Hierarchical mobility management method and system for satellite-ground coordination in low earth orbit satellite network

Abstract

This invention discloses a hierarchical mobility management method and system for satellite-ground collaboration in low-Earth orbit (LEO) satellite networks. It primarily addresses the problems of excessive mobility signaling overhead and over-reliance on ground stations leading to network robustness issues in existing technologies for user mobility management. The solution involves: selecting key nodes for deploying Mobility Management Functions (AMFs) through network deployment planning to eliminate the maintenance overhead of virtual clusters; decoupling location updates from high-speed satellite movement by constructing dynamic activity areas for users and implementing a lazy location update strategy; establishing a mapping relationship between user identifiers and multiple nodes to achieve distributed on-board storage of user data, avoiding strong dependence on ground stations; and establishing a hierarchical paging mechanism, calculating the PCI index based on the user mobility model, and dynamically selecting paging strategies to achieve an optimal balance between paging overhead and paging latency. This invention can significantly reduce the overall signaling interaction cost while ensuring service continuity and eventual data consistency, effectively improving the robustness and autonomous operation efficiency of large-scale LEO satellite networks in scenarios with limited satellite-ground links, and can be applied to satellite communication networks.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, and specifically relates to a hierarchical mobility management method and system that can be used in satellite communication networks. Background Technology

[0002] With the rapid advancement of integrated space-ground information networks, low-Earth orbit (LEO) satellite internet constellations, represented by OneWeb and Starlink, have become an important supplement and extension to terrestrial cellular networks due to their advantages such as seamless global coverage, low transmission latency, and low link loss. However, LEO satellite networks have extremely high dynamic characteristics. LEO satellites typically move relative to the ground at high speeds of approximately 27,000 km / h, resulting in extremely short visibility time for a single satellite to ground terminals, usually only a few minutes. This rapidly changing time-varying nature of the network topology means that ground terminals face extremely frequent beam switching and inter-satellite switching during communication, posing a severe challenge to network mobility management (MM) technology. Currently, the main technical solutions for solving the mobility management problem of LEO satellites are: the traditional centralized solution MIPv6, the distributed solution DMM, and the dynamic virtualization / cluster solution, but all of these solutions have significant technical limitations.

[0003] The first type is the traditional centralized MIPv6 solution. Patent document CN202110048597.7 discloses a typical implementation that transmits terminal location information to a ground database for storage, with the ground handling the paging path calculation. However, this architecture has significant limitations: all data packets must circumvent ground anchor points, causing a severe "triangular routing" problem and significantly increasing transmission latency; simultaneously, the high dynamic characteristics of low-Earth orbit satellites lead to frequent satellite-to-ground signaling interactions, increasing registration and paging costs, and the highly centralized ground anchor points easily become single-point performance bottlenecks in the network.

[0004] The second category is the distributed solution DMM. Chan HA et al., in an IEEE Communications Journal, advocated for moving the anchor point function to the network edge to solve the single point of failure problem and improve scalability. However, this solution faces serious challenges: to ensure the network-wide reachability of users, the distributed anchor points need to frequently synchronize global location information, which will generate huge "mesh synchronization" signaling pressure in a large satellite network; in addition, due to geographical environments such as oceans and polar regions and geopolitical constraints, it is difficult to achieve uniform global deployment of ground gateways, and it is impossible to completely eliminate the impact of satellite-to-ground link propagation delay on handover performance.

[0005] The third category is dynamic virtualization / cluster solutions. Li T. et al. proposed building "satellite virtual gateways" based on NFV or using multiple satellites to form dynamic clusters in an attempt to reduce dependence on the ground. Its limitations are: the high-speed relative motion of satellites makes the service domain topology extremely unstable, and a huge amount of signaling is required to reconstruct the cluster in order to maintain its integrity; and it is limited by the limited computing and storage resources of the onboard platform, making it difficult to support high-load virtualization management and frequent intra-domain synchronization operations, and thus unable to meet the needs of large-scale commercial applications. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing methods by proposing a hierarchical mobility management method based on satellite-ground collaboration in low-Earth orbit satellite networks. This method aims to significantly reduce the signaling interaction cost of mobility management while ensuring service continuity and eventual data consistency in highly dynamic environments, thereby improving the robustness and autonomous operation efficiency of the mobility management system in scenarios with limited satellite-ground links.

[0007] The technical approach to achieving the objective of this invention is as follows: by selecting key nodes to carry AMF functions, the virtual cluster architecture is abandoned, thereby eliminating the overhead of maintaining strong consistency of virtual clusters; by using a lazy position update mechanism based on dynamic active regions, position updates are decoupled from the high-speed movement of satellites, eliminating invalid synchronization triggered by satellite movement and reducing user mobility management signaling overhead; and by using a hierarchical intelligent paging decision mechanism, paging efficiency and network resources are systematically and collaboratively optimized, improving system robustness.

[0008] Based on the above ideas, the technical solution of the present invention includes:

[0009] 1. A hierarchical mobility management method for satellite-ground coordination in a low-Earth orbit satellite network, characterized in that it includes:

[0010] (1) Construct a total cost management function to determine the optimal number of Access and Mobility Management Functions (AMFs) to deploy in the satellite network and assign network nodes without deployed AMFs to the nearest AMF node for management;

[0011] (2) Construct a dynamic activity area for the user terminal and monitor out-of-area behavior, and generate a location update request only when the user moves out of the area, so as to suppress unnecessary signaling interactions to the greatest extent;

[0012] (3) Based on the satellite orbit topology characteristics, establish a fixed logical mapping relationship between the user terminal and multiple AMF nodes, and use it as the user's logical anchor point. After the user's location update request is initiated, the AMF node under the jurisdiction of the access satellite to which the user belongs will synchronize the information between the logical anchor points.

