Path acquisition method and apparatus, electronic device, storage medium, and computer program

Abstract

This disclosure provides a path acquisition method, apparatus, electronic device, storage medium, and computer program. The method includes: acquiring link device information of each autonomous system (AS) of a first target path; the link device information is used to characterize the cross-domain occupancy of the AS; determining a first link set based on the link device information of each AS; the first link set includes suspected conflicting links of the first target path in each AS; suspected conflicting links are links that may be occupied; acquiring at least one candidate replacement link corresponding to a suspected conflicting link based on the first link set; the AS is located in the same AS as the corresponding candidate replacement link; replacing the suspected conflicting link in the first target path with the candidate replacement link to obtain a second target path. This disclosure avoids the probability of occupancy conflicts, improves path stability, and is no longer extremely sensitive to delayed updates.

Description

Technical Field

[0001] This disclosure relates to the field of communications, and more particularly to a path acquisition method, apparatus, electronic device, storage medium, and computer program. Background Technology

[0002] In bearer networks designed for 5G, cloud computing, and multi-service convergence, Internet Protocol / Multi-Protocol Label Switching (IP / MPLS) technology has long been widely deployed as the mainstream end-to-end service bearer. MPLS technology achieves high-speed forwarding through label switching and can provide service quality assurance by combining traffic engineering mechanisms.

[0003] However, in the actual operation of cross-domain IP / MPLS networks, especially in scenarios involving multiple autonomous systems, resource usage and path state information synchronization between domains are often delayed due to factors such as control plane propagation latency, cross-domain interface bandwidth limitations, and information exchange strategies. This can lead to situations where the source domain considers a path available, but the actual target domain has already allocated it to other services, causing immediate conflicts after path establishment and triggering backoff and secondary convergence, thus significantly increasing service establishment latency. Summary of the Invention

[0004] This disclosure is made in view of the above-mentioned problems. This disclosure provides a path acquisition method, apparatus, electronic device, storage medium, and computer program.

[0005] Firstly, this disclosure provides a path acquisition method, the method including: Obtain link device information for each autonomous system (AS) along the first target path; the link device information is used to characterize the cross-domain occupancy status of the AS; Based on the link device information of each autonomous system, a first link set is determined; the first link set includes suspected conflict links of the first target path in each autonomous system; the suspected conflict links are links that may be occupied. Based on the first link set, at least one candidate replacement link corresponding to a suspected conflict link is obtained; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system. The suspected conflicting links in the first target path are replaced with the candidate replacement links to obtain the second target path.

[0006] In one embodiment of the first aspect, the link device information includes at least one of the following: link utilization, queue latency, packet loss rate, ratio of instantaneous peak to average traffic, and slice isolation status; before determining the first link set based on the link device information of each autonomous system, the method further includes: determining occupancy intensity information based on the link device information; the occupancy intensity information includes at least one of the following: occupancy rate change, cross-domain coupling, boundary conservatism, current occupancy level, and autonomous system rating information; determining whether each autonomous system has cross-domain occupancy based on a first preset condition; if so, determining the first link set based on the link device information of the autonomous systems with cross-domain occupancy; wherein, the first preset condition includes at least one of the following: the number of link segments in the autonomous system that meet the second preset condition meets a first quantity threshold, and the number of times the critical link segments in the autonomous system meet the second preset condition meets a second quantity threshold; the second preset condition includes at least one of the following: the comprehensive rating information of the autonomous system is greater than the first preset threshold, the occupancy change rate is greater than the second preset threshold, and the stability coefficient is less than the third preset threshold.

[0007] In one embodiment of the first aspect, determining the first link set based on link device information of autonomous systems with cross-domain occupancy includes: determining the link occupancy strength of each link in the autonomous system based on the link device information of the autonomous system with cross-domain occupancy; determining whether the link occupancy strength is greater than a dynamic conflict threshold; if so, determining the link as a suspected conflict link to obtain the first link set.

