Space-based service load migration method and system based on thermodynamic diagram dynamic region division

Abstract

The invention provides a space-based service load migration method and system based on thermodynamic diagram dynamic region division. The system comprises a thermodynamic diagram generation module, a dynamic region division module, a migration decision module and a fault-tolerant execution module. The dynamic region division module generates a service load thermodynamic diagram according to the thermodynamic diagram generation module, divides a network region, identifies a service hotspot region and quantifies the level; the migration decision module plans a service load migration path according to the service hotspot area; and the fault-tolerant execution module searches for replacement nodes to generate a plurality of paths, and executes load migration according to the optimal migration path. According to the method, global calculation is decomposed into parallel calculation of multiple orbit planes, global balance of space-based satellite networking platform loads is achieved, and the defect that under low-orbit rasterized giant constellation networking, resource waste and network congestion of partial regions are caused by uneven service loads is overcome.

Description

Technical Field

[0001] The present invention belongs to the field of space-based network intelligence and satellite edge computing technology. Specifically, it relates to a space-based business load migration method and system based on dynamic area division of heat maps, especially a space-based business efficient load migration method and device based on dynamic area division of heat maps. Background Art

[0002] Due to the spatial non-uniformity of user geographic distribution and the temporal non-uniformity of user activity periods, user services accessing space-based platforms result in non-uniform satellite loads, which in turn leads to resource waste or network congestion in some areas. In existing space-based service systems, load migration technology is a key means of ensuring service continuity and efficient resource utilization.

[0003] However, traditional load migration methods typically use static or simple dynamic threshold strategies for regional division and migration decisions, resulting in significant limitations in practical applications. Specifically, traditional methods often rely on fixed geographic or logical boundaries, making it difficult to accurately distinguish actual high-load areas. During the load migration process, repeated migrations within high-load areas can lead to a large number of invalid handoffs. Invalid handoffs introduce additional signaling interactions and migration delays, increasing system overhead, impacting space-based services with high real-time requirements (such as remote sensing data processing and emergency communications), and reducing service quality.

[0004] The patent document "Region-Based QoS Routing Method for Satellite Network Multi-Services" (CN113067625A) discloses a region-based QoS routing method for satellite networks. This method uses a virtual topology to divide time slices, determine service weights, and divide heavy-load and light-load areas. The method uses a ground control center to calculate static routing paths offline and dynamically calculates routing paths in heavy-load areas, reducing onboard computational complexity and routing signaling overhead. However, this is a static routing planning strategy with completely different routing paths, making it impossible to achieve adaptive partitioning, migration path optimization, and fault-tolerant execution.

[0005] The patent document "Polar Constellation Intersatellite Load Balancing Routing Method and System Based on Topology Switching" (CN117938238A) discloses a polar constellation intersatellite load balancing routing method based on topology switching. This method divides the ground control center into blocks, calculates block connectivity and link load strength, generates a routing table, and calculates switching time based on satellite operating patterns, reducing computational complexity and errors. However, this method targets polar constellations rather than low-orbit gridded mega-constellations, as the two differ significantly and cannot be migrated. Furthermore, it does not consider parallel computing and relies on offline ground-based computing.

[0006] Therefore, there is an urgent need for an efficient load migration method for space-based services based on dynamic area division of heat maps, which can dynamically perceive, adaptively divide and migrate the service hotspot areas, and achieve global load balancing of the space-based satellite networking platform. Summary of the Invention

[0007] In view of the defects in the prior art, the purpose of the present invention is to provide a method and system for migrating space-based business load based on dynamic area division of heat maps.

[0008] According to the present invention, a space-based service load migration method based on dynamic area division of a heat map is provided, comprising: Step S1: Collect data and build a business load heat map; Step S2: Divide the network area according to the business load heat map, identify the business hotspot area and quantify the level; Step S3: planning a service load migration path based on the divided service hotspot areas; Step S4: Find a replacement node to generate multiple paths, determine and select the optimal migration path to perform load migration.