[0013] (4) When it is necessary to paging a user, the hierarchical intelligent paging decision of satellite-ground coordination is executed. The AMF node under the jurisdiction of the access satellite to which the paging initiator belongs is given priority to conduct range paging. If paging fails, the latest status data of the user is obtained from the logical anchor point corresponding to the user, and the global paging strategy is dynamically selected based on the assessment of location uncertainty.

[0014] Furthermore, the total cost management function constructed in (1) includes the following implementation:

[0015] (1a) Define the cross-domain handover signaling cost function ,

[0016] (1b) Define the load balancing cost function :

[0017] (1c) Define the operation and maintenance complexity cost function :

[0018] (1d) Based on the three functions defined above, the total cost management function is obtained. :

[0019] .

[0020] The determination of the optimal number of Access and Mobility Management Functions (AMFs) to be deployed in the satellite network in (1) includes the following implementation:

[0021] (1e) Solve the total cost management function to obtain the optimal number of deployments. :

[0022] (1f) Deploy based on the optimal deployment quantity:

[0023] (1f1) Satellite network operating cycle Discretize into Each time slice yields a set of dynamic network topology graphs. ;

[0024] (1f2) Definition set For the AMF candidate set, construct the average access hop count over the entire cycle. As the objective function for evaluating the merits of site selection:

[0025] (1f3) Initialize the selected set Empty;

[0026] (1f4) In each iteration, compute the value in the selected set. Average number of access hops throughout the entire cycle before adding nodes Average number of access hops throughout the entire cycle after joining The difference between :

[0027] (1f5) Select the above Add the largest node to the selected set ;

[0028] (1f6) Repeat steps (1f4) and (1f5) until... The number of nodes in the array reaches the optimal deployment number in (1e). .

[0029] (1f7) will set All satellite nodes included are deployed as nodes with Access and Mobility Management Functions (AMF).

[0030] The process of assigning network nodes without deployed AMF to nearby AMF nodes in (1) includes the following implementation:

[0031] (1g) Based on ephemeris data, the ground control station calculates the network topology for different time slices within the network operating cycle, and for each time slice... Generate a compressed topology snapshot And take a snapshot of the corresponding topology before the start of each time slice. Flood to all access nodes in the network;

[0032] (1h) Access Node Based on the current system time, read the topology snapshot for the next time slice. Calculate its own node based on the shortest path search algorithm To AMF node set Each node Logical hop count, construct distance set :

[0033] (1i) When the time slice is switched, the access node Based on the current time slice Select the nearest AMF node as its managed node. :

[0034] (1j) Access Node Based on the above calculation results It initiates a unicast registration request to the selected AMF node to complete the synchronization of the jurisdiction relationship.

[0035] Furthermore, the hierarchical intelligent paging decision in (4) includes: first-level paging and second-level paging: if the AMF node exceeds the time threshold... If a valid paging response is received from the access node, the first-level paging is considered successful, and the paging process ends; if the AMF node's timer exceeds the time threshold... If no response is received, or if a clear message is received that the user is not in the domain, the first-level paging process is deemed to have failed, and the second-level paging process is immediately triggered.

[0036] 2. A hierarchical mobility management system for satellite-ground coordination in a low-Earth orbit satellite network, characterized in that it includes:

[0037] The network deployment planning and optimization module is used to construct a total cost management function that includes cross-domain failover, load balancing, and operational complexity; the optimal deployment number of AMFs is obtained by iterating through and solving this function. The system determines the physical location scheme of the AMF based on a greedy algorithm, and assigns nodes that have not deployed AMF to the nearest AMF for management.

[0038] The terminal mobility monitoring module is used to construct a virtual circular geofence with the last reported coordinates of the user terminal as the center, and adaptively adjust the radius of the fence according to the historical average movement rate; it collects its own location coordinates in real time and calculates the relative distance, and only triggers a location update request when it determines that a substantial out-of-area behavior has occurred, so as to suppress unnecessary signaling interactions;

[0039] The dynamic access and routing management module is used to receive periodic network topology snapshots uploaded by ground telemetry and control stations at the access satellite nodes. ; Utilize onboard computing power to autonomously calculate the set of inter-satellite link hop counts from this node to each AMF node. Based on this, it selects the best AMF node under its jurisdiction and establishes directional routing and jurisdiction relationships;

[0040] The distributed multi-anchor mapping module is used to map AMF nodes according to preset parameters. A differentiated ordered list and user terminal identifier Calculate the uniform index pointer and extract the corresponding A set of physical nodes serves as the user's fixed logical anchor points. This enables the uniform discretization and binding of user anchor points in the constellation's spatiotemporal domain.

[0041] The state consensus and synchronization control module is used to build a dynamic synchronization cluster for user information on AMF nodes, and execute majority write logic to update user state, that is, only when a majority write is received... Submit logs regularly;

[0042] The hierarchical intelligent paging decision module is used to calculate the location certainty index (PCI) based on the retrieved user mobility context when performing second-level paging at the AMF node, and dynamically select the nearest anchor paging, majority anchor paging, or full anchor paging strategy according to the PCI level.

