priority is determined as the first priority. When the target time slot allocation ratio meets the preset time slot allocation ratio condition and the latency requirement meets the preset latency condition, the routing priority is determined to be the second priority; wherein, the preset time slot allocation ratio condition is that the target time slot allocation ratio is not less than the second time slot allocation ratio threshold and is less than the first time slot allocation ratio threshold, and the preset latency condition is that the latency requirement is not less than the first latency threshold and is not greater than the second latency threshold. When the target time slot allocation ratio is less than the second time slot allocation ratio threshold and the latency requirement is greater than the second latency threshold, the routing priority is determined to be the third priority. Among them, the second time slot allocation ratio threshold is less than the first time slot allocation ratio threshold; the first delay threshold is less than the second delay threshold; the second priority is higher than the first priority and lower than the third priority.

[0069] In this embodiment of the invention, the routing priority of each user can be determined based on latency requirements and the target time slot allocation ratio. Specifically, when the target time slot allocation ratio is not less than a first time slot allocation ratio threshold and the latency requirement is less than a first latency threshold, the routing priority is determined to be the first priority, i.e., the highest priority, and inter-satellite direct links are preferentially selected. When the target time slot allocation ratio is not less than a second time slot allocation ratio threshold and less than the first time slot allocation ratio threshold, and the latency requirement is not less than the first latency threshold and not greater than the second latency threshold, the routing priority is determined to be the second priority, and a "satellite-ground relay-satellite" link can be selected. When the target time slot allocation ratio is less than the second time slot allocation ratio threshold and the latency requirement is greater than the second latency threshold, the routing priority is determined to be the third priority, i.e., the highest priority, and multi-hop links can be selected to reduce interference to high-priority users.

[0070] In this embodiment of the invention, after determining the target resource allocation scheme and the beamforming weights, encoding / decoding types, and routing priorities of each user, the spectrum bandwidth to be allocated to each user, i.e., the target spectrum resources, is first calculated based on the target spectrum allocation ratio of each user in the target resource allocation scheme and the available spectrum bandwidth in the satellite system resource data; the transmit power to be allocated to each user, i.e., the target transmit power, is calculated based on the target power allocation ratio of each user in the target resource allocation scheme and the available power in the satellite system resource data; the transmission time slot to be allocated to each user, i.e., the target transmission time slot, is calculated based on the target time slot allocation ratio of each user in the target resource allocation scheme and the available time slot in the satellite system resource data; thus, the target resource quantization value, i.e., the target resources, is obtained for each user.

[0071] Based on this, and combined with the pre-determined beamforming weights for each user, the target transmit power and target spectrum resources are precisely directed to the corresponding user in the form of directional beams to suppress inter-user interference. According to each user's encoding and decoding type, resource block size and modulation / coding format are matched to their error correction capability, transmission rate, and latency characteristics to ensure transmission reliability and efficiency. Simultaneously, based on each user's routing priority, time slots and forwarding links are preferentially allocated to high-priority users at the network layer to ensure the latency and QoS requirements of critical services. Finally, the target spectrum resources, target transmit power, and target transmission time slots are allocated to the corresponding users according to their channel characteristics and service levels.

[0072] Implementing the above embodiments has the following effects: This invention provides a satellite system resource scheduling method based on multi-user satisfaction. It collects real-time service demand data, channel state data, and QoS satisfaction data from each user, as well as satellite system resource data for scheduling. The collected data is input into an adversarial neural network (ANN), which repeatedly executes resource allocation operations to output a target resource allocation scheme for resource scheduling of each user. The resource allocation operation involves: generating a current resource allocation scheme based on the input data of the ANN and current network parameters; for each user, calculating current user satisfaction based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range; calculating the user satisfaction variance based on the current user satisfaction of each user; when the user satisfaction variance is less than a preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the ANN are adjusted according to the user satisfaction variance. This invention utilizes adversarial neural networks to generate target resource allocation schemes for each user, avoiding the bias of system resources towards fixed service types. Furthermore, when generating target resource allocation schemes, the adversarial neural networks optimize and adjust the schemes using the QoS satisfaction index range of each user collected in real time, which can better adapt to the needs of each user. Executing the target resource allocation schemes can improve the resource utilization efficiency of the satellite system.