[0008] In one embodiment of the first aspect, when the number of failures of the second target path is greater than a fourth preset threshold, or when the number of candidate replacement links corresponding to the suspected conflicting links is less than a fifth preset threshold, the method further includes: obtaining link occupancy information of each autonomous system, determining a set of idle links corresponding to each suspected conflicting link; matching the attribute information of each link in the set of idle links with the attribute information of the target slice to determine at least one backhaul candidate link; the attribute information includes at least one of the following: bandwidth quota, isolation policy, latency level; the target slice is used to execute the service corresponding to the target path; sorting each backhaul candidate link to obtain a first candidate link; and replacing the suspected conflicting link in the second target path with the first candidate link to obtain a third target path.

[0009] In one embodiment of the first aspect, when the third target path does not meet the quality of service requirements, the method further includes: marking conflicting links in the third target path; selecting from the backhaul candidate links based on the conflicting links to obtain a second candidate link corresponding to the conflicting link; the autonomous system to which the second candidate link belongs is different from the autonomous system to which the first candidate link belongs. The fourth target path is obtained by replacing the conflicting links in the third target path with the second candidate links.

[0010] In one embodiment of the first aspect, the method further includes: determining a path delivery mode based on path information of the target path; the path delivery mode includes at least one of the following: general mode, satellite network mode, and privacy constraint mode; the path information includes at least one of the following: number of autonomous systems in the target path, link type, packet header length, device processing capacity, service information, number of links, control channel bandwidth, link delay, link occupancy fluctuation, and service requirements.

[0011] Secondly, this disclosure provides a path acquisition device, specifically including: an information acquisition module, configured to acquire link device information of each autonomous system of a first target path; the link device information is used to characterize the cross-domain occupancy of the autonomous system; The set determination module is configured to determine a first set of links based on the link device information of each autonomous system; the first set of links includes suspected conflicting links of the first target path in each autonomous system; the suspected conflicting links are links that may be occupied. The link acquisition module is configured to acquire at least one candidate replacement link corresponding to a suspected conflict link based on the first link set; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system. The path acquisition module is configured to replace the suspected conflicting links in the first target path with the candidate replacement links to obtain the second target path.

[0012] Thirdly, this disclosure provides an electronic device, including: a memory for storing computer-readable instructions; and a processor for executing the computer-readable instructions, causing the electronic device to perform the method as described in any embodiment of the first or second aspect.

[0013] Fourthly, this disclosure provides a non-transitory computer-readable storage medium for storing computer-readable instructions that, when executed by a processor, cause the processor to perform the method as described in any embodiment of the first or second aspect.

[0014] Fifthly, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the method as described in any of the embodiments of the first and / or second aspects.

[0015] This disclosure provides a path acquisition method. The method involves acquiring link device information for each autonomous system (AS) along a first target path. This link device information characterizes the cross-domain occupancy status of the AS. Based on the link device information of each AS, a first link set is determined. This first link set includes suspected conflicting links of the first target path in each AS. The suspected conflicting links are links that may be occupied. Based on the first link set, at least one candidate replacement link corresponding to a suspected conflicting link is acquired. The suspected conflicting link and the corresponding candidate replacement link reside in the same AS. The candidate replacement link replaces the suspected conflicting link in the first target path to obtain a second target path. In other words, this disclosure uses the acquired link device information to predict the status of each link in the first target path in advance, obtaining suspected conflicting links that may cause problems, thereby switching paths in advance. This avoids, to some extent, the path status misjudgment problem caused by delayed link device information updates, efficiently completing path selection and switching, thus avoiding the probability of occupancy conflicts, improving path stability, and reducing extreme sensitivity to delayed updates.

[0016] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0017] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 A schematic diagram illustrating a path acquisition method provided in an embodiment of this disclosure; Figure 2 This diagram illustrates a cross-domain determination method provided in an embodiment of the present disclosure. Figure 3 This is a schematic diagram of an occupancy gradation provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram illustrating a method for filtering suspected conflicting links provided in an embodiment of this disclosure; Figure 5 This is a schematic diagram of cross-domain occupancy provided in an embodiment of this disclosure; Figure 6 This is a schematic diagram of a cross-slice rollback provided in an embodiment of this disclosure; Figure 7 This is a schematic diagram of another cross-slice rollback provided in an embodiment of this disclosure; Figure 8 This is a schematic diagram illustrating a legality verification and mode selection method provided in an embodiment of this disclosure; Figure 9 This is a schematic diagram illustrating a general mode of delivery provided in an embodiment of