Multi-track elastic communication method and system

Abstract

The invention relates to the technical field of wireless communication, in particular to a multi-orbit elastic communication method and system, and the method comprises the steps: monitoring the performance parameters of a multi-orbit satellite link in real time, and calculating the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function; dynamically adjusting a configuration strategy of the main backhaul link according to the comprehensive score and a reinforcement learning model; when a network fault or congestion is detected, activating an alternative backhaul path to complete flow redirection and breakpoint resume, and executing a corresponding disaster recovery strategy according to a fault level; and dynamically adjusting bandwidth allocation according to the authorization level of the user equipment. According to the invention, the elastic communication service with wide-area coverage, low time delay and high reliability can be realized through multi-orbit satellite cooperation, efficient disaster recovery switching and priority resource guarantee in a complex scene.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and particularly to a multi-orbit elastic communication method and system. Background Art

[0002] With the acceleration of the global digitalization process, the integration of 5G / 6G technologies and satellite communication has become a key direction for achieving seamless global coverage. In scenarios such as public safety, emergency rescue, and communication in remote areas, the ground network often fails to provide coverage due to natural disasters, geographical limitations, or human attacks, and there is an urgent need to rely on satellite communication to provide reliable emergency support.

[0003] Traditional satellite communication systems mostly use single geostationary orbit (GSO) satellites. Although they can achieve wide-area coverage, it is difficult to meet the requirements of low-latency services. Non-geostationary orbit (NGSO) satellites, although having the advantages of low latency and high throughput, cannot independently provide global seamless services due to their limited coverage area and the need for frequent satellite handovers. How to build an elastic communication system with both wide-area coverage and low-latency characteristics has become a technical difficulty that the industry urgently needs to solve. Existing solutions attempt to coordinate and schedule satellites and ground base stations, but only use satellites as an extension of a single path, without fully utilizing the synergistic complementarity of GSO / NGSO, nor solving the disaster tolerance problem when the ground core network fails. Although the hierarchical slicing and resource scheduling solutions perform unified management at the core network level, they lack disaster tolerance design for the failure of the ground core network in extreme scenarios and do not provide a differentiated priority guarantee mechanism for authorized user equipment (UE). In addition, in the face of the diverse access requirements of heterogeneous terminals such as drones, vehicle-mounted terminals, and Internet of Things devices, existing solutions are also difficult to achieve flexible resource scheduling and mobility management, resulting in low network resource utilization and the inability to guarantee the quality of critical services.

[0004] In summary, the existing technologies cannot achieve elastic communication services with wide-area coverage, low latency, and high reliability through multi-orbit satellite collaboration, efficient disaster tolerance handover, and priority resource guarantee in complex scenarios. Summary of the Invention

[0005] In view of this, the present invention provides a multi-orbit elastic communication method and system to solve the problem that the existing technologies cannot achieve elastic communication services with wide-area coverage, low latency, and high reliability through multi-orbit satellite collaboration, efficient disaster tolerance handover, and priority resource guarantee in complex scenarios.

[0006] In a first aspect, the present invention provides a multi-orbit elastic communication method, and the method includes:

[0007] Real-time monitoring of the performance parameters of the multi-orbit satellite link; the multi-orbit satellite link includes at least one of a geostationary orbit GSO satellite link and a non-geostationary orbit NGSO satellite link;

[0008] Calculate the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function;

[0009] Dynamically adjust the configuration strategy of the main backhaul link according to the comprehensive score and the reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching satellite links, and performing load diversion;

[0010] When a network fault or congestion is detected, activate an alternative backhaul path to complete traffic redirection and resume interrupted downloads, and execute corresponding disaster tolerance strategies according to the fault level; and dynamically adjust the bandwidth allocation according to the authorization level of the user device.

[0011] In an alternative embodiment, the performance parameters include available bandwidth, latency, packet loss rate, and coverage stability.

[0012] In an alternative embodiment, the preset scoring function is:

[0013]

[0014] where S i represents the comprehensive score of satellite link i, W i represents the available bandwidth of satellite link i, W max represents the maximum available bandwidth of satellite link i, T i represents the latency of satellite link i, T max represents the maximum latency, P i represents the packet loss rate of satellite link i, P max represents the maximum packet loss rate of each satellite link, O i represents the orbital coverage stability of satellite link i, and α, β, γ, δ represent the weight coefficients in cross-orbit satellite collaboration.

[0015] In an alternative embodiment, the reward function of the reinforcement learning model is:

[0016] R t = α·ThroughputGain - β·SwitchCost + γ·Qos Satisfaction ;

[0017] where R t represents the reward value, ThroughputGain represents the throughput gain caused by satellite link switching, SwitchCost represents the satellite link switching cost, QoS Satisfaction represents the service quality satisfaction, and α, β, γ represent the weight coefficients in cross-orbit satellite collaboration.

[0018] In an alternative embodiment, dynamically adjusting the configuration policy of the primary backhaul link according to the comprehensive score and the reinforcement learning model includes:

[0019] Inputting the performance parameters and the comprehensive score into the reinforcement learning model for training, and generating an orbit switching or load splitting instruction in combination with the prediction of the movement trajectory of the user equipment;

[0020] Dynamically adjusting the configuration policy of the primary backhaul link based on the orbit switching or load splitting instruction.

[0021] In an alternative embodiment, when detecting a network failure or congestion, activating an alternative backhaul path to complete traffic redirection and resume interrupted transmission, and executing corresponding disaster tolerance policies according to the failure level, including:

[0022] When detecting a failure of the terrestrial core network or the primary satellite link, activating an alternative backhaul path based on a hierarchical disaster tolerance architecture, performing traffic redirection and resume interrupted transmission through a multipath transmission protocol, and executing corresponding disaster tolerance policies according to the failure level; the alternative backhaul path includes at least one of the geostationary orbit (GSO) satellite link, the non-geostationary orbit (NGSO) satellite link, or an adjacent emergency base station.

[0023] In an alternative embodiment, the method further includes:

[0024] After the primary satellite link is restored, performing a progressive fallback test during a low traffic period to gradually migrate traffic to the original primary backhaul link.

[0025] In an alternative embodiment, the alternative backhaul path is obtained through a multi-objective optimization algorithm, and the objective function of the multi-objective optimization algorithm is:

[0026]

[0027] where W R represents the available bandwidth of the alternative backhaul path R, W max represents the maximum available bandwidth among the selected alternative backhaul paths, T R represents the delay of the alternative backhaul path R, T max represents the maximum delay, Cost(S) represents the comprehensive cost of the set S of alternative backhaul paths, Reliability(S) represents the combined availability when multiple alternative backhaul paths are in parallel, and ω1, ω2, ω3, ω4 represent the weight coefficients in network disaster tolerance management; represents a set of selected alternative backhaul paths {R1,…R n}.

[0028] In an alternative embodiment, dynamically adjusting the bandwidth allocation according to the authorization level of the user device includes:

[0029] According to the authorization level of the user device, allocate an independent network slice to the target high-priority user device, and dynamically adjust the bandwidth allocation.

[0030] In a second aspect, the present invention provides a multi-orbit elastic communication system, which is used to execute a multi-orbit elastic communication method as described above. The system includes:

[0031] A cross-orbit satellite cooperation module, which is used to monitor the performance parameters of the multi-orbit satellite link in real time; calculate the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function; dynamically adjust the configuration strategy of the main backhaul link according to the comprehensive score and the reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching the satellite link, and performing load diversion; the multi-orbit satellite link includes at least one of a geostationary orbit (GSO) satellite link and a non-geostationary orbit (NGSO) satellite link;

[0032] An alternative backhaul and network disaster recovery management module, which is used to activate an alternative backhaul path to complete traffic redirection and resume interrupted transmission when detecting a network failure or congestion, and execute corresponding disaster recovery strategies according to the failure level;

[0033] An authorized user device priority guarantee module, which is used to dynamically adjust the bandwidth allocation according to the authorization level of the user device when detecting a network failure or congestion.