[0073] like Figure 3 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a satellite system resource scheduling device based on multi-user satisfaction, comprising: a data acquisition module, a resource allocation scheme generation module, and a resource scheduling module; The data acquisition module is used to collect satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range in real time. The resource allocation scheme generation module is used to input the satellite system resource data, the service demand data of each user and the channel status data into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme. The resource scheduling module is used to schedule resources for each user based on the target resource allocation scheme. The resource allocation operation includes: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, a current resource allocation scheme is generated. For each user, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. Calculate the variance of user satisfaction based on the current user satisfaction level of each user. When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

[0074] As a preferred embodiment, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range, including: The current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel status data, the service requirement data, and the latency satisfaction index range in QoS satisfaction. Calculate the current bandwidth satisfaction based on the power allocation ratio and time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the service demand data, and the bandwidth satisfaction index range in the QoS satisfaction. Calculate the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel status data, and the bit error rate satisfaction index range in QoS satisfaction. The current user satisfaction is obtained by weighting the current latency satisfaction, the current bandwidth satisfaction, and the current bit error rate satisfaction.

[0075] As a preferred embodiment, the current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel state data, the service demand data, and the latency satisfaction index range in QoS satisfaction, including: Determine available spectrum bandwidth based on satellite system resource data; The available spectrum bandwidth is allocated according to the spectrum allocation ratio in the current resource allocation scheme to determine the spectrum bandwidth allocated to the user; The user channel transmission rate is calculated based on the user-allocated spectrum bandwidth and the user signal-to-noise ratio and channel attenuation coefficient in the channel state data. Obtain the business type and latency requirements from the business requirement data, and determine the amount of data for a single business based on the business type; Calculate the time required to transmit the single service data volume at the user channel transmission rate to obtain the data transmission time; The user transmission delay is obtained by superimposing the data transmission time and the preset system fixed delay. When the user's transmission latency is less than or equal to the latency requirement, the current latency satisfaction level is set to 1. When the user transmission latency is greater than the latency requirement, the difference between the user transmission latency and the latency requirement is determined as the consumed latency tolerance resource amount; the difference between the latency requirement and the maximum latency value in the latency satisfaction index range is determined as the total latency tolerance resource amount; the remaining latency tolerance resource percentage is calculated based on the consumed latency tolerance resource amount and the total latency tolerance resource amount, and the remaining latency tolerance resource percentage is determined as the current latency satisfaction level; The user channel transmission rate is calculated using the following formula: In the formula, Let i be the user channel transmission rate; Allocate spectrum bandwidth to user i's users; For user i, the signal-to-noise ratio; Let be the channel attenuation coefficient for user i.

[0076] As a preferred embodiment, the current bandwidth satisfaction is calculated based on the power allocation ratio, time slot allocation ratio, satellite system resource data, service demand data, and bandwidth satisfaction index range in the current resource allocation scheme, including: Available power is determined based on satellite system resource data, and bandwidth requirements are obtained from service demand data; The available power is allocated according to the power allocation ratio in the current resource allocation scheme to determine the user's allocated transmission power; Based on the user-allocated transmit power, the nonlinear gain of power with respect to effective bandwidth is calculated, and the transmit power gain correction coefficient is obtained. The user-allocated spectrum bandwidth is corrected based on the transmit power gain correction coefficient to obtain the actual bandwidth occupied by the user; When the actual bandwidth used by the user is greater than or equal to the bandwidth requirement, the current bandwidth satisfaction level is set to 1. When the actual bandwidth used by the user is less than the bandwidth requirement, the difference between the bandwidth requirement and the actual bandwidth used by the user is determined as the amount of bandwidth resources consumed; the total amount of bandwidth resources is calculated based on the maximum and minimum bandwidth values ​​in the bandwidth satisfaction index range; the proportion of remaining bandwidth resources is calculated based on the amount of bandwidth resources consumed and the total amount of bandwidth resources, and the proportion of remaining bandwidth resources is determined as the current bandwidth satisfaction.

[0077] As a preferred embodiment, the current bit error rate satisfaction is calculated based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel state data, and the bit error rate satisfaction index range in QoS satisfaction, including: Based on the channel attenuation coefficient in the channel state data, the effective received power of the user-allocated transmit power after channel attenuation is calculated. Based on the natural exponential function, the effective received power is converted into an exponential decay factor for the bit error rate as the effective received power changes. The user's actual bit error rate is obtained by correcting the preset benchmark bit error rate based on the exponential decay factor. Obtain the bit error rate requirement from the business requirements data; When the user's actual bit error rate is less than or equal to the bit error rate requirement, the current bit error rate satisfaction level is set to 1. When the user's actual bit error rate is greater than the bit error rate requirement, the difference between the user's actual bit error rate and the bit error rate requirement is determined as the amount of bit error rate resources consumed; the total amount of all bit error rate resources is calculated based on the maximum and minimum bit error rate values ​​within the bit error rate satisfaction index range; the remaining bit error rate resource ratio is calculated based on the consumed bit error rate resources and the total amount of all bit error rate resources, and the remaining bit error rate resource ratio is determined as the current bit error rate satisfaction level.