[0009] Preferably, in step S1, the global data collection is decomposed into data collection of multiple track planes:

[0010] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; Represents the data set of the p-th orbital plane.

[0011] The orbital master serves as a computing node, receiving the load indicators of all satellites in its orbital plane, and calculating the corresponding mean and variance to generate a cluster summary; The ground station receives cluster summary information of all orbital plane masters, constructs orbital plane-to-plane mapping, and generates a business load heat map; An unsupervised clustering method is used on the received cluster summary information, and the satellites constituting the high-load area are regarded as the source of migration, and the satellites constituting the low-load area are regarded as the destination candidates of migration.

[0012] Preferably, the load index of local service access of each satellite on the orbital plane preprocessing orbital plane by the orbital plane master satellite is:

[0013] Perform Min-Max normalization to generate standardized feature vectors and obtain cluster summary information:

[0014] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; express The kth load indicator.

[0015] The data collection period is 30 seconds.

[0016] Preferably, in step S2, the network area is divided according to the number of inter-satellite link hops, time-space continuity detection is performed, isolated points are merged, service hotspot areas are identified, and load intensity is graded and quantified.

[0017] The network area is divided into two areas by judging the number of intersatellite link hops. If two satellites are interconnected through ≤2 hops, they are divided into the same network area. If two satellites are interconnected through >2 hops, they are divided into different network areas.

[0018] The spatiotemporal continuity detection is to determine whether the points in the same divided network area are in the continuous time window. If 80% of the time is in the same load state, if so, it is determined to be a stable hotspot and merged into the same load intensity level. If not, no adjustment is made.

[0019] The load intensity grading quantification is evenly divided into ranges according to the number of levels set for the task, and the processing strategies are divided into receivable migration, monitoring status and triggered migration.

[0020] Preferably, in step S3, the migration priority of each business hotspot area is calculated: Migration priority = average regional load × regional duration Through hierarchical routing planning, the path with the lowest latency is planned within the area with the same load intensity level; When crossing load intensity levels, select the area with the lowest load intensity level and select the area with the highest migration priority as the migration target area based on the migration priority. Plan a path that meets the specified constraints or requirements for bandwidth and latency between areas.

[0021] In step S4, a spatial neighborhood search method is used to seek a fast replacement node in a path within 3 hops, excluding satellites that share an orbital plane with the main path, and generating a new path. Each satellite maintains a local routing table containing the real-time delay / load status of 1-hop neighbors and the average delay of 2-hop neighbors.

[0022] Select the path with the lowest comprehensive cost. If there is only one path with the lowest comprehensive cost, it will be used as the optimal migration path. If there are multiple paths with the same comprehensive cost, determine whether they are in the same load intensity level area under the existing constraints.

[0023] If the load intensity levels of the regions are the same, the path with the lowest latency is selected as the optimal migration path. If the load intensity levels of the regions are different, the path with the best latency and bandwidth is selected as the optimal migration path.

[0024] According to the present invention, a space-based business load migration system based on dynamic area division of heat map is provided, comprising: a heat map generation module, a dynamic area division module, a migration decision module and a fault-tolerant execution module; The dynamic area division module divides the network area according to the business load heat map generated by the heat map generation module, identifies the business hotspot areas and quantifies their levels; The migration decision module plans the business load migration path based on the business hotspot areas; The fault-tolerant execution module searches for replacement nodes to generate multiple paths, determines and selects the optimal migration path to perform load migration.

[0025] Preferably, the heat map generation module decomposes the global data collection into data collection of multiple track planes:

[0026] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; Represents the data set of the p-th orbital plane.

[0027] The orbital master serves as a computing node, receiving the load indicators of all satellites in its orbital plane, and calculating the corresponding mean and variance to generate a cluster summary; The ground station receives cluster summary information of all orbital plane masters, constructs orbital plane-to-plane mapping, and generates a business load heat map; An unsupervised clustering method is used on the received cluster summary information, and the satellites constituting the high-load area are regarded as the source of migration, and the satellites constituting the low-load area are regarded as the destination candidates of migration.