[0034] In a third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute a multi-orbit elastic communication method according to the first aspect or any corresponding embodiment thereof.

[0035] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute a multi-orbit elastic communication method according to the first aspect or any corresponding embodiment thereof.

[0036] In a fifth aspect, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute a multi-orbit elastic communication method according to the first aspect or any corresponding embodiment thereof.

[0037] The technical solution provided by the present invention may include the following beneficial effects:

[0038] 1. The present invention combines the advantages of geosynchronous orbit (GSO) satellite links and non-geosynchronous orbit (NGSO) satellite links, which can not only ensure wide coverage but also provide low latency and high throughput, achieving comprehensive cross-orbit coordination.

[0039] 2. Through alternative backhaul paths and multi-redundant configurations, the present invention can quickly switch and recover in case of ground network failures or satellite failures, achieving strong resilience and disaster tolerance capabilities.

[0040] 3. According to the authorization level of user equipment, the present invention dynamically adjusts bandwidth allocation and adopts resource preemption scheduling to meet the high requirements for communication in public safety and emergency rescue.

[0041] 4. After the main satellite link is restored, the present invention performs progressive cut-back tests during low-traffic periods, gradually migrating traffic to the original main backhaul link. By adopting low-traffic period cut-back tests and progressive migration strategies, it smoothly cuts back after the main satellite link is restored, avoiding sudden network fluctuations and improving cut-back reliability.

[0042] 5. The present invention uses a multi-objective optimization algorithm to select alternative backhaul paths and combines reinforcement learning to adaptively adjust the optimal backhaul path, improving network resource utilization and backhaul stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0044] Figure 1 FIG. is a schematic structural diagram of a multi-orbit elastic communication system according to an embodiment of the present invention;

[0045] Figure 2 FIG. is a flowchart of a multi-orbit elastic communication method according to an embodiment of the present invention;

[0046] Figure 3 FIG. is a flowchart of another multi-orbit elastic communication method according to an embodiment of the present invention;

[0047] Figure 4 FIG. is a schematic flowchart of cross-orbit satellite coordination and dynamic backhaul switching according to an embodiment of the present invention;

[0048] Figure 5 FIG. is a schematic flowchart of satellite alternative backhaul access and network disaster tolerance according to an embodiment of the present invention;

[0049] Figure 6It is a schematic flowchart of an authorized UE priority guarantee strategy according to an embodiment of the present invention;

[0050] Figure 7 It is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Detailed implementation manners

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

[0052] It should be noted that the following explains each term involved in the embodiments of the present invention:

[0053] 1. Geostationary Orbit (GSO): It refers to a high-orbit satellite that is synchronized with the Earth's rotation period. It has a wide coverage range, but due to an orbital altitude of about 35,786 kilometers, there is usually a relatively large communication delay.

[0054] 2. Non-Geostationary Orbit (NGSO): It refers to a satellite system including low and medium orbits (LEO, MEO, etc.). It has a relatively low orbital altitude and delay, but a relatively limited coverage range, and frequent satellite handovers are required.

[0055] 3. User Equipment (UE): It refers to various devices such as mobile terminals, Internet of Things devices, emergency communication vehicle-mounted terminals, and unmanned aerial vehicles that can access a communication network.

[0056] 4. Quality of Service (QoS): It refers to the guarantee of the transmission quality of service data in a network, including indicators such as bandwidth, delay, packet loss rate, and jitter.

[0057] 5. Backhaul: It refers to the communication link between a base station or an access point (such as a satellite ground station) and the core network. In the present invention, it also includes satellite alternative backhaul for emergency or disaster recovery scenarios.

[0058] 6. Alternative backhaul: When the terrestrial core network or the main satellite link fails or is overloaded, an emergency link provided by other orbital satellites, neighboring stations, or temporary access points is used.

[0059] 7. Internet of Things (IoT): It refers to connecting various sensing devices to the cloud or server through a network to achieve the collection, analysis, and control of device data.

[0060] The present invention aims to provide an elastic operation mechanism based on multi-orbit satellites. In disaster scenarios or high-priority service requirements, through the cooperation of geostationary orbit (GSO) and non-geostationary orbit (NGSO) satellites, alternative backhaul, and priority guarantee for authorized UEs, a network service with strong resilience, high availability, and high security is achieved; and through a scheduling method for heterogeneous terminals, the network resource utilization rate and service quality are improved.

[0061] According to an embodiment of the present invention, an embodiment of a multi-orbit elastic communication system is provided. Figure 1 It is a schematic structural diagram of a multi-orbit elastic communication system according to an embodiment of the present invention. This system is used to execute a multi-orbit elastic communication method as shown in Figure 2 A multi-orbit elastic communication method is shown. This system includes:

[0062] A cross-orbit satellite cooperation module, which is used to monitor the performance parameters of the multi-orbit satellite link in real time; calculate the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function; and dynamically adjust the configuration strategy of the main backhaul link according to the comprehensive score and the reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching satellite links, and performing load diversion; the multi-orbit satellite link includes at least one of the geostationary orbit GSO satellite link and the non-geostationary orbit NGSO satellite link.

[0063] An alternative backhaul and network disaster tolerance management module, which is used to activate an alternative backhaul path to complete traffic redirection and resume interrupted transmission when detecting a network failure or congestion, and execute corresponding disaster tolerance strategies according to the failure level.

[0064] An authorized user equipment priority guarantee module, which is used to dynamically adjust the bandwidth allocation according to the authorization level of the user equipment when detecting a network failure or congestion.

[0065] Furthermore, this system includes three core modules: a cross-orbit satellite cooperation module, an alternative backhaul and network disaster tolerance management module, and an authorized user equipment (UE) priority guarantee module. Each module realizes the elastic scheduling, fault tolerance, and priority resource guarantee of the multi-orbit satellite network through data interaction and policy cooperation.

[0066] First, the cross-orbit satellite cooperation module is used to achieve dynamic cooperation between geostationary orbit (GSO) and non-geostationary orbit (NGSO) satellites and optimize the main backhaul link configuration. The cross-orbit satellite cooperation module can implement real-time performance monitoring functions, including real-time collection of performance parameters of GSO / NGSO satellite links and uploading them to the central control unit through satellite gateways, ground stations or edge nodes. Build and dynamically update the global coverage database, including satellite ephemeris, beam coverage, historical availability and other information, providing a data basis for subsequent decisions. After that, the cross-orbit satellite cooperation module also performs comprehensive score calculation and dynamic policy adjustment processing, including quantifying the comprehensive performance of each satellite link based on a preset scoring function to identify the optimal main backhaul candidate link; and combining with a reinforcement learning model (such as Q-Learning), taking the output of the scoring function as the state input, optimizing long-term decisions through a reward function, and generating configuration policies.

[0067] Secondly, the alternative backhaul and network disaster recovery management module can ensure service continuity through alternative backhaul in case of network failures or congestion. The alternative backhaul and network disaster recovery management module can implement fault detection and grading functions, including real-time monitoring of fault types (such as ground core network interruption, satellite link anomaly, local overload, etc.), and identifying the fault level through active detection (heartbeat packets) and passive monitoring (traffic anomaly analysis). After that, the alternative backhaul and network disaster recovery management module can also implement the alternative backhaul activation function, including selecting the optimal alternative backhaul based on a hierarchical disaster recovery architecture (centralized decision-making at the core layer + autonomous switching at the edge layer). Guide new traffic to take the backup path through flow table update, cache data using TCP / QUIC protocols to avoid transmission interruption caused by handover; support multi-path parallel transmission to enhance disaster resistance.