[0043] The satellite-ground coordination and global control module is used to generate and distribute network-wide topology snapshots based on ephemeris data at ground telemetry and control stations; it also provides asynchronous fallback recovery of the network status when user location information synchronization fails and wide-area auxiliary paging services when anchor paging fails.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] Firstly, by abandoning the virtual cluster architecture and introducing a geofence-based lazy location update mechanism, this invention eliminates the virtual cluster consistency overhead while decoupling the user's logical location from the high-speed movement of satellites, significantly reducing the signaling overhead for user mobility management.

[0046] Secondly, this invention utilizes a hierarchical intelligent paging strategy based on the Location Determinism Index (PCI) to dynamically switch between three paging modes—"single anchor point delegation, majority anchor point collaboration, and full anchor point flooding"—according to the user's status, thereby achieving an optimal balance between paging success rate and wireless resource usage.

[0047] Third, relying on distributed multi-anchor mapping and on-board majority consensus mechanism, this invention ensures the eventual consistency of user mobility data and service continuity in scenarios where satellite-to-ground links are limited or nodes fail, significantly improving the robustness and autonomy of the system. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the implementation of the hierarchical mobility management method for satellite-ground coordination in low-Earth orbit satellite networks according to the present invention.

[0049] Figure 2 This is a flowchart of the user paging sub-process in the method of this invention.

[0050] Figure 3 This is a block diagram of the hierarchical mobility management system for satellite-ground coordination in a low-orbit satellite network according to the present invention;

[0051] Figure 4 This is a comparison chart of the mobility management signaling overhead between the present invention and existing mobility management solutions;

[0052] Figure 5 A comparison chart of the signaling overhead of this invention and existing paging methods is shown. Detailed Implementation

[0053] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the present invention and not all of it. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative effort should all fall within the protection scope of the present invention.

[0054] Example 1: A hierarchical mobility management method for satellite-ground coordination in low-Earth orbit satellite networks.

[0055] Reference Figure 1 The implementation steps of this example include the following:

[0056] Step 1: Construct the total cost function.

[0057] (1.1) Define the cross-domain handover signaling cost function :

[0058] ,

[0059] in, For switching frequency coefficient, This refers to the inter-satellite transmission overhead during a single cross-domain handover. For satellite-to-ground backhaul frequency coefficients, Cost of satellite-to-ground link backup;

[0060] (1.2) Define the load balancing cost function :

[0061] ,

[0062] in, This is the congestion penalty coefficient. The load-sharing efficiency factor. To maintain the basic processing overhead of the entire network connection;

[0063] (1.3) Define the operation and maintenance complexity cost function :

[0064] ,

[0065] in, The basic complexity coefficient, A non-linear growth factor greater than 1 Standard operation and maintenance cost per node;

[0066] (1.4) Based on the three functions defined above, the total cost management function is obtained. :

[0067] .

[0068] Step 2: Solve for the total cost function Determine the optimal number of AMF deployments and deploy the AMF.

[0069] (2.1) Solve the total cost function to obtain the optimal deployment quantity for AMF:

[0070] (2.1.1) Deploy the number of AMFs The range of values ​​is set to a closed interval. ,in This represents the total number of satellite nodes in the current low-Earth orbit satellite constellation.

[0071] (2.1.2) Iterate through each integer in the range of values. ( Substitute this number as the assumed AMF deployment quantity into the total cost management function defined in step 1, and calculate the corresponding total cost function value one by one. This yields the cost set for different deployment numbers. ;

[0072] (2.1.3) Take the minimum value in the cost set. Value as the optimal deployment quantity for AMF ;

[0073] (2.2) The deployment set of AMF is obtained by solving based on the optimal deployment number of AMF. :

[0074] (2.2.1) Define the minimum common operating period for a satellite constellation. , will cycle Divided into Each time slot is a discrete time slice of equal duration. The ground control center uses satellite ephemeris data for each time sampling point. Calculate the spatial location and inter-satellite link connection status of all satellite nodes in the entire network, and generate a corresponding set of static network topology snapshots. ;

[0075] (2.2.2) Define a set For the AMF candidate set, construct the average access hop count over the entire cycle. As the objective function for evaluating the merits of site selection:

[0076] ,

[0077] in, To access the satellite collection, Represents a set of topology snapshots The total number of time slices included represents the complete runtime covered by the evaluation. Indicates the first Topology snapshot Under the defined specific network connection relationship, access nodes To AMF node The shortest path hop count;

[0078] (2.2.3) Initialize the selected set Empty;

[0079] (2.2.4) In each iteration, calculate the value in the selected set. Average number of access hops throughout the entire cycle before adding nodes Average number of access hops throughout the entire cycle after joining The difference between :

[0080] ;

[0081] (2.2.5) Select the option that makes the above Add the largest node to the selected set ;

[0082] (2.2.6) Repeat steps (2.2.4) and (2.2.5) until... The number of nodes reaches the optimal deployment number of AMF in (2.1.3). The set that has been selected at this time As an AMF node set .

[0083] (2.3) Set All satellite nodes included are deployed as nodes with Access and Mobility Management Functions (AMF).

[0084] Step 3: Assign satellite nodes that have not deployed AMF as access nodes to the nearest AMF node for management.