[0028] Preferably, the load index of local service access of each satellite on the orbital plane preprocessing orbital plane by the orbital plane master satellite is:

[0029] Perform Min-Max normalization to generate standardized feature vectors and obtain cluster summary information:

[0030] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; express The kth load indicator.

[0031] The data collection period is 30 seconds.

[0032] Preferably, the dynamic area division module divides the network area according to the number of inter-satellite link hops, performs spatiotemporal continuity detection, merges isolated points, identifies service hotspot areas, and quantifies load intensity by levels.

[0033] The network area is divided into two areas by judging the number of intersatellite link hops. If two satellites are interconnected through ≤2 hops, they are divided into the same network area. If two satellites are interconnected through >2 hops, they are divided into different network areas.

[0034] The spatiotemporal continuity detection is to determine whether the points in the same divided network area are in the continuous time window. If 80% of the time is in the same load state, if so, it is determined to be a stable hotspot and merged into the same load intensity level. If not, no adjustment is made.

[0035] The load intensity grading quantification is evenly divided into ranges according to the number of levels set for the task, and the processing strategies are divided into receivable migration, monitoring status and triggered migration.

[0036] Preferably, the migration decision module calculates the migration priority of each business hotspot area: Migration priority = average regional load × regional duration Through hierarchical routing planning, the path with the lowest latency is planned within the area with the same load intensity level; When crossing load intensity levels, select the area with the lowest load intensity level and select the area with the highest migration priority as the migration target area based on the migration priority. Plan a path that meets the specified constraints or requirements for bandwidth and latency between areas.

[0037] The fault-tolerant execution module uses a spatial neighborhood search method to seek a fast replacement node in a path within three hops, excluding satellites that share an orbital plane with the main path, and generates a new path. Each satellite maintains a local routing table containing the real-time delay / load status of one-hop neighbors and the average delay of two-hop neighbors.

[0038] Select the path with the lowest comprehensive cost. If there is only one path with the lowest comprehensive cost, it is used as the optimal migration path. If there are multiple paths with the same comprehensive cost, determine whether they are in the same load intensity level area under the existing constraints. If the load intensity levels of the regions are the same, the path with the lowest latency is selected as the optimal migration path. If the load intensity levels of the regions are different, the path with the best latency and bandwidth is selected as the optimal migration path.

[0039] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention achieves global load balancing of the space-based satellite networking platform by dynamically sensing, adaptively dividing and migrating service hotspots.

[0040] 2. The present invention solves the drawbacks of both regional resource waste and network congestion caused by uneven service load in low-orbit gridded giant constellation networking by decomposing global computation into multiple orbital planes for parallel computation.

[0041] 3. The present invention implements a dynamic routing solution through migration priority calculation and fault tolerance mechanism, thereby improving the efficiency and loss of migration while meeting bandwidth and delay requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings: Figure 1 The figure is a flow chart of a space-based business load migration method based on dynamic area division of heat maps; Figure 2 Build a process diagram for business load heat map based on unsupervised clustering method; Figure 3 A flowchart of the adaptive service hotspot identification and load classification method; Figure 4 Schematic diagram of the global business load migration decision process for dynamic routing planning; Figure 5 This is a diagram of the process of performing high-reliability load migration based on a fault-tolerant mechanism; Figure 6 Schematic diagram of the space-based business load migration system based on dynamic area division of heat maps. DETAILED DESCRIPTION

[0043] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0044] According to the present invention, a space-based service load migration method based on dynamic area division of heat maps is provided. By dynamically sensing service hotspots, adaptively dividing and migrating the load, the load of the space-based satellite networking platform is balanced across the entire region, solving the problems of resource waste and network congestion in some areas caused by uneven service load in low-orbit gridded giant constellation networks. Figure 1 For example, the specific steps include: Step S1: construct a business load heat map based on an unsupervised method, and generate a business load heat map by spatial hierarchical management clustering; Specifically, Figure 2 For example, the global data collection is decomposed into data collection of multiple orbital planes:

[0045] in, is the data of the i-th satellite on the same orbital plane, is the total satellite data of the p-th orbital plane, Represents data for all orbital planes. Global computations are broken down into multiple orbital planes for parallel computation, with each orbital plane processing only the satellite data for that plane.