[0078] As a preferred embodiment, resource scheduling for each user is performed based on the target resource allocation scheme, including: In the target resource allocation scheme, the target spectrum allocation ratio, target power allocation ratio, and target time slot allocation ratio for each user are obtained respectively. For each user, the beamforming weights are adjusted according to the target spectrum allocation ratio and the target power allocation ratio. The codec type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio. Determine routing priority based on the latency requirements and target time slot allocation ratio; The target resources for each user are determined based on the target resource allocation scheme, and the target resources are allocated to the corresponding users based on each user's beamforming weight, codec type, and routing priority.

[0079] As a preferred embodiment, adjusting the beamforming weights according to the target spectrum allocation ratio and the target power allocation ratio includes: Determine the user's channel matrix based on channel state data; Based on the channel matrix, the target spectrum allocation ratio, the target power allocation ratio, and the available power, the beamforming weights for each user are calculated using the following formula: In the formula, Beamforming weights for user i; Let be the channel matrix for user i; Let be the transpose of the channel matrix of user i; Preset noise power; The target spectrum allocation ratio for user i; The target power allocation ratio for user i; This represents the available power.

[0080] As a preferred embodiment, determining the encoding / decoding type based on the service type, the bit error rate requirement, and the target power allocation ratio includes: When the business type is real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Turbo code; When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset real-time power allocation ratio, the code type is determined to be Turbo code. When the business type is non-real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Polar code; When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset non-real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset non-real-time power allocation ratio, the encoding and decoding type is determined to be Polar code.

[0081] As a preferred embodiment, determining the routing priority based on the latency requirement and the target time slot allocation ratio includes: When the target time slot allocation ratio is not less than the first time slot allocation ratio threshold and the latency requirement is less than the first latency threshold, the routing priority is determined as the first priority. When the target time slot allocation ratio meets the preset time slot allocation ratio condition and the latency requirement meets the preset latency condition, the routing priority is determined to be the second priority; wherein, the preset time slot allocation ratio condition is that the target time slot allocation ratio is not less than the second time slot allocation ratio threshold and is less than the first time slot allocation ratio threshold, and the preset latency condition is that the latency requirement is not less than the first latency threshold and is not greater than the second latency threshold. When the target time slot allocation ratio is less than the second time slot allocation ratio threshold and the latency requirement is greater than the second latency threshold, the routing priority is determined to be the third priority. Among them, the second time slot allocation ratio threshold is less than the first time slot allocation ratio threshold; the first delay threshold is less than the second delay threshold; the second priority is higher than the first priority and lower than the third priority.

[0082] Implementing the above embodiments has the following effects: This invention provides a satellite system resource scheduling device based on multi-user satisfaction. It collects real-time service demand data, channel state data, and QoS satisfaction data from each user, as well as satellite system resource data for scheduling. The collected data is input into an adversarial neural network (ANN), which repeatedly executes resource allocation operations to output a target resource allocation scheme for resource scheduling of each user. The resource allocation operation involves: generating a current resource allocation scheme based on the input data of the ANN and current network parameters; for each user, calculating current user satisfaction based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range; calculating the user satisfaction variance based on the current user satisfaction of each user; when the user satisfaction variance is less than a preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the ANN are adjusted according to the user satisfaction variance. This invention utilizes adversarial neural networks to generate target resource allocation schemes for each user, avoiding the bias of system resources towards fixed service types. Furthermore, when generating target resource allocation schemes, the adversarial neural networks optimize and adjust the schemes using the QoS satisfaction index range of each user collected in real time, which can better adapt to the needs of each user. Executing the target resource allocation schemes can improve the resource utilization efficiency of the satellite system.

[0083] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the satellite system resource scheduling method based on multi-user satisfaction provided by any of the above-described method embodiments of the present invention.