[0085] (3.1) Based on ephemeris data, the ground control station calculates the network topology for different time slices within the future period, and for each time slice... Generate a compressed topology snapshot And take a snapshot of the corresponding topology before the start of each time slice. Flood to all access nodes in the network;

[0086] (3.2) Access Node Received the next time-slice topology snapshot Then, calculate the value of its own node based on the shortest path algorithm. To AMF node set Each node Logical hop count, construct distance set :

[0087] ,

[0088] in, Indicates topology Calculated access nodes To AMF node The shortest path hop count;

[0089] (3.3) When switching time slices, the access node Based on the next time slice Select the nearest AMF node as its governing node. :

[0090] ;

[0091] (3.4) Access Node To its own managed nodes Initiate a unicast registration request to complete the synchronization of jurisdictional relationships.

[0092] Step 4: Construct the dynamic activity area of ​​the user terminal and perform location detection and lazy update.

[0093] (4.1) The location coordinates last reported by the user terminal Starting from, construct with With the center, and A virtual circular geofence with a radius of 1000 is defined, and the radius of the geofence is adaptively adjusted based on the average movement rate of the user terminal. :

[0094] ,

[0095] in, To preset the basic coverage radius, The average mobile speed of the user terminal during the statistical period. The velocity sensitivity coefficient;

[0096] (4.2) The user terminal periodically acquires its real-time location coordinates. And calculate the current position relative to the starting point. Great circle distance :

[0097] ;

[0098] (4.3) Calculate the spatial distance With the current fence radius A comparison will be performed only if the following conditions are met. If a substantial out-of-area violation is detected, the user terminal immediately initiates a location update with the network side and updates its current real-time location. Update to a new starting point Simultaneously reset the fence radius .

[0099] Step 5: Establish a fixed mapping between the user terminal and k logical anchor points.

[0100] (5.1) Select the Each AMF physical node is sorted according to the orbital plane number of its satellite. If the orbital plane numbers of the satellites are the same, they are sorted according to the initial phase angle within the satellite's orbital plane. After sorting, an initial ordered list that does not change with the real-time motion of the satellites is generated. :

[0101] ;

[0102] (5.2) Generate a random seed based on the system time. ;

[0103] (5.3) In order to achieve uniform discretization of user anchor points on different orbital planes and phases, the number of logical anchor points is defined. and build A differentiated sorted list :

[0104] (5.3.1) For the sequence number is A sorted list based on a random seed Calculate the fixed offset of the list ;

[0105] (5.3.2) Based on offset For the initial list Perform a circular left shift operation to generate an ordered list. ;

[0106] (5.3.3) Repeat (5.3.1) and (5.3.2) until generated. A sorted list;

[0107] (5.4) Based on the user terminal's identity identifier Calculate the uniform index of the user in the sorted list :

[0108] ,

[0109] in To determine the optimal number of AMF deployments;

[0110] (5.5) Based on the index For the product generated in step (5.3) Addressing is performed on each differentiated sorted list, that is, for each list Extract the first item from the list. The physical nodes corresponding to each bit are combined to form the user's fixed logical anchor point set. :

[0111]

[0112] .

[0113] Step 6: Establish a synchronization mechanism for user information across multiple anchor points.

[0114] To ensure eventual consistency of user location data and system fault tolerance in the event of partial node failure or link interruption, a majority confirmation strategy needs to be established between the synchronization master node and the synchronization slave nodes. Specifically, this is implemented as follows:

[0115] (6.1) Define the governing AMF node of the access star to which the user belongs as the synchronization master node, and the user Each fixed logical anchor point is a synchronous slave node;

[0116] (6.2) When the synchronization master node receives a user's location update request, it sends the user's location update to all slave nodes with the current term number, and at the same time starts the synchronization timer and sets the timer timeout threshold. ;

[0117] (6.3) The master node counts the number of acknowledgment responses received during the timing period. And when the timer reaches the timeout threshold Perform synchronous judgment at the same time:

[0118] like Satisfying the majority consensus condition If so, the status log is officially submitted locally and a success response is returned to the user, thus completing the synchronization.

[0119] like If the master node determines that strong consistency synchronization has failed, it will asynchronously send the status to the ground telemetry and control station for on-board data correction.

[0120] Step 7: Establish a hierarchical paging mechanism.

[0121] To achieve accurate matching between paging range and user status, it is necessary to prioritize low-cost searching for AMF nodes in the local access domain, and then dynamically match single anchor point, majority anchor point, or full anchor point paging strategies based on the user location certainty index (PCI) after the initial search fails.

[0122] Reference Figure 2 This step specifically includes:

[0123] (7.1) First-level paging:

[0124] (7.1.1) When an access node receives a paging request for a target user, it first routes the request to its current supervising AMF node, which then activates the paging service for a specified time threshold. The paging timer is activated, and paging commands are simultaneously sent to all access nodes within its local management domain to search for the target user within the current coverage area.

[0125] (7.1.2) The AMF node determines the paging result based on the paging timer status and the feedback result from the access node:

[0126] If the AMF node's timer exceeds the time threshold If a valid paging response is received from the access node, the first-level paging is considered successful, and the paging process ends.

[0127] If the AMF node's timer exceeds the time threshold If no valid paging response is received from the access node, or if a clear "user is not in this domain" feedback is received, the first-level paging is determined to have failed, and the second-level paging process is immediately triggered.

[0128] (7.2) Data preparation before performing the second-level paging:

[0129] (7.2.1) The AMF node calculates its own and the target user's... The status query request is sent to the anchor with the fewest hops among a fixed logical anchor point, based on the number of hops between them.