[0046] The main satellite of the orbital plane serves as a computing node, receiving the normalized indicator statistics of the payloads of all satellites in this plane and calculating the mean and variance of each corresponding indicator.

[0047] Specifically, the load indicators of local service access of each satellite in the pre-processing orbital plane (processor utilization, memory occupancy, current ISL connection status, and orbital position) are as follows:

[0048] And normalize the heterogeneous multi-index to eliminate the dimension effect, generate standardized feature vectors, and reduce the transmission volume. Using the Min-Max normalization method, it is expressed as:

[0049] At the same time, the data collection period is set to 30 seconds to accommodate the rapid motion of the constellations.

[0050] The ground station receives cluster summaries of primary satellites on each orbital plane and constructs a mapping from orbital planes to a plane to generate a global view, known as a traffic load heat map. Unsupervised clustering methods (such as K-means) are applied to the received data to identify satellites that form high-load areas, serving as migration sources, and satellites that form low-load areas, serving as migration destinations.

[0051] Step S2: Adaptive service hotspot identification and load level classification: Through spatiotemporal continuity detection and load level quantification methods, service hotspots are identified and quantified, providing a basis for migration direction; Specifically, Figure 3 For example, based on the number of intersatellite link hops, if two satellites can communicate with each other through ≤2 hops, they belong to the same network area. Considering that a single continuous high-load area is misjudged as multiple isolated hotspots, spatiotemporal continuity detection is performed and isolated points are regionally merged.

[0052] Among them, spatial continuity means merging points with adjacent hops (≤2 hops); temporal continuity means merging points with adjacent hops (≤2 hops) in the time window. For example, when the continuous time window in the spatial region r If the area is under the same load state for 80% of the time, it is determined to be a stable hotspot and merged into the same load area.

[0053] Specifically, the spatiotemporal continuity is determined in a simple way, that is, by considering points in the same area and whether they are in a continuous time window. The same area is the business hotspot area divided according to the business load heat map, and the continuous time window The time interval is set relatively but not too long, preferably in the order of seconds. This ensures spatial consistency in the area division and reduces the instability and fluctuation of the division in time.

[0054] The load intensity is quantified into 8 levels (0-7) and the following regulations are made:

[0055] In more preferred embodiments, the load intensity levels are manually set according to specific needs, and the range is generally evenly divided according to the number of levels required.

[0056] Step S3: Based on the global service load migration decision of dynamic routing planning, a hierarchical routing planning method is used to plan the lowest latency path within the same load level area, and to plan the inter-area path that meets the stability conditions of the inter-satellite link (ISL) across load areas; Specifically, Figure 4For example, we select the neighborhood of the high-load cluster center as the starting point and the neighborhood of the low-load cluster center as the end point. We also calculate the migration priority of each area using the following formula: Migration priority = average regional load × regional duration A hierarchical routing planning strategy is adopted. Within the same load area, the path with the lowest latency is prioritized. When migrating across regions, the divided low-load area is selected, and the area with a high migration priority is selected as the migration target area based on the migration priority. The inter-satellite link stability conditions must be met under the specified bandwidth and latency constraints between regions.

[0057] The intersatellite link stability condition refers to a stable link that can guarantee the required bandwidth and delay constraints or requirements. A link that meets the bandwidth and delay requirements is considered stable.

[0058] Step S4: Perform high-reliability load migration based on the fault-tolerant mechanism.

[0059] Specifically, Figure 5 For example, based on the existing routing planning, considering the sudden abnormal unreachability of nodes, the spatial neighborhood search method is used to consider the replacement of nodes in paths within 3 hops and select multiple paths at the same time.