[0068] Furthermore, the authorized user equipment priority guarantee module provides differentiated resource guarantees for high-priority UEs (such as emergency rescue terminals, military and police equipment). The authorized user equipment priority guarantee module can implement UE grading and authentication and dynamic bandwidth adjustment functions. In case of network congestion or failure, increase the queue weight of authorized UEs through multi-queue scheduling and forcibly preempt the resources of ordinary UEs. Combine with network slicing technology to reserve a dedicated resource pool for authorized UEs and synchronize slice parameters during cross-orbit handover to ensure uninterrupted QoS.

[0069] In summary, in this embodiment, the cross-orbit satellite cooperation module combines the characteristics of geostationary orbit (GSO) and non-geostationary orbit (NGSO) satellites, and realizes dynamic switching and load sharing of the main backhaul through real-time monitoring and intelligent scoring (or reinforcement learning), taking into account wide-area coverage and low latency. The alternative backhaul and network disaster recovery management module automatically enables backup satellites or adjacent base stations for traffic redirection and resume from breakpoint when the ground core network or the main satellite link fails, ensuring service continuity; and can adopt a disaster recovery strategy with multiple orbits in parallel to improve the network's disaster resistance ability. The authorized user equipment priority guarantee module establishes a multi-queue or slicing mechanism for high-priority user equipment such as public safety and disaster emergency. When network congestion or failure occurs, resources (such as bandwidth, scheduling time slots, beam power) are preferentially allocated to authorized UEs to ensure the communication quality of their critical services.

[0070] In summary, the technical solution provided in this embodiment may include the following beneficial effects:

[0071] 1. This embodiment integrates the advantages of geostationary orbit (GSO) satellite links and non-geostationary orbit (NGSO) satellite links, which can not only ensure large-scale coverage but also provide low latency and high throughput, realizing comprehensive cross-orbit cooperation.

[0072] 2. Through alternative backhaul paths and multi-redundancy configurations, this embodiment can quickly switch and recover when the ground network fails or the satellite fails, realizing strong elasticity and disaster recovery capabilities.

[0073] 3. According to the authorization level of user equipment, this embodiment dynamically adjusts bandwidth allocation and adopts resource preemption scheduling to meet the high requirements for communication in public safety and emergency rescue.

[0074] 4. After the main satellite link is restored, this embodiment performs a progressive backhaul test during low-traffic periods, gradually migrating traffic to the original main backhaul link. By adopting a low-traffic period backhaul test and progressive migration strategy, it smoothly backhauls after the main satellite link is restored, avoiding sudden network fluctuations and improving the reliability of backhaul.

[0075] 5. This embodiment uses a multi-objective optimization algorithm to select alternative backhaul paths, combined with reinforcement learning, to adaptively adjust the optimal backhaul path, improving network resource utilization and backhaul stability.

[0076] According to an embodiment of the present invention, an embodiment of a multi-orbit elastic communication method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0077] In this embodiment, a multi-orbit elastic communication method is provided, which can be used in Figure 1 a multi-orbit elastic communication system shown in Figure 2 is a flowchart of a multi-orbit elastic communication method according to an embodiment of the present invention. As Figure 2 shown, the process includes the following steps:

[0078] Step S201, monitor the performance parameters of the multi-orbit satellite link in real time; the multi-orbit satellite link includes at least one of the geosynchronous orbit (GSO) satellite link and the non-geosynchronous orbit (NGSO) satellite link.

[0079] Further, in this embodiment, the performance parameters of the geosynchronous orbit (GSO) and non-geosynchronous orbit (NGSO) satellite links are continuously collected. The performance parameters include available bandwidth, delay, packet loss rate, and coverage stability; among them, the bandwidth refers to the available transmission rate of the current link; the delay refers to the round-trip time of data; the packet loss rate refers to the proportion of lost data packets during transmission; the coverage stability refers to the link reliability score calculated based on satellite ephemeris and historical availability data.

[0080] Step S202, calculate the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function.

[0081] Further, in this embodiment, a preset scoring function is used to quantitatively evaluate the performance of each satellite link. The preset scoring function combines available bandwidth, delay, packet loss rate, and coverage stability. That is to say, the preset scoring function comprehensively considers multiple performance parameters, assigns different weights to each parameter, and converts the performance parameters of each satellite link into a quantitative comprehensive score through weighted summation or other mathematical operation methods. The comprehensive score can intuitively reflect the overall performance and applicability of each satellite link at the current moment, providing a quantitative basis for subsequent link selection and resource allocation. A link with a high score represents better comprehensive performance and may be preferentially selected.

[0082] Step S203, dynamically adjust the configuration strategy of the main backhaul link according to the comprehensive score and the reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching satellite links, and performing load splitting.

[0083] Further, the reinforcement learning model is a reinforcement learning model based on Q-Learning. The state space of the reinforcement learning model includes link performance and system load, and the action space includes orbit switching or load shunting operations. In this embodiment, the comprehensive scoring and the reinforcement learning model are integrated. Based on historical data and continuously learned experience, the reinforcement learning model can predict the optimal primary backhaul link configuration strategy according to the current comprehensive score and link status information. The configuration strategy includes maintaining the current orbit, switching the satellite link, and performing load shunting. If the comprehensive score of the current link is good and the reinforcement learning model determines that no adjustment is required, the current orbit is maintained; if the scores of other satellite links are higher and it is beneficial after considering comprehensive factors such as switching costs, the satellite link is switched; when the link load is too high, load shunting is performed to transfer some traffic to other links to balance the link load and improve the overall transmission efficiency.

[0084] Step S204, when a network failure or congestion is detected, activate the alternative backhaul path to complete traffic redirection and resume interrupted transmission, and execute corresponding disaster recovery strategies according to the failure level; and dynamically adjust the bandwidth allocation according to the authorization level of the user equipment.

[0085] Further, once a network failure or congestion is detected, this embodiment immediately activates the alternative backhaul path. The alternative backhaul path can be other satellite links or ground backup links, etc. By redirecting the traffic to the alternative backhaul path, the continuity of data transmission is ensured, and the resume interrupted transmission technology is adopted to avoid data loss. At the same time, this embodiment executes corresponding levels of disaster recovery strategies according to the severity and impact range of the failure to minimize the impact of the failure on the communication system (for example, if the failure level is L1 mild, that is, the delay or packet loss rate slightly exceeds the threshold, only load shunting processing needs to be performed; if the failure level is L3 severe, that is, the ground core network is completely interrupted, standby satellites or emergency base stations need to be enabled, and the multi-orbit parallel strategy operation can also be started). In addition, when a network failure or congestion is detected, this embodiment also dynamically adjusts the bandwidth allocation according to the authorization level of the user equipment. For high-priority authorized user equipment, such as emergency communication equipment and key business equipment, more bandwidth resources are preferentially allocated to ensure that their communication quality is not affected; while for ordinary user equipment, the bandwidth may be appropriately restricted when necessary to ensure the normal operation of key services.

[0086] In summary, the technical solution provided by this embodiment may include the following beneficial effects:

[0087] 1. This embodiment combines the advantages of geostationary orbit (GSO) satellite links and non-geostationary orbit (NGSO) satellite links, which can not only ensure wide coverage but also provide low latency and high throughput, achieving comprehensive cross-orbit collaboration.

[0088] 2. In this embodiment, through the alternative return path and multi-redundancy configuration, it can be quickly switched and restored in case of ground network failure or satellite failure, achieving strong elasticity and disaster tolerance capabilities.