[0130] (7.2.2) After receiving the status response returned by the anchor point, parse the user's latest mobility context parameters, which include at least the user's last reported location coordinates. Dynamic activity area radius Average moving speed and the timestamp of the last update ;

[0131] (7.2.3) Calculate the user's location certainty index based on the user's latest mobility context parameters. :

[0132] ,

[0133] in, The average moving speed obtained from the retrieval; The timestamp of the current paging trigger moment. The timestamp of the last successful update of the user's location information; The radius of the user's current dynamic activity area; This is the normalized scaling adjustment factor;

[0134] (7.2.4) Load the preset hierarchical decision threshold parameters, that is, define the probability boundary for high effectiveness of location prediction. With a high confidence threshold of 0.7, a probability boundary is defined for location prediction facing a high risk of failure. The low confidence threshold is 0.3;

[0135] (7.3) The AMF nodes are calculated based on the With the preset boundary and Perform a comparison and dynamically execute the paging strategy based on the comparison results:

[0136] when If the user's state is stable, the AMF node will only select... The anchor point closest to itself among the anchor points is designated as the sole executor of paging, and a lightweight delegation instruction is sent to it. This anchor point independently calculates the paging satellite and issues the paging instruction.

[0137] when When ze determines that the user may be experiencing state drift, the AMF node selects... Anchor points form a “cooperative computing group” and send cooperative instructions to the group, requiring each anchor point to compare the user information it holds, determine the unique and valid user location through a consensus mechanism, and then the anchor point closest to the user calculates the paging execution satellite and issues the paging signaling.

[0138] when If the user's location is extremely dispersed, the AMF node selects all of the user's anchor points and performs paging in parallel. Each anchor point calculates the paging satellite based on the user data it holds and issues paging commands. At the same time, the ground core network is introduced to assist in paging and improve the paging success rate.

[0139] It should be noted that the step numbers in the above embodiments and claims are only for the purpose of clearly describing the implementation schemes of the present invention and facilitating understanding, and their sequence order is not limited.

[0140] Example 2: A hierarchical mobility management system for satellite-ground collaboration in a low-Earth orbit satellite network.

[0141] Reference Figure 3 This example includes: a network deployment planning and optimization module 1, a terminal mobility monitoring module 2, a dynamic allocation and routing management module 3, a distributed multi-anchor mapping module 4, a state consensus and synchronization control module 5, a hierarchical intelligent paging decision module 6, and a satellite-ground coordination and global management module 7. The hierarchical intelligent paging decision module 6 includes a context feature extraction submodule 61, a user location uncertainty calculation submodule 62, and a policy routing and execution submodule 63.

[0142] The working principle of the entire system is as follows:

[0143] The network deployment planning and optimization module 1 is used to construct a total cost management function that includes cross-domain switching, load balancing and operation and maintenance complexity. The optimal number of AMFs is obtained by traversing and solving the function, and the physical location scheme of AMFs is determined based on a greedy algorithm. Finally, the deployment planning results are transmitted to the dynamic allocation and routing management module 3 and the distributed multi-anchor mapping module 4.

[0144] The terminal mobility monitoring module 2 is used to construct a virtual circular geofence with the user terminal's last reported coordinates as the center, and adaptively adjust the fence radius according to the historical average movement rate; collect its own location coordinates in real time and calculate the relative distance, and trigger a location update request only when it is determined that a substantial out-of-area behavior has occurred, so as to suppress unnecessary signaling interactions, and send the user's location update request to the dynamic allocation and routing management module 3.

[0145] The dynamic allocation and routing management module 3 is used to determine the AMF node to which it belongs and establish a directional route based on the deployment planning results and real-time topology snapshot, and transmit the user's location update request to module 4 through the directional route to trigger the subsequent mapping and synchronization process.

[0146] The distributed multi-anchor mapping module 4 is used to calculate the receiving node list of the user's location update request, that is, to extract the user terminal identifier, calculate the user's unified index pointer based on the user ID, and calculate the anchor set of the user's k anchor points according to the preset k differentiated ordered lists, and transmit the anchor set as the receiving node set of the above request to the state consensus and synchronization control module 5.

[0147] The state consensus and synchronization control module 5 is used to construct a dynamic synchronization cluster pointing to these receiving nodes based on the set of receiving nodes for the user's location update request, and execute majority write logic, that is, only submit the log when more than half of the receiving nodes confirm. If the majority write is successful, the user data is shared to the hierarchical intelligent paging decision module 6. If it fails, the abnormal information is transmitted to the satellite-ground coordination and global management module 7.

[0148] The hierarchical intelligent paging decision module 6 is used to retrieve user mobility context and calculate the location certainty index (PCI) based on shared data at the AMF node, and dynamically select different paging strategies according to the PCI level. Specifically, the context feature extraction submodule 61 is used to extract the user's historical average movement rate, location update time, and current dynamic activity area radius from the shared data when the second-level paging is triggered at the AMF node, establish a user mobility model, and transmit the user model parameters to the user location uncertainty calculation submodule 62. This user location uncertainty calculation submodule 62 is used at the AMF node... The PCI index, calculated based on user mobility model parameters, is output to quantify user location confidence. This index is then transmitted to the policy routing and execution submodule 63. The policy routing and execution submodule 63 selects different paging strategies at the AMF node based on the PCI index: a single-anchor paging strategy is selected when the PCI is greater than the high confidence threshold; a majority anchor paging strategy is selected when the PCI is between the low and high confidence thresholds; and a full-anchor paging strategy is selected when the PCI is lower than the low confidence threshold. The auxiliary paging request generated by this strategy is then sent to the satellite-ground coordination and global control module 7.