[0060] Seek to quickly replace nodes, exclude satellites that share the orbital plane with the main path, and generate a new path to avoid cascading failures within the plane and ensure the reliability of the overall migration path.

[0061] Each satellite maintains a local routing table containing the real-time status (latency / load) of its one-hop neighbors and summary information (average latency) of its two-hop neighbors. It selects the path with the lowest overall cost. If multiple paths are available, it chooses the path that offers the best combination of low latency and high bandwidth under the existing constraints. This is a dynamic routing solution. Within the same region, the lowest latency is prioritized, while across regions, both latency and bandwidth constraints are considered.

[0062] The present invention also provides a space-based business load migration system based on dynamic area division of heat maps. The space-based business load migration system based on dynamic area division of heat maps can be implemented by executing the process steps of the space-based business load migration method based on dynamic area division of heat maps. That is, those skilled in the art can understand the space-based business load migration method based on dynamic area division of heat maps as an optimal implementation of the space-based business load migration system based on dynamic area division of heat maps.

[0063] According to the present invention, a space-based business load migration system based on dynamic area division of heat map is provided. Figure 6 For example, including: Heat map generation module: used for heat map construction and generation; Dynamic area division module: used for dynamic division of load areas; Migration decision module: used for migration direction selection and dynamic route planning; Fault-tolerant execution module: used for standby node switching and rapid route replanning.

[0064] The dynamic area division module divides the network area according to the business load heat map generated by the heat map generation module, identifies the business hotspot areas and quantifies their levels; The migration decision module plans the business load migration path based on the business hotspot areas; The fault-tolerant execution module searches for replacement nodes to generate multiple paths, determines and selects the optimal migration path to perform load migration.

[0065] In more preferred embodiments, the heat map generation module decomposes the global data collection into data collection of multiple track surfaces:

[0066] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; Represents the data set of the p-th orbital plane.

[0067] The orbital master serves as a computing node, receiving the load indicators of all satellites in its orbital plane, and calculating the corresponding mean and variance to generate a cluster summary; The ground station receives cluster summary information of all orbital plane masters, constructs orbital plane-to-plane mapping, and generates a business load heat map; An unsupervised clustering method is used on the received cluster summary information, and the satellites constituting the high-load area are regarded as the source of migration, and the satellites constituting the low-load area are regarded as the destination candidates of migration.

[0068] In more preferred embodiments, the load index of local service access of each satellite in the orbital plane preprocessing orbital plane by the orbital plane master satellite is:

[0069] Perform Min-Max normalization to generate standardized feature vectors and obtain cluster summary information:

[0070] in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; express The kth load indicator.

[0071] The data collection period is 30 seconds.

[0072] In more preferred examples, the dynamic area division module divides the network area according to the number of inter-satellite link hops, performs spatiotemporal continuity detection, merges isolated points, identifies service hotspot areas, and quantifies load intensity by levels.

[0073] The network area is divided into two areas by judging the number of intersatellite link hops. If two satellites are interconnected through ≤2 hops, they are divided into the same network area. If two satellites are interconnected through >2 hops, they are divided into different network areas.

[0074] The spatiotemporal continuity detection is to determine whether the points in the same divided network area are in the continuous time window If 80% of the time is in the same load state, if so, it is determined to be a stable hotspot and merged into the same load intensity level. If not, no adjustment is made.

[0075] The load intensity grading quantification is evenly divided into ranges according to the number of levels set for the task, and the processing strategies are divided into receivable migration, monitoring status and triggered migration.

[0076] In more preferred embodiments, the migration decision module calculates the migration priority of each business hotspot area: Migration priority = average regional load × regional duration Through hierarchical routing planning, the path with the lowest latency is planned within the area with the same load intensity level; When crossing load intensity levels, select the area with the lowest load intensity level and select the area with the highest migration priority as the migration target area based on the migration priority. Plan a path that meets the specified constraints or requirements for bandwidth and latency between areas.