[0089] 3. In this embodiment, according to the authorization level of the user equipment, the bandwidth allocation is dynamically adjusted, and resource preemption scheduling is adopted to meet the high requirements for communication in public safety and emergency rescue.

[0090] 4. After the main satellite link is restored, in the low-traffic period, a progressive backhaul test is performed, and the traffic is gradually migrated to the original main return link. By adopting the low-traffic period backhaul test and progressive migration strategy, smooth backhaul is achieved after the main satellite link is restored, avoiding sudden network fluctuations and improving the reliability of backhaul.

[0091] 5. In this embodiment, a multi-objective optimization algorithm is used to select the alternative return path, combined with reinforcement learning, to adaptively adjust the optimal return path, improving the network resource utilization rate and the stability of the return path.

[0092] In this embodiment, a multi-orbit elastic communication method is provided, which can be used in Figure 1 a multi-orbit elastic communication system shown in Figure 3 is a flowchart of another multi-orbit elastic communication method according to an embodiment of the present invention, as shown in Figure 3 shown, and the process includes the following steps:

[0093] Step S301, monitor the performance parameters of the multi-orbit satellite link in real time, and construct a dynamically updated global coverage database based on the performance parameters; the multi-orbit satellite link includes at least one of the geostationary orbit GSO satellite link and the non-geostationary orbit NGSO satellite link. The performance parameters include available bandwidth, delay, packet loss rate, and coverage stability.

[0094] Furthermore, in this embodiment, the performance parameters of the multi-orbit satellite link are continuously collected, and these links include at least one of the GSO or NGSO satellite links to ensure that the system can timely obtain the latest status information of the link. And, based on the collected performance parameters, a global coverage database is constructed and dynamically updated. The global coverage database can provide real-time information about the coverage range and quality of the satellite link, providing data support for subsequent link selection and switching decisions. The performance parameters include key indicators such as available bandwidth, delay, packet loss rate, and coverage stability, which directly reflect the communication ability and reliability of the satellite link.

[0095] Step S302, calculate the comprehensive score of each satellite link of the multi-orbit satellite link based on a preset scoring function.

[0096] In an optional implementation manner, the preset scoring function is:

[0097]

[0098] Among them, S i represents the comprehensive score of satellite link i, and W i represents the available bandwidth of satellite link i, and W max represents the maximum available bandwidth of the satellite link, and T i represents the delay of satellite link i, and T max represents the maximum delay, and P i represents the packet loss rate of satellite link i, and P max represents the maximum packet loss rate of each satellite link, and O i represents the orbital coverage stability of satellite link i (such as remaining coverage time, constellation operation status), and α, β, γ, δ represent the weight coefficients in cross-orbit satellite cooperation.

[0099] Furthermore, in this embodiment, a pre-defined scoring function is used. This scoring function comprehensively considers various performance parameters of the satellite link and assigns corresponding weights to each parameter. By substituting the values of each performance parameter into the scoring function, the comprehensive score of each satellite link can be calculated. The comprehensive score reflects the overall performance of the satellite link at the current moment and can be used to compare the advantages and disadvantages of different links.

[0100] Step S303: Input the performance parameter and the comprehensive score into the reinforcement learning model for training, and generate an orbit switching or load splitting instruction in combination with the prediction of the mobile trajectory of the user equipment; based on the orbit switching or load splitting instruction, dynamically adjust the configuration strategy of the primary backhaul link; the configuration strategy includes maintaining the current orbit, switching the satellite link, and performing load splitting.

[0101] In an alternative embodiment, the reward function of the reinforcement learning model is:

[0102] R t =α·ThroughputGain - β·SwitchCost + γ·Qos Satisfaction ;

[0103] Among them, R t represents the reward value, ThroughputGain represents the throughput gain caused by satellite link switching, SwitchCost represents the satellite link switching cost, that is, the interruption time or packet loss amount caused by switching, and QoS Satisfaction represents the service quality satisfaction, that is, whether services such as authorized UEs are guaranteed (such as the delay being less than the threshold), and α, β, γ represent the weight coefficients in cross-orbit satellite cooperation.

[0104] Furthermore, in this embodiment, the monitored performance parameters and the calculated comprehensive score are input into the reinforcement learning model for training. The reinforcement learning model uses this data to learn the optimal link configuration strategy. Considering that the user equipment may move between different satellite coverage areas, therefore, this embodiment also predicts the movement trajectory of the user equipment to make decisions on link switching or load diversion in advance. Then, according to the output result of the reinforcement learning model, orbit switching or load diversion instructions are generated, and based on these instructions, the configuration strategy of the main backhaul link is dynamically adjusted to optimize the network performance.

[0105] Furthermore, the following summarizes steps S301 to S303: Please refer to Figure 4 the schematic flow diagram of cross-orbit satellite cooperation and dynamic backhaul switching shown in. In this embodiment, the adaptive multi-score fusion technology and the orbit switching and terminal mobility linkage technology are first adopted in the process of cross-orbit satellite cooperation and dynamic backhaul switching. The adaptive multi-score fusion technology allows the combination of Q-Learning (reinforcement learning model) and the scoring function; when the system detects a sudden situation (such as extreme weather), it can temporarily rely on a fixed threshold judgment with a higher priority to avoid excessive uncertainty of the pure reinforcement learning strategy in rare scenarios. The orbit switching and terminal mobility linkage technology enables if the ground terminal (UE) also moves at high speed (such as in a vehicle or an unmanned aerial vehicle), the position of the UE can be predicted synchronously during cross-orbit switching, and the satellite beam with more stable coverage can be pre-switched to reduce the cost of secondary switching.

[0106] When conducting multi-orbit depth monitoring, the realization of coverage map construction involves GSO satellites, NGSO satellites, and a coverage database. Among them, GSO satellites are geostationary orbit satellites, which have the characteristic of large-scale global coverage due to their orbital height (~35,786 km), but have a relatively high latency (usually >240 ms); NGSO satellites are low-earth orbit (LEO, 1200 km) or medium-earth orbit (MEO, 20000 km) satellites, which have the advantages of low latency and high throughput, but the coverage area is relatively limited, and the satellites need to continuously move and switch coverage; the coverage database refers to the cloud / ground control center, which pre-loads all satellite orbit ephemeris and antenna beam information in advance to generate a global or regional coverage map (including signal strength, possible interference, historical availability, etc.), and dynamically updates it according to actual observations.

[0107] When conducting real-time telemetry and feedback, the satellite gateway (GW) and the ground station (GS) regularly report key indicators such as the signal strength (RSSI), carrier-to-interference ratio (C / N0), and link round-trip delay (RTT) they observe. This embodiment also deploys edge monitoring nodes at each base station or the upper-layer cloud, which can monitor the satellite channel performance in two ways: probe and passive actual traffic flow, and give a real-time alarm when an anomaly occurs.

[0108] When performing intelligent cross-orbit scoring and switching, this implementation uses the above configurable and scenario-adjustable preset scoring function. When the comprehensive score S calculated by the preset scoring function i is lower than a certain threshold for a long time, or shows a sudden cliff-like drop, the system will determine that this link is no longer suitable as the main backhaul and needs to perform a switching / shunting operation. And the reinforcement learning model of this embodiment adopts the Q-Learning reinforcement learning mechanism and is a reinforcement learning model based on Q-Learning.

[0109] For the reinforcement learning model, its state is:

[0110] S t ={W g ,T g ,P g ,W g ,W n ,T n ,P[[ID=z5]] n ,…,UE Priority}, which can be extended to multiple NGSO / GSO combinations.

[0111] Its action is: {A1 = maintain GSO, A = switch to NGSO1, A3 = load shunting, …}.

[0112] Its reward function (Reward) is as shown above.