[0084] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0085] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A satellite system resource scheduling method based on multi-user satisfaction, characterized in that, include: Real-time acquisition of satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range; The satellite system resource data, the service demand data of each user, and the channel status data are input into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme. Resource scheduling is performed for each user based on the target resource allocation scheme; The resource allocation operation includes: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, a current resource allocation scheme is generated. For each user, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. Calculate the variance of user satisfaction based on the current user satisfaction level of each user. When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

2. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 1, characterized in that, Based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, and the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range, the current user satisfaction is calculated, including: The current latency satisfaction is calculated based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel status data, the service requirement data, and the latency satisfaction index range in QoS satisfaction. Calculate the current bandwidth satisfaction based on the power allocation ratio and time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the service demand data, and the bandwidth satisfaction index range in the QoS satisfaction. Calculate the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel status data, and the bit error rate satisfaction index range in QoS satisfaction. The current user satisfaction is obtained by weighting the current latency satisfaction, the current bandwidth satisfaction, and the current bit error rate satisfaction.

3. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 2, characterized in that, The calculation of current latency satisfaction based on the time slot allocation ratio in the current resource allocation scheme, the satellite system resource data, the channel state data, the service requirement data, and the latency satisfaction index range in QoS satisfaction includes: Determine available spectrum bandwidth based on satellite system resource data; The available spectrum bandwidth is allocated according to the spectrum allocation ratio in the current resource allocation scheme to determine the spectrum bandwidth allocated to the user; The user channel transmission rate is calculated based on the user-allocated spectrum bandwidth and the user signal-to-noise ratio and channel attenuation coefficient in the channel state data. Obtain the business type and latency requirements from the business requirement data, and determine the amount of data for a single business based on the business type; Calculate the time required to transmit the single service data volume at the user channel transmission rate to obtain the data transmission time; The user transmission delay is obtained by superimposing the data transmission time and the preset system fixed delay. When the user's transmission latency is less than or equal to the latency requirement, the current latency satisfaction level is set to 1. When the user transmission latency is greater than the latency requirement, the difference between the user transmission latency and the latency requirement is determined as the consumed latency tolerance resource amount; the difference between the latency requirement and the maximum latency value in the latency satisfaction index range is determined as the total latency tolerance resource amount; the remaining latency tolerance resource percentage is calculated based on the consumed latency tolerance resource amount and the total latency tolerance resource amount, and the remaining latency tolerance resource percentage is determined as the current latency satisfaction level; The user channel transmission rate is calculated using the following formula: In the formula, Let i be the user channel transmission rate; Allocate spectrum bandwidth to user i's users; For user i, the signal-to-noise ratio; Let be the channel attenuation coefficient for user i.

4. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 3, characterized in that, The calculation of current bandwidth satisfaction based on the power allocation ratio, time slot allocation ratio, satellite system resource data, service demand data, and bandwidth satisfaction index range in QoS satisfaction includes: Available power is determined based on satellite system resource data, and bandwidth requirements are obtained from service demand data; The available power is allocated according to the power allocation ratio in the current resource allocation scheme to determine the user's allocated transmission power; Based on the user-allocated transmit power, the nonlinear gain of power with respect to effective bandwidth is calculated, and the transmit power gain correction coefficient is obtained. The user-allocated spectrum bandwidth is corrected based on the transmit power gain correction coefficient to obtain the actual bandwidth occupied by the user; When the actual bandwidth used by the user is greater than or equal to the bandwidth requirement, the current bandwidth satisfaction level is set to 1. When the actual bandwidth used by the user is less than the bandwidth requirement, the difference between the bandwidth requirement and the actual bandwidth used by the user is determined as the amount of bandwidth resources consumed; the total amount of bandwidth resources is calculated based on the maximum and minimum bandwidth values ​​in the bandwidth satisfaction index range; the proportion of remaining bandwidth resources is calculated based on the amount of bandwidth resources consumed and the total amount of bandwidth resources, and the proportion of remaining bandwidth resources is determined as the current bandwidth satisfaction.

5. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 4, characterized in that, The step of calculating the current bit error rate satisfaction based on the power allocation ratio in the current resource allocation scheme, the service demand data, the channel state data, and the bit error rate satisfaction index range in QoS satisfaction includes: Based on the channel attenuation coefficient in the channel state data, the effective received power of the user-allocated transmit power after channel attenuation is calculated. Based on the natural exponential function, the effective received power is converted into an exponential decay factor for the bit error rate as the effective received power changes. The user's actual bit error rate is obtained by correcting the preset benchmark bit error rate based on the exponential decay factor. Obtain the bit error rate requirement from the business requirements data; When the user's actual bit error rate is less than or equal to the bit error rate requirement, the current bit error rate satisfaction level is set to 1. When the user's actual bit error rate is greater than the bit error rate requirement, the difference between the user's actual bit error rate and the bit error rate requirement is determined as the amount of bit error rate resources consumed; the total amount of all bit error rate resources is calculated based on the maximum and minimum bit error rate values ​​within the bit error rate satisfaction index range; the remaining bit error rate resource ratio is calculated based on the consumed bit error rate resources and the total amount of all bit error rate resources, and the remaining bit error rate resource ratio is determined as the current bit error rate satisfaction level.

6. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 5, characterized in that, The resource scheduling for each user based on the target resource allocation scheme includes: In the target resource allocation scheme, the target spectrum allocation ratio, target power allocation ratio, and target time slot allocation ratio for each user are obtained respectively. For each user, the beamforming weights are adjusted according to the target spectrum allocation ratio and the target power allocation ratio. The codec type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio. Determine routing priority based on the latency requirements and target time slot allocation ratio; The target resources for each user are determined based on the target resource allocation scheme, and the target resources are allocated to the corresponding users based on each user's beamforming weight, codec type, and routing priority.

7. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 6, characterized in that, Adjusting beamforming weights based on target spectrum allocation ratios and target power allocation ratios includes: Determine the user's channel matrix based on channel state data; Based on the channel matrix, the target spectrum allocation ratio, the target power allocation ratio, and the available power, the beamforming weights for each user are calculated using the following formula: In the formula, Beamforming weights for user i; Let be the channel matrix for user i; Let be the transpose of the channel matrix of user i; Preset noise power; The target spectrum allocation ratio for user i; The target power allocation ratio for user i; This represents the available power.

8. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 7, characterized in that, The codec type is determined based on the service type, the bit error rate requirement, and the target power allocation ratio, including: When the business type is real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Turbo code; When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset real-time power allocation ratio, the code type is determined to be LDPC code. When the service type is real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset real-time power allocation ratio, the code type is determined to be Turbo code. When the business type is non-real-time data business and the bit error rate requirement is not greater than the preset bit error rate threshold, the code type is determined to be Polar code; When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is not less than the preset non-real-time power allocation ratio, the code-decoding type is determined to be LDPC code. When the service type is non-real-time data service, the bit error rate requirement is greater than the preset bit error rate threshold, and the target power allocation ratio is less than the preset non-real-time power allocation ratio, the encoding and decoding type is determined to be Polar code.

9. The satellite system resource scheduling method based on multi-user satisfaction as described in claim 8, characterized in that, Determining routing priorities based on the latency requirements and target time slot allocation ratio includes: When the target time slot allocation ratio is not less than the first time slot allocation ratio threshold and the latency requirement is less than the first latency threshold, the routing priority is determined as the first priority. When the target time slot allocation ratio meets the preset time slot allocation ratio condition and the latency requirement meets the preset latency condition, the routing priority is determined to be the second priority; wherein, the preset time slot allocation ratio condition is that the target time slot allocation ratio is not less than the second time slot allocation ratio threshold and is less than the first time slot allocation ratio threshold, and the preset latency condition is that the latency requirement is not less than the first latency threshold and is not greater than the second latency threshold. When the target time slot allocation ratio is less than the second time slot allocation ratio threshold and the latency requirement is greater than the second latency threshold, the routing priority is determined to be the third priority. Among them, the second time slot allocation ratio threshold is less than the first time slot allocation ratio threshold; the first delay threshold is less than the second delay threshold; the second priority is higher than the first priority and lower than the third priority.

10. A satellite system resource scheduling device based on multi-user satisfaction, characterized in that, include: Data acquisition module, resource allocation scheme generation module, and resource scheduling module; The data acquisition module is used to collect satellite system resource data, service demand data of several users, channel status data, and QoS satisfaction index range in real time. The resource allocation scheme generation module is used to input the satellite system resource data, the service demand data of each user and the channel status data into the adversarial neural network, so that the adversarial neural network repeatedly executes the resource allocation operation to generate the target resource allocation scheme. The resource scheduling module is used to schedule resources for each user based on the target resource allocation scheme. The resource allocation operation includes: Based on the satellite system resource data, the service demand data and channel status data of each user, and the current network parameters of the adversarial neural network, a current resource allocation scheme is generated. For each user, the current user satisfaction is calculated based on the spectrum allocation ratio, power allocation ratio, and time slot allocation ratio in the current resource allocation scheme, as well as the latency satisfaction index range, bandwidth satisfaction index range, and bit error rate satisfaction index range in the QoS satisfaction index range. Calculate the variance of user satisfaction based on the current user satisfaction level of each user. When the user satisfaction variance is less than the preset variance, the current resource allocation scheme is determined as the target resource allocation scheme; otherwise, the current network parameters of the adversarial neural network are adjusted according to the user satisfaction variance.

Images