[0077] The fault-tolerant execution module uses a spatial neighborhood search method to seek a fast replacement node in a path within three hops, excluding satellites that share an orbital plane with the main path, and generates a new path. Each satellite maintains a local routing table containing the real-time delay / load status of one-hop neighbors and the average delay of two-hop neighbors.

[0078] Select the path with the lowest comprehensive cost. If there is only one path with the lowest comprehensive cost, it is used as the optimal migration path. If there are multiple paths with the same comprehensive cost, determine whether they are in the same load intensity level area under the existing constraints. If the load intensity levels of the regions are the same, the path with the lowest latency is selected as the optimal migration path. If the load intensity levels of the regions are different, the path with the best latency and bandwidth is selected as the optimal migration path.

[0079] Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, and units provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same functions of the system and its various devices, modules, and units provided by the present invention in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; the devices, modules, and units for implementing various functions can also be considered as both software modules implementing the method and structures within the hardware component.

[0080] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A space-based service load migration method based on dynamic area division of heat maps, characterized in that: include: Step S1: Collect data and build a business load heat map; Step S2: Divide the network area according to the business load heat map, identify the business hotspot area and quantify the level; Step S3: planning a service load migration path based on the divided service hotspot areas; Step S4: Find a replacement node to generate multiple paths, determine and select the optimal migration path to perform load migration.

2. The space-based service load migration method based on dynamic area division of heat map according to claim 1 is characterized in that: In step S1, the global data collection is decomposed into data collection of multiple track planes: in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; Represents the data set of the p-th orbital plane; The orbital master serves as a computing node, receiving the load indicators of all satellites in its orbital plane, and calculating the corresponding mean and variance to generate a cluster summary; The ground station receives cluster summary information of all orbital plane masters, constructs orbital plane-to-plane mapping, and generates a business load heat map; An unsupervised clustering method is used on the received cluster summary information, and the satellites constituting the high-load area are regarded as the source of migration, and the satellites constituting the low-load area are regarded as the destination candidates of migration.

3. The space-based service load migration method based on dynamic area division of heat map according to claim 2 is characterized in that: The load index of local business access of each satellite in the orbital plane master satellite pre-processing orbital plane is: Perform Min-Max normalization to generate standardized feature vectors and obtain cluster summary information: in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; express The kth load index of The data collection period is 30 seconds.

4. The method for migrating space-based service load based on dynamic area division of heat map according to claim 1, characterized in that: In step S2, the network area is divided according to the number of inter-satellite link hops, time-space continuity detection is performed, isolated points are merged, service hotspots are identified, and load intensity is graded and quantified; The network area is divided by judging the number of intersatellite link hops. If two satellites are interconnected through ≤2 hops, they are divided into the same network area. If two satellites are interconnected through >2 hops, they are divided into different network areas. The spatiotemporal continuity detection is to determine whether the points in the same divided network area are in the continuous time window If 80% of the time is in the same load state, if so, it is determined to be a stable hotspot and merged into the same load intensity level. If not, no adjustment is made; The load intensity grading quantification is evenly divided into ranges according to the number of levels set for the task, and the processing strategies are divided into receivable migration, monitoring status and triggered migration.

5. The method for migrating space-based business load based on dynamic area division of heat map according to claim 1, characterized in that: In step S3, the migration priority of each service hotspot area is calculated: Migration priority = average regional load × regional duration Through hierarchical routing planning, the path with the lowest latency is planned within the area with the same load intensity level; When crossing load intensity levels, select the area with the lowest load intensity level and, based on the migration priority, select the area with the highest migration priority as the migration target area. Plan a path that meets the specified constraints or requirements for bandwidth and latency between areas. In step S4, a spatial neighborhood search method is used to find a fast replacement node in a path within 3 hops, excluding satellites that share the orbital plane with the main path, and generating a new path. Each satellite maintains a local routing table containing the real-time delay / load status of 1-hop neighbors and the average delay of 2-hop neighbors. Select the path with the lowest comprehensive cost. If there is only one path with the lowest comprehensive cost, it is used as the optimal migration path. If there are multiple paths with the same comprehensive cost, determine whether they are in the same load intensity level area under the existing constraints. If the load intensity levels of the regions are the same, the path with the lowest latency is selected as the optimal migration path. If the load intensity levels of the regions are different, the path with the best latency and bandwidth is selected as the optimal migration path.