[0113] The learning process of this reinforcement learning model can be:

[0114] Initially, the Q-table or policy network has not been fully trained, and the system can first use the scoring function for conservative switching; as time goes by and real network operation data accumulates, the RL model continuously updates the Q value or neural network weights; after convergence, it can still have better switching decisions in a changing environment (different weather, interference, satellite coverage). Among them, as described above, the state space of this reinforcement learning model is S t ={link performance, system load, number of authorized UEs, …}; the action space is {maintain the current orbit}, {switch to GSO}, {switch to NGSO}, {load shunting}. The reward function comprehensively considers high throughput, low latency, switching overhead (to avoid overly frequent switching), satisfaction of authorized UEs, etc., and assigns different weights. In the training stage, it can perform real-time online learning in a simulation environment or edge nodes, and long-term convergence obtains a set of policies suitable for a wide range of environments. In the inference stage, according to the current state, it can quickly make a switching decision by looking up the table or inferring through a neural network.

[0115] Such as Figure 4As shown in the figure, in this embodiment, scoring, decision-making, and execution switching are performed among multiple satellite orbits, and the process includes:

[0116] A(Start): The system starts a loop for cross-orbit monitoring. It can be called periodically (e.g., every 1 second or 5 seconds), or triggered by an event (when the satellite beam detects abnormal performance).

[0117] B(Obtain GSO&NGSO real-time performance data): Collect the available bandwidth W of each link of GSO and NGSO from the satellite gateway, ground station, or edge node i , delay T i , packet loss rate P i and coverage stability / remaining time O i . These data are uploaded to the disaster recovery control center or cloud scheduling platform through the interface.

[0118] C(Calculate link score Si / RL inference): Input the parameters into the reinforcement learning model (RL) according to the scoring function, and let it output a decision of "optimal orbit selection" or "whether to maintain / switch" based on the Q value or policy network obtained through long-term learning.

[0119] D(Scoring / policy result): If the obtained preferred orbit is the same as the current orbit, that is, the system state is good, there is no need to perform a switch; if the preferred orbit is different from the current orbit, it means that the current link performance has decreased significantly or other links are significantly better.

[0120] E(Maintain the status quo): Keep the current connection state unchanged to avoid unnecessary switching overhead or interruption to the UE.

[0121] F(Prepare for cross-orbit handover / check UE status): Collect the number of currently connected UEs, service types, and authorized UE priority information, and estimate the interruption duration, traffic loss, or additional signaling overhead caused by the handover. If the total number of UEs is too large and the service is extremely sensitive to handover (such as the URLLC scenario), the handover may be temporarily not performed or part of the traffic may be diverted first.

[0122] G(Whether the handover conditions are met): Include multi-faceted judgments such as the benefit-cost ratio (bandwidth gain after handover / interruption time caused by handover), requirements of authorized UEs (guarantee critical services first), etc. If the conditions are met, go to H; otherwise, return to E and maintain the existing state.

[0123] H(Perform handover / traffic splitting + send update instructions): Send instructions to network devices (such as satellite gateways, routers), allocate new orbit resources, and establish new connections; if it is traffic splitting, transfer part of the traffic to the preferred orbit to relieve congestion on the current orbit.

[0124] I (Update routing table + notify ground core / edge): After successful handover, the new link ID or IP path needs to be registered in the routing table to prevent subsequent traffic from still being sent to the old orbit; at the same time, notify the ground core and edge nodes to ensure consistent data uplink and downlink paths.

[0125] J (End): The cross-orbit collaborative decision-making process for this round ends, waiting for the next round of monitoring or event trigger.

[0126] Step S304, when a network failure or congestion is detected, activate the alternative backhaul path to complete traffic redirection and resume interrupted transmission, and execute corresponding disaster tolerance strategies according to the failure level.

[0127] In an optional implementation, this step S304 includes:

[0128] When a ground core network or main satellite link failure is detected, activate the alternative backhaul path based on the hierarchical disaster tolerance architecture, perform traffic redirection and resume interrupted transmission through the multi-path transmission protocol, and execute corresponding disaster tolerance strategies according to the failure level; the alternative backhaul path includes at least one of the geostationary orbit GSO satellite link, the non-geostationary orbit NGSO satellite link, or the adjacent emergency base station.

[0129] In an optional implementation, the alternative backhaul path is obtained through a multi-objective optimization algorithm, and the objective function of the multi-objective optimization algorithm is:

[0130]

[0131] where W R represents the available bandwidth of the alternative backhaul path R, W max represents the maximum available bandwidth among the selected alternative backhaul paths, T R represents the delay of the alternative backhaul path R, T max represents the maximum delay, Cost(S) represents the comprehensive cost of the alternative backhaul path set S, Reliability(S) represents the joint availability when multiple alternative backhaul paths are in parallel (if one fails, there are still others available), and ω1, ω2, ω3, ω4 represent the weight coefficients in network disaster tolerance management; represents a set of selected alternative backhaul paths {R1,…R n}.

[0132] Furthermore, in this embodiment, the network status is monitored in real time. Once a network failure or congestion is detected, the emergency response mechanism is immediately activated to activate a pre-planned alternative backhaul path, which may include other satellite links or ground emergency base stations, etc. Through the multi-path transmission protocol, this embodiment redirects the traffic to the alternative backhaul path to ensure the continuity of data transmission. In addition, this embodiment also executes disaster recovery strategies at corresponding levels according to the severity and impact scope of the failure to minimize the impact of the failure on the communication system.

[0133] Step S305, after the main satellite link is restored, a progressive cut-back test is performed during a low-traffic period to gradually migrate the traffic to the original main backhaul link.

[0134] Furthermore, after detecting that the main satellite link failure is repaired or restored to normal, this embodiment selects to perform a progressive cut-back test during a low-traffic period to reduce the impact on user services, and gradually migrates the traffic from the alternative backhaul path back to the original main backhaul link to ensure the smoothness and stability of the link switch.

[0135] Furthermore, the following summarizes steps S304 to S305: Please refer to Figure 5 the schematic flowchart of satellite alternative backhaul access and network disaster recovery shown. In the process of satellite alternative backhaul access and network disaster recovery, this embodiment first adopts hierarchical disaster recovery architecture technology, fault detection and traffic redirection technology, and multi-orbit + multi-base station backup parallel technology.

[0136] Among them, for the hierarchical disaster recovery architecture technology, its core layer is the disaster recovery control center, which adopts a centralized brain to master all main backhaul, satellite status, and candidate backup link information, and receives fault reports from edge nodes to quickly decide whether to activate the alternative backhaul or only perform partial traffic diversion; this hierarchical disaster recovery architecture also provides a fault classification strategy, and for light faults (L1), only fine-tuning is performed when the QoS deviation is not large; for major faults (L3), all critical services are immediately transferred. The edge layer of the hierarchical disaster recovery architecture is the local disaster recovery unit, which is deployed on the satellite ground station or gateway (GW) side and has a fast fault detection function. If it loses contact with the core center, it can also locally determine whether to enable the emergency channel to ensure autonomy. This edge layer adopts a distributed routing table or blockchain-based storage to avoid the failure of the entire network caused by a single point of failure of the core node.

[0137] For the fault detection and traffic redirection technology, the fault types detected in this embodiment when network faults or congestion occur include ground core interruption, satellite link anomaly, and local overload. Ground core interruption means that the optical fiber or microwave backbone line is cut off by natural disasters / attacks, resulting in the central node being unreachable. Satellite link anomaly includes satellite's own faults, space interference, antenna damage, etc.; if the primary GSO or NGSO fails severely, a backup satellite is needed. Local overload means that although the link is not completely disconnected, the congestion is severe, and traffic needs to be actively diverted to a more idle alternative path. Redirection and resume from breakpoint involve flow table reset (i.e., the above-mentioned traffic redirection) and TCP / QUIC resume from breakpoint operations (i.e., the above-mentioned resume from breakpoint). The operation of traffic redirection includes updating the forwarding tables of edge nodes and the core network at the routing layer, so that the traffic of new or unfinished sessions all takes the backup return path. The operation of resume from breakpoint includes adopting a caching and resume from breakpoint mechanism at the link layer or transport layer to reduce a large number of retransmissions caused by interruptions.