[0149] The satellite-ground coordination and global control module 7 is used to generate and distribute a network topology snapshot to the dynamic allocation and routing management module 3 based on ephemeris data at the ground telemetry and control station; to provide asynchronous fallback recovery of the network status when a majority write fails; and to provide wide-area auxiliary paging service when an auxiliary paging request is received.

[0150] It should be noted that the above functional modules can be implemented, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented, in whole or in part, as program instruction products. A program instruction product includes one or a set of program instructions. When the program instructions are loaded and executed on a computer, the described process or function is generated, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The program instructions can be stored in a computer-readable and writable storage medium, or transferred from one computer's readable and writable storage medium to another.

[0151] The direct coupling or communication connections between the modules shown or discussed in this embodiment can be achieved through indirect coupling or communication connections via interfaces, devices, or modules. The various functional modules and sub-modules in this embodiment can dynamically reside within a single processing unit, or each module can exist physically independently, or two or more modules can dynamically reside within a single processing unit. When these dynamic components are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable and writable storage medium. This storage medium can be a memory, disk, or optical disc, etc.

[0152] The effectiveness of this invention can be further illustrated by the following simulation results.

[0153] I. Simulation Conditions

[0154] The simulation scenario was created using the simulation software MATLAB for simulation purposes.

[0155] The main simulation parameters are shown in Table 1:

[0156] Table 1. Main Simulation Parameters

[0157]

[0158] II. Simulation Content

[0159] Simulation 1: Under the above simulation parameter settings, user mobility is managed using the present invention and existing centralized ground management methods, network-wide flooding management methods, and virtual gateway cluster management methods, respectively. The simulation obtains the results of each scheme under different concurrent active user counts. The total system management cost is as follows: Figure 4 As shown.

[0160] from Figure 4 It is evident that the total system management cost of the three traditional solutions increases exponentially or linearly with network load, making it difficult to adapt to large-scale concurrent user scenarios. However, this invention effectively solves the problem of frequent updates of user access information caused by rapid changes in user access satellites through a lazy update mechanism for user location and on-board distributed storage of user information. This ensures that the total system management cost remains low and grows slowly under different numbers of users, demonstrating extremely high system resource utilization.

[0161] Simulation 2: Under the above simulation parameter settings, user mobility is managed using the present invention and existing centralized ground management methods, network-wide flooding management methods, and virtual gateway cluster management methods, respectively. The simulation obtains the paging overhead of each scheme at different user movement speeds, and the results are as follows: Figure 5 As shown.

[0162] from Figure 5 It is evident that the three traditional schemes, because they ignore the differences in user movement speed, maintain a consistently high and rigid paging overhead for the target user regardless of the user's movement speed. In contrast, the hierarchical paging strategy of this invention keeps the paging overhead for the target user consistently low while gradually increasing in a stepwise manner with the user's movement speed. This demonstrates that this invention can achieve an optimal balance between connection reliability and minimizing signaling overhead at different user movement speeds.

[0163] The simulation results show that the present invention can significantly reduce the overall signaling cost of the system while ensuring the continuity of user services, and effectively improve the robustness and autonomous operation efficiency of large-scale low-Earth orbit satellite networks.

Claims

1. A hierarchical mobility management method for satellite-ground coordination in a low-Earth orbit satellite network, characterized in that it includes: (1) Construct a total cost management function to determine the optimal number of Access and Mobility Management Functions (AMFs) to deploy in the satellite network and assign network nodes without deployed AMFs to the nearest AMF node for management; (2) Construct a dynamic activity area for the user terminal and monitor out-of-area behavior, and generate a location update request only when the user moves out of the area, so as to suppress unnecessary signaling interactions to the greatest extent; (3) Based on the satellite orbit topology characteristics, establish a fixed logical mapping relationship between the user terminal and multiple AMF nodes, and use it as the user's logical anchor point. After the user's location update request is initiated, the AMF node under the jurisdiction of the access satellite to which the user belongs will synchronize the information between the logical anchor points. (4) When it is necessary to paging a user, the hierarchical intelligent paging decision of satellite-ground coordination is executed. The AMF node under the jurisdiction of the access satellite to which the paging initiator belongs is given priority to conduct range paging. If paging fails, the latest status data of the user is obtained from the logical anchor point corresponding to the user, and the global paging strategy is dynamically selected based on the assessment of location uncertainty.