6. A space-based business load migration system based on dynamic area division of heat maps, characterized in that: include: Heat map generation module, dynamic region division module, migration decision module and fault-tolerant execution module; The dynamic area division module divides the network area according to the business load heat map generated by the heat map generation module, identifies the business hotspot areas and quantifies their levels; The migration decision module plans the business load migration path based on the business hotspot areas; The fault-tolerant execution module searches for replacement nodes to generate multiple paths, determines and selects the optimal migration path to perform load migration.

7. The space-based service load migration system based on dynamic area division of heat map according to claim 6 is characterized in that: The heat map generation module decomposes the global data collection into data collection of multiple track planes: in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; Represents the data set of the p-th orbital plane; The orbital master serves as a computing node, receiving the load indicators of all satellites in its orbital plane, and calculating the corresponding mean and variance to generate a cluster summary; The ground station receives cluster summary information of all orbital plane masters, constructs orbital plane-to-plane mapping, and generates a business load heat map; An unsupervised clustering method is used on the received cluster summary information, and the satellites constituting the high-load area are regarded as the source of migration, and the satellites constituting the low-load area are regarded as the destination candidates of migration.

8. The space-based service load migration system based on dynamic area division of heat map according to claim 7 is characterized in that: The load index of local business access of each satellite in the orbital plane master satellite pre-processing orbital plane is: Perform Min-Max normalization to generate standardized feature vectors and obtain cluster summary information: in, Represents the data of the i-th satellite on the same orbital plane; Represents all satellite data of the p-th orbital plane; Data representing all orbital surfaces; express The kth load index of The data collection period is 30 seconds.

9. The space-based service load migration system based on dynamic area division of heat map according to claim 6 is characterized in that: The dynamic area division module divides the network area according to the number of inter-satellite link hops, performs spatiotemporal continuity detection, merges isolated points, identifies service hotspots, and quantifies load intensity by level; The network area is divided by judging the number of intersatellite link hops. If two satellites are interconnected through ≤2 hops, they are divided into the same network area. If two satellites are interconnected through >2 hops, they are divided into different network areas. The spatiotemporal continuity detection is to determine whether the points in the same divided network area are in the continuous time window. If 80% of the time is in the same load state, if so, it is determined to be a stable hotspot and merged into the same load intensity level. If not, no adjustment is made; The load intensity grading quantification is evenly divided into ranges according to the number of levels set for the task, and the processing strategies are divided into receivable migration, monitoring status and triggered migration.

10. The space-based service load migration system based on dynamic area division of heat map according to claim 6, characterized in that: The migration decision module calculates the migration priority of each business hotspot area: Migration priority = average regional load × regional duration Through hierarchical routing planning, the path with the lowest latency is planned within the area with the same load intensity level; When crossing load intensity levels, select the area with the lowest load intensity level and, based on the migration priority, select the area with the highest migration priority as the migration target area. Plan a path that meets the specified constraints or requirements for bandwidth and latency between areas. The fault-tolerant execution module uses a spatial neighborhood search method to seek a fast replacement node in a path within 3 hops, excluding satellites that share an orbital plane with the main path, and generates a new path. Each satellite maintains a local routing table containing the real-time delay / load status of 1-hop neighbors and the average delay of 2-hop neighbors; Select the path with the lowest comprehensive cost. If there is only one path with the lowest comprehensive cost, it is used as the optimal migration path. If there are multiple paths with the same comprehensive cost, determine whether they are in the same load intensity level area under the existing constraints. If the load intensity levels of the regions are the same, the path with the lowest latency is selected as the optimal migration path. If the load intensity levels of the regions are different, the path with the best latency and bandwidth is selected as the optimal migration path.

Images