[0138] For the multi-orbit + multi-base station backup parallel technology, in extreme cases, this embodiment enables both GSO return path and NGSO return path simultaneously, and can even combine neighboring stations or emergency base stations (balloon stations, UAV relays) for parallel redundant transmission. In addition, this embodiment also combines the MP-TCP (Multi-Path TCP) or MPTCP-like protocol to let the same transmission session be fragmented and sent among multiple paths to improve the data reach rate.

[0139] This embodiment obtains the alternative return path through a multi-objective optimization algorithm. The objective function of this multi-objective optimization algorithm is as shown above. In the backup resource library, there may be multiple available satellites / base stations (multiple return paths of satellites or ground base stations), denoted as {R1,…,R n}. If a group of return paths needs to be selected for parallel transmission simultaneously, an optimal solution needs to be found in the combination space P({R1,…,R n}). The objective function of this multi-objective optimization algorithm aims to balance bandwidth, delay, cost, and reliability to select the optimal set of return paths.

[0140] This embodiment adopts an improved algorithm process. This algorithm is used to search for the optimal set of return paths S in the combination solution space P({R2,…,R n}). First, this embodiment initializes the candidate paths, collects all possible satellite or neighboring station return path information, and obtains the candidate set:

[0141] {R1,R2…,R n};

[0142] After that, this embodiment collects index data such as bandwidth, delay, path cost, and reliability, and conducts evaluation and scoring, and calculates the above-mentioned multi-objective scores for each path:

[0143]

[0144] Cost(R) includes the bandwidth lease cost, power consumption, and the possible security risk score of using this backhaul. If multiple paths need to be selected, calculate the overall score of different path combinations of SSS:

[0145] S core (S) = ∑ R∈S S core (R);

[0146] Quantify the comprehensive score of each backhaul path for screening in the next optimization search.

[0147] This embodiment also uses heuristic search. If parallel multi-paths are required, the ant colony algorithm or genetic algorithm can be used to search for the optimal subset in the combination space; the simple "greedy + local search" method can also be used to quickly approach the optimal solution. It includes: first select the single path R with the highest score * ; gradually add additional paths R i , if the overall score improves after combination, keep it; if there is no further improvement, stop the search; after selecting the path, make a decision and send a handover instruction and breakpoint resume configuration to the gateway and router.

[0148] As Figure 5 shown, the process of satellite alternative backhaul access and network disaster tolerance includes:

[0149] A(Start): Execute the start process;

[0150] B(Disaster tolerance control center monitors network health): Combine active detection (timed Ping / Traceroute / heartbeat) and passive monitoring (analyze the packet loss rate and delay fluctuation of actual traffic). Once a certain threshold (packet loss > 30% or delay > 500ms, etc.) is triggered, enter the C judgment link.

[0151] C(Is it a serious fault?): If it does not reach the serious level (for example, the packet loss is slightly high, but there is no disconnection), large-scale disaster tolerance can be temporarily not executed, and only fine-tuning is done; if it is confirmed to be a major fault (L3 level, such as the complete disconnection of the ground core network), then transfer to E.

[0152] E(Judge the fault level): According to the fault manifestations (disconnection ratio, duration, recoverable probability), it is divided into L1 (mild), L2 (moderate), L3 (severe); different levels correspond to different actions: L1 can simply perform load sharing, and L3 must immediately enable the standby backhaul.

[0153] F(Query backup resource repository): During system initialization, resources such as GSO backups, NGSO backups, neighboring base stations, and emergency vehicle-mounted base stations and their statuses (bandwidth, latency, cost, etc.) are registered. If some backup resources are unavailable (for example, the GSO backup fails due to weather reasons), other feasible paths need to be selected.

[0154] G(Choose the optimal or parallel backhaul): Use multi-objective functions or heuristic algorithms (genetic algorithm, ant colony algorithm) to find the optimal single backhaul or parallel multiple backhauls; if the disaster is very serious, multiple backhauls can be transmitted concurrently, and technologies such as MP-TCP can be combined to improve disaster tolerance capabilities.

[0155] H(Breakpoint resume & traffic redirection + notify GW): Start breakpoint resume for ongoing traffic (such as video sessions, large file transfers) to ensure that there is no need to start over after the handover; the routing layer updates the forwarding table to direct subsequent traffic to the new backhaul; notify each satellite ground station / edge node to synchronously switch to the new path.

[0156] I(Has the terrestrial network recovered?): Continuously monitor the repair status of the terrestrial core or main backhaul link (manual or automatic detection); if it has not been repaired yet, enter J and continue to use the alternative backhaul; if it has been repaired, try to switch back (see K).

[0157] K(Low-traffic period backhaul switch test): To reduce the impact on UE services, the backhaul switch test is usually carried out at night or during low-load periods; after confirming that the main backhaul has recovered stably, gradually migrate the traffic back to normal.

[0158] D, J, and L are all nodes leading to the next round of loop monitoring, and the entire process continues to execute in the background until the disaster is completely over or the system exits.

[0159] Step S306: According to the authorization level of the user equipment, allocate an independent network slice for the target high-priority user equipment and dynamically adjust the bandwidth allocation.

[0160] Furthermore, in this embodiment, high-priority user equipment is identified according to the authorization level of the user equipment. An independent network slice is allocated for the target high-priority user equipment to ensure that these users can obtain exclusive network resources. According to the network conditions and user requirements, the bandwidth allocation is dynamically adjusted to meet the communication needs of the target high-priority user equipment and ensure the smooth progress of its critical services.

[0161] Furthermore, the following summarizes step S306: Please refer to Figure 6Schematic diagram of the process of the authorized UE priority guarantee strategy. In this embodiment, during the process of the authorized UE priority guarantee strategy, UE classification and encryption identification technology, multi-queue priority scheduling and slice reservation technology, dynamic power and beamforming priority technology, and priority scheduling algorithms are first adopted. Among them, for the UE classification and encryption identification technology, the authorized UE refers to a terminal with clear high-priority requirements and may carry a specific SIM / USIM / security certificate. The ordinary UE refers to the general public users, enterprise terminals, etc. If the system requirements permit, more levels can also be divided for commercial payment differentiation, such as VIP / paying premium UE (optional). The data link of the authorized UE can default to end-to-end encryption (such as IPSec, TLS1.3 or higher), multi-factor authentication, and separate sharding or dedicated bearers at the network slice level. In the event of a disaster scenario, cyber warfare, or extreme attack, it can be upgraded to a higher level of encryption and switched to the "secure satellite beam" (if this mechanism is reserved) to avoid eavesdropping or interference.

[0162] For the multi-queue priority scheduling and slice reservation technology, this embodiment performs multi-queue priority queuing, using methods such as PQ (Priority Queuing), WFQ (Weighted Fair Queuing), or DRR (Deficit Round Robin); the scheduling priority of the authorized UE queue is the highest, and the ordinary UE data is only scheduled when it is idle or has a low load. When the network is congested or there is a link failure, the bandwidth of the ordinary UE can be restricted to yield resources to the authorized UE. Moreover, this embodiment also performs cross-orbit slice reservation processing. The network slice controller uniformly manages the resource pools of all orbits on the ground / cloud and globally reserves the "authorized UE slice". For example, set the minimum guaranteed bandwidth BW min and the maximum delay T max . Regardless of whether the UE is connected to GSO, NGSO, or an alternative backhaul, the slice information can be synchronized to the corresponding satellite gateway and fulfilled.