2. The method according to claim 1, characterized in that: The total cost management function constructed in (1) includes the following implementation: (1a) Define the cross-domain handover signaling cost function , ； in, For switching frequency coefficient, This refers to the inter-satellite transmission overhead during a single cross-domain handover. For satellite-to-ground backhaul frequency coefficients, Cost of satellite-to-ground link backup; (1b) Define the load balancing cost function : ； in, This is the congestion penalty coefficient. The load-sharing efficiency factor. To maintain the basic processing overhead of the entire network connection; (1c) Define the operation and maintenance complexity cost function : ； in, The basic complexity coefficient, A non-linear growth factor greater than 1 Standard operation and maintenance cost per node; (1d) Based on the three functions defined above, the total cost management function is obtained. : 。 3. The method according to claim 1, characterized in that: The determination of the optimal number of Access and Mobility Management Functions (AMFs) to be deployed in the satellite network in (1) includes the following implementation: (1e) Solve the total cost management function to obtain the optimal number of deployments: (1e1) Set the number of AMF deployments The range of values ,in Given the total number of satellite nodes in the constellation, iterate through each integer in the range. Calculate the corresponding total cost function value. ; (1e2) Compare all Value, select to make Reaching the global minimum corresponds to The value is output as the optimal number of AMF deployments. ; (1f) Deploy based on the optimal deployment quantity: (1f1) Satellite network operating cycle Discretize into Each time slice yields a set of dynamic network topology graphs. ; (1f2) Definition set For the AMF candidate set, construct the average access hop count over the entire cycle. As the objective function for evaluating the merits of site selection: ， in, To access the satellite collection, Indicates in Always access node To AMF node The shortest path hop count; (1f3) Initialize the selected set Empty; (1f4) In each iteration, compute the value in the selected set. Average number of access hops throughout the entire cycle before adding nodes Average number of access hops throughout the entire cycle after joining The difference between : ； (1f5) Select the above Add the largest node to the selected set ; (1f6) Repeat steps (1f4) and (1f5) until... The number of nodes in the array reaches the optimal deployment number in (1e). ; (1f7) will set All satellite nodes included are deployed as nodes with Access and Mobility Management Functions (AMF).

4. The method according to claim 1, characterized in that: The process of assigning network nodes without deployed AMF to nearby AMF nodes in (1) includes the following implementation: (1g) Based on ephemeris data, the ground control station calculates the network topology for different time slices within the network operating cycle, and for each time slice... Generate a compressed topology snapshot And take a snapshot of the corresponding topology before the start of each time slice. Flood to all access nodes in the network; (1h) Access Node Based on the current system time, read the topology snapshot for the next time slice. Calculate its own node based on the shortest path algorithm To AMF node set Each node Logical hop count, construct distance set : ， in, Indicates topology Calculated access nodes To AMF node The shortest path hop count; (1i) When the time slice is switched, the access node Based on the next time slice Select the nearest AMF node as its managed node. : ； (1j) Access Node To its own managed nodes Initiate a unicast registration request to complete the synchronization of jurisdictional relationships.

5. The method according to claim 1, characterized in that: The implementation of (2) for constructing a dynamic activity area for the user terminal and monitoring cross-area behavior includes: (2a) The location coordinates last reported by the user terminal Starting from, construct with With the center, and A virtual circular geofence with a radius of 1000 is defined, and the radius of the geofence is adaptively adjusted based on the average movement rate of the user terminal. : ； in, Based on the coverage radius, The average mobile speed of the user terminal during the statistical period. The velocity sensitivity coefficient; (2b) The user terminal periodically acquires its real-time location coordinates. And calculate the current position relative to the starting point. Great circle distance : ； (2c) Calculate the spatial distance With the current fence radius A comparison will be performed only if the following conditions are met. If a substantial out-of-area violation is detected, the user terminal immediately initiates a location update with the network side and updates its current real-time location. Update to a new starting point Simultaneously reset the fence radius .

6. The method according to claim 1, characterized in that: The implementation of (3) establishing a fixed mapping relationship between the user terminal and multiple AMF nodes based on satellite orbit topology features includes: (3a) Select the Each AMF physical node is sorted according to the orbital plane number of its satellite. If the orbital plane numbers of the satellites are the same, they are sorted according to the initial phase angle within the satellite's orbital plane. After sorting, an initial ordered list that does not change with the real-time motion of the satellites is generated. : ； (3b) In order to achieve uniform discretization of user anchor points on different orbital planes and phases, the number of logical anchor points is defined. And construct using a circular shift mechanism A differentiated sorted list : (3b1) For the sequence number is A sorted list based on a random seed Calculate the fixed offset of the list ; (3b2) Based on offset For the initial list Perform a circular left shift operation to generate an ordered list. ; (3b3) Repeat (3b1) and (3b2) until generated. A sorted list; (3c) Based on the user terminal's identity identifier Calculate the uniform index of the user in the sorted list : , in To determine the optimal number of AMF deployments; (3d) Based on the index For the product generated in step (3b) Addressing is performed on each differentiated sorted list, that is, for each list Extract the first item from the list. The physical nodes corresponding to each bit are combined to form the user's fixed logical anchor point set. : ， 。 7. The method according to claim 1, characterized in that: The information synchronization between logical anchor points by the governing AMF node of the user's access star in (3) includes the following: (3e) Define the governing AMF node of the user's access star as the synchronization master node, and define... Each fixed logical anchor point is a synchronous slave node; (3f) When the synchronization master node receives a user's location update request, it sends the user's location update to all slave nodes with the current term number, and at the same time starts the synchronization timer and sets the timer timeout threshold. ; (3g) The master node counts the number of acknowledgment responses received during the timing period. And when the timer reaches the timeout threshold Perform synchronous judgment at the same time: like Satisfying the majority consensus condition If so, the status log is officially submitted locally and a success response is returned to the user, thus completing the synchronization. like If the master node determines that strong consistency synchronization has failed, it will asynchronously send the status to the ground telemetry and control station for on-board data correction.

8. The method according to claim 1, characterized in that: The hierarchical intelligent paging decision in (4) includes: (4a) First-level paging: (4a1) When an access node receives a paging request for a target user, it first routes the request to its current supervising AMF node, which then activates the time threshold. The paging timer is activated, and paging commands are simultaneously sent to all access nodes within its local management domain to search for the target user within the current coverage area. (4a2) Determine the paging result based on the paging timer status and feedback results: If the AMF node's timer exceeds the time threshold If a valid paging response is received from the access node, the first-level paging is considered successful, and the paging process ends. If the AMF node's timer exceeds the time threshold If no response is received, or if a clear "user is not in this domain" feedback is received, the first-level paging is determined to have failed, and the second-level paging process is immediately triggered. (4b) Second-level paging: (4b1) The above AMF nodes calculate themselves and the target user's... The status query request is sent to the anchor with the fewest hops among a fixed logical anchor point, based on the number of hops between them. After receiving the status response from the anchor point, the user's latest mobility context parameters are parsed, including at least the user's last reported location coordinates. Dynamic activity area radius Average moving speed and the timestamp of the last update ; (4b2) The AMF node performs a status assessment based on the retrieved context parameters, and dynamically selects and executes the optimal paging strategy based on the location uncertainty.

9. The method according to claim 8, characterized in that: The implementation of (4b2) in which the optimal paging strategy is dynamically selected and executed based on location uncertainty includes: (4b2a) The AMF node uses the user mobility context parameters received in step (4b1) to calculate the user's location certainty index: ； in, The average moving speed obtained from the retrieval; The timestamp of the current paging trigger moment. The timestamp of the last successful update of the user's location information; The radius of the user's current dynamic activity area; This is the normalized scaling adjustment factor; (4b2b) AMF nodes are calculated based on Different paging strategies are assigned to user paging requests: when When the user's state is determined to be stable, the AMF node only selects... The anchor point closest to itself among the anchor points is designated as the sole executor of paging, and a lightweight delegation instruction is sent to it. This anchor point independently calculates the paging satellite and issues the paging instruction. when When it is determined that the user may be experiencing state drift, the AMF node selects... Anchor points form a "cooperative computing group" and send cooperative instructions to the group, requiring each anchor point to compare the user information it holds, determine the unique and valid user location through a consensus mechanism, and then the anchor point closest to the user calculates the paging execution satellite and issues the paging signaling. when When the user's location is determined to be extremely dispersed, the AMF node selects all the user's anchor points and performs paging in parallel. Each anchor point performs paging satellite calculations based on the user data it holds and issues paging commands. At the same time, the ground core network is introduced to assist in paging, thereby improving the paging success rate.

10. A hierarchical mobility management system for satellite-ground cooperation in a low-Earth orbit satellite network, characterized in that it includes: The network deployment planning and optimization module is used to build total cost management functions that include cross-domain switching, load balancing, and operational complexity. The optimal deployment quantity of AMF is obtained by iterating through and solving this function. And determine the AMF physical location scheme based on a greedy algorithm; The terminal mobility monitoring module is used to construct a virtual circular geofence with the last reported coordinates of the user terminal as the center, and adaptively adjust the radius of the fence according to the historical average movement rate; it collects its own location coordinates in real time and calculates the relative distance, and only triggers a location update request when it determines that a substantial out-of-area behavior has occurred, so as to suppress unnecessary signaling interactions; The dynamic allocation and routing management module is used to receive periodic network topology snapshots uploaded by ground telemetry and control stations at the access satellite nodes. ; Utilize onboard computing power to autonomously calculate the set of inter-satellite link hop counts from this node to each AMF node. Based on this, it selects the best AMF node under its jurisdiction and establishes directional routing and jurisdiction relationships; The distributed multi-anchor mapping module is used to map AMF nodes according to preset parameters. A differentiated ordered list and user terminal identifier Calculate the uniform index pointer and extract the corresponding A set of physical nodes serves as the user's fixed logical anchor points. This enables the uniform discretization and binding of user anchor points in the constellation's spatiotemporal domain. The state consensus and synchronization control module is used to build a dynamic synchronization cluster for user information on AMF nodes, and execute majority write logic to update user state, meaning that only user states received within a specified time limit are updated. Logs should be submitted regularly, if The abnormal information will then be transmitted to the ground station; The hierarchical intelligent paging decision module is used to calculate the location certainty index (PCI) based on the retrieved user mobility context when performing second-level paging at the AMF node, and dynamically select the nearest anchor paging, majority anchor paging, or full anchor paging strategy according to the PCI level. The satellite-ground coordination and global control module is used to generate and distribute network-wide topology snapshots based on ephemeris data at ground telemetry and control stations; it also provides asynchronous fallback recovery of the network status when user location information synchronization fails and wide-area auxiliary paging services when anchor paging fails.

11. The system according to claim 10, characterized in that: The hierarchical intelligent paging decision module includes: Context Feature Extraction Submodule: Used to extract the user's historical average mobility rate from the database when a second-level paging is triggered at the AMF node. Location update time and the radius of the user's current dynamic activity area And establish a user mobility model; User location uncertainty calculation submodule: used to load a preset normalized attenuation model into the AMF node. The extracted user mobility model is calculated, and the output is a quantified representation of user location reliability. index; Policy routing and execution submodule: used to perform calculations on AMF nodes. Different paging strategies are assigned to the index, i.e., when... Lock the nearest logical anchor point and issue a single-point delegation instruction; when When a majority of anchor points are formed and a state consensus command is issued, the nearest anchor point to the user is locked for paging execution; when The system generates and issues a full set of anchor point paging commands in real time, and simultaneously generates ground station auxiliary paging commands.

Images