[0163] For the dynamic power and beamforming priority technology, in the satellite physical layer of this embodiment, if it is detected that the traffic volume of the critical authorized UE is large and the satellite resources are tight, the transmit power of this beam can be dynamically increased or beamforming can be performed to reduce the coverage range but increase the signal strength, enabling the authorized UE to enjoy better channel quality.

[0164] For the priority scheduling algorithm, this embodiment uses weighted fair queue or priority queue for scheduling. The weighted fair queue (WFQ, Weighted Fair Queuing) is a strategy commonly used for network traffic scheduling. By assigning different weights to different types of traffic, it ensures that high-priority data streams obtain more stable bandwidth guarantees.

[0165] Weighted Fair Queuing (WFQ) can be calculated by the following formula:

[0166]

[0167] Where, W i represents the weight of queue i. The larger the weight value of queue i, the higher the proportion of bandwidth Rate i obtained by this queue; represents the sum of the weights of all queues, and R total represents the total available bandwidth.

[0168] The Authorized UE Queue can be given a higher weight W auth to ensure that its traffic is processed first. The weighted fair queue in this embodiment can ensure the bandwidth guarantee for authorized UEs. By increasing the weight W auth , authorized UEs can obtain a larger bandwidth share, and the communication quality can still be guaranteed when resources are scarce or the network is congested. When congestion occurs, W auth can be increased in real time to ensure the stability of critical services. The queue weight of non-authorized UEs can be appropriately reduced to free up resources and achieve dynamic adjustment of bandwidth allocation.

[0169] For the priority queue, the Priority Queue (PQ) is an extreme queue scheduling strategy that processes data packets in the order of queue priorities.

[0170] Authorized UE traffic → highest priority queue PQ auth : If this queue is not empty, the scheduler will always give priority to processing the data packets in this queue until the queue is empty, and then it will consider processing queues with lower priorities. This can ensure that the tasks of authorized UEs (such as emergency video calls, command and control traffic) are not affected by low-priority traffic.

[0171] Ordinary UE traffic → ordinary queue PQ normal : Only when the authorized UE queue is empty, the scheduler will process the traffic of ordinary UEs. It is applicable to scenarios such as emergency communication, private networks, and military networks to ensure unconditional priority transmission of critical task traffic. This embodiment can also implement a dynamic power / beam scheduling function, perform more refined scheduling at the satellite physical layer or MAC layer. If the signal of an authorized UE is weak but the service is urgent, the beam power will be temporarily increased or beamforming will be used to focus the coverage to ensure its SNR, and negotiate the power allocation with the ground gateway to avoid causing serious interference to UEs in other areas.

[0172] As Figure 6 shown, the process of the authorized UE priority guarantee strategy includes:

[0173] A(Start): The UE starts to access the network, and the system enters the authorization identification phase.

[0174] B(UE Access / AAA Authentication): Check the UE's SIM / USIM or security certificate to determine whether it has an authorized identity (such as a rescue agency); and corresponding encryption or security policies can be loaded.

[0175] C(Is the UE authorized?): If not on the authorized list, enter the normal user process D; if so, enter E.

[0176] D(Allocate Normal Slice / QoS): The system allocates a normal priority to the normal UE (the available WFQ queue weight = 1, etc.), and does not start special encryption or resource reservation.

[0177] E(Allocate High-Level Slice / QoS): Create or map to a high-priority slice in the network slice controller; this slice usually has the lowest bandwidth guarantee and the maximum latency limit; perform end-to-end encryption configuration on the UE (such as AES-256 or IPSec, etc.).

[0178] F(Daily Scheduling): When there is no congestion or failure, both normal UEs and authorized UEs share normal resources. The authorized UE queue generally does not preempt actively because of less load or the scheduler not triggering speed limit.

[0179] G(Is congestion or failure detected?): The monitoring module or the disaster recovery control center discovers that the satellite or ground network is congested or has a sudden failure, and it is necessary to check whether the QoS of the authorized UE is seriously affected.

[0180] I(Enter the Priority Preemption Process): At this time, increase the priority weight of the authorized UE queue, limit or reduce the bandwidth of the normal queue; the traffic of the normal UE can be migrated to the sub-optimal beam / orbit; if the resources are still insufficient, the priority of the authorized UE queue can be continuously increased until the minimum requirements are met.

[0181] J(Reduce the Bandwidth of the Normal Queue / Switch Orbit): Send an instruction through the scheduler to reduce the queuing weight of the normal UE or directly switch them to an orbit with lower load (such as the standby NGSO / GSO) to release resources for the authorized UE.

[0182] K(Does the performance of the authorized UE meet the standard?): Judge whether the authorized UE meets the minimum bandwidth (BW min ) and the maximum latency (T max ); if not, go back to I to continue increasing the preemption intensity; if so, it means that the priority guarantee has reached the target.

[0183] L (Completion of Priority Guarantee / Update of Slicing & Scheduling Table): The preemption process ends, and the system records the current priority configuration status, and maintains the resource allocation subsequently or gradually restores the normal UE bandwidth after load mitigation.

[0184] M (End): The current round of priority guarantee operation for authorized UEs ends, waiting for the next monitoring or event trigger.

[0185] In summary, this embodiment combines AI / ML with a scoring function to dynamically optimize GSO / NGSO orbit switching in extreme environments, improve communication stability, and achieve intelligent cross-orbit satellite switching, that is, integrate the advantages of GSO and NGSO to achieve comprehensive cross-orbit coordination, which can not only ensure wide coverage but also provide low latency and high throughput; through the global satellite coverage database and edge telemetry, the satellite link status is updated in real time to improve the switching accuracy and achieve multi-orbit in-depth monitoring. By integrating parameters such as bandwidth, latency, and packet loss rate, Q-Learning is used to optimize cross-orbit decisions, reduce unnecessary handovers, and achieve intelligent scoring and reinforcement learning optimization; based on the L1-L3 hierarchical strategy, GSO / NGSO / emergency base station backhaul is dynamically enabled during a failure to ensure uninterrupted services and achieve a multi-level disaster tolerance and alternative backhaul mechanism, that is, through alternative backhaul and multi-redundancy configuration, it can quickly switch and recover in case of ground network failure or satellite failure, achieving strong resilience and disaster tolerance capabilities; for authorized UEs such as emergency rescue and private networks, QoS and beam resources are dynamically allocated to ensure stable communication for high-priority services, that is, resource preemption scheduling is adopted to meet the high requirements of public safety and emergency rescue for communication and achieve priority guarantee for authorized UEs; combined with the WFQ / PQ scheduling strategy, it is ensured that authorized UEs can obtain priority bandwidth and low-latency services in a high-load environment, achieving a priority queue and dynamic regulation; based on indicators such as bandwidth, latency, cost, and reliability, a multi-objective optimization algorithm is used to select the optimal backhaul path to achieve intelligent backhaul path optimization with multi-objective optimization, that is, a multi-objective optimization algorithm (bandwidth, latency, cost, reliability) is adopted, combined with genetic algorithms, ant colony algorithms, or reinforcement learning, to adaptively adjust the optimal backhaul path, improve network resource utilization and backhaul stability; through low-traffic period testing and progressive backhaul, the service is smoothly migrated after the main link is restored, avoiding network jitter or packet loss, and achieving optimization of the disaster tolerance backhaul strategy, that is, low-traffic period backhaul testing and progressive migration strategy are adopted to smoothly backhaul after the main link is restored, avoiding sudden network fluctuations, improving the reliability of backhaul, and achieving intelligent backhaul optimization and smooth restoration of the main link.

[0186] In addition, this embodiment can achieve cross-orbit switching and load balancing through reinforcement learning, dynamic programming, or rule-based algorithms. If complex model training is not allowed in the actual environment, a fixed threshold can also be used to achieve it; multiple encryption protocols or blockchain distributed storage can be used to implement security reinforcement technology, or it can be replaced with other distributed databases or lighter security verification mechanisms; in addition to satellites, this embodiment can also be connected to temporary base stations, drone relays, or balloon base stations, etc. as alternative backhaul access points (only meeting the reliability requirements of alternative backhaul is sufficient).

[0187] In summary, the technical solution provided by this embodiment may include the following beneficial effects:

[0188] 1. This embodiment combines the advantages of geostationary orbit (GSO) satellite links and non-geostationary orbit (NGSO) satellite links, which can not only ensure wide coverage but also provide low latency and high throughput, achieving comprehensive cross-orbit coordination.

[0189] 2. Through alternative backhaul paths and multi-redundancy configurations, this embodiment can quickly switch and recover in case of ground network failures or satellite failures, achieving strong elasticity and disaster tolerance capabilities.

[0190] 3. According to the authorization level of user equipment, this embodiment dynamically adjusts bandwidth allocation and adopts resource preemption scheduling to meet the high requirements for communication in public safety and emergency rescue.

[0191] 4. After the main satellite link is restored, this embodiment performs progressive backhaul tests during low-traffic periods, gradually migrating traffic to the original main backhaul link. By adopting low-traffic period backhaul tests and progressive migration strategies, smooth backhaul can be achieved after the main satellite link is restored, avoiding sudden network fluctuations and improving backhaul reliability.

[0192] 5. This embodiment uses a multi-objective optimization algorithm to select alternative backhaul paths, combined with reinforcement learning, to adaptively adjust the optimal backhaul path, improving network resource utilization and backhaul stability.

[0193] The embodiment of the present invention also provides a computer device. Please refer to Figure 7 , Figure 7 which is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention, as shown in Figure 7As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting the components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if needed, multiple processors and / or multiple buses can be used with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 7 Taking one processor 10 as an example in

[0194] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.

[0195] Among them, the memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiments.

[0196] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory and can also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0197] The memory 20 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memories.

[0198] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or communication networks.

[0199] Embodiments of the present invention also provide a computer-readable storage medium. The method according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0200] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the method and / or technical solution according to the present invention can be called or provided. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible by the computer.

[0201] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the defined scope.

Claims

1. A multi-track elastic communication method, characterized in that, The method includes: Real-time monitoring of the performance parameters of a multi-orbit satellite link; the multi-orbit satellite link includes at least one of a geostationary orbit (GSO) satellite link and a non-geostationary orbit (NGSO) satellite link; Calculating the comprehensive score of each satellite link in the multi-orbit satellite link based on a preset scoring function; Dynamically adjusting the configuration strategy of the primary backhaul link according to the comprehensive score and a reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching satellite links, and performing load splitting; When detecting a network fault or congestion, activating an alternative backhaul path to complete traffic redirection and resume interrupted transmission, and executing corresponding disaster recovery strategies according to the fault level; and dynamically adjusting bandwidth allocation according to the authorization level of the user equipment.

2. The method according to claim 1, wherein The performance parameters include available bandwidth, latency, packet loss rate, and coverage stability; Before calculating the comprehensive score of each satellite link in the multi-orbit satellite link based on the preset scoring function, the method further includes: Constructing a dynamically updated global coverage database based on the performance parameters.

3. The method according to claim 1, wherein The preset scoring function is: Among them, S i represents the comprehensive score of satellite link i, W i represents the available bandwidth of satellite link i, W max represents the maximum available bandwidth of the satellite link, T i represents the delay of satellite link i, T max represents the maximum delay, P i represents the packet loss rate of satellite link i, P max represents the maximum packet loss rate of each satellite link, O i represents the orbital coverage stability of satellite link i, and α, β, γ, δ represent the weight coefficients in cross-orbit satellite collaboration.

4. The method according to claim 1, wherein The reward function of the reinforcement learning model is: R t = α·ThroughputGain - β·SwitchCost + γ·Qos Satisfaction ; Among them, R t represents the reward value, ThroughputGain represents the throughput gain caused by satellite link switching, SwitchCost represents the satellite link switching cost, QoS Satisfaction represents the service quality satisfaction, and α, β, and γ represent the weight coefficients in cross-orbit satellite collaboration.

5. The method according to claim 1, characterized in that, The dynamically adjusting the configuration strategy of the primary backhaul link according to the comprehensive score and the reinforcement learning model includes: Inputting the performance parameters and the comprehensive score into the reinforcement learning model for training, and generating an orbit switching or load splitting instruction in combination with the prediction of the movement trajectory of the user equipment; Dynamically adjusting the configuration strategy of the primary backhaul link based on the orbit switching or load splitting instruction.

6. The method according to claim 1, characterized in that The activating an alternative backhaul path to complete traffic redirection and resume interrupted transmission, and executing corresponding disaster recovery strategies according to the fault level when detecting a network fault or congestion includes: When detecting a ground core network or primary satellite link fault, activating an alternative backhaul path based on a hierarchical disaster recovery architecture, performing traffic redirection and resume interrupted transmission through a multi-path transmission protocol, and executing corresponding disaster recovery strategies according to the fault level; the alternative backhaul path includes at least one of the geostationary orbit (GSO) satellite link, the non-geostationary orbit (NGSO) satellite link, or an adjacent emergency base station.

7. The method according to claim 6, wherein The method further includes: After the primary satellite link is restored, performing a progressive fallback test during a low-traffic period to gradually migrate traffic to the original primary backhaul link.

8. The method according to claim 6, wherein Obtaining the alternative backhaul path through a multi-objective optimization algorithm, and the objective function of the multi-objective optimization algorithm is: Among them, W R represents the available bandwidth of the alternative backhaul path R, and W max represents the maximum available bandwidth in the selected alternative backhaul path. T R represents the delay of the alternative backhaul path R, and T max represents the maximum delay. Cost(S) represents the comprehensive cost of the set S of alternative backhaul paths, and Reliability(S) represents the joint availability when multiple alternative backhaul paths are in parallel. ω1, ω2, ω3, ω4 represent the weight coefficients in network disaster tolerance management; represents a set of selected alternative backhaul paths {R1, …, R n}.

9. The method according to any one of claims 1 to 8, characterized in that, The dynamically adjusting bandwidth allocation according to the authorization level of the user equipment includes: Allocating an independent network slice to target high-priority user equipment according to the authorization level of the user equipment, and dynamically adjusting bandwidth allocation.

10. A multi-track elastic communication system, characterized in that, The system is used to execute a multi-orbit elastic communication method according to any one of claims 1 to 9, and the system includes: The cross-orbit satellite collaboration module is used to monitor the performance parameters of multi-orbit satellite links in real time; calculate the comprehensive scores of each satellite link in the multi-orbit satellite links based on a preset scoring function; dynamically adjust the configuration strategy of the main backhaul link according to the comprehensive scores and the reinforcement learning model; the configuration strategy includes maintaining the current orbit, switching satellite links, and performing load diversion; the multi-orbit satellite links include at least one of the geostationary orbit (GSO) satellite links and non-geostationary orbit (NGSO) satellite links; The alternative backhaul and network disaster recovery management module is used to activate the alternative backhaul path to complete traffic redirection and resume interrupted transmission when detecting network failures or congestion, and execute corresponding disaster recovery strategies according to the failure levels; The authorized user equipment priority guarantee module is used to dynamically adjust the bandwidth allocation according to the authorization levels of user equipment when detecting network failures or congestion.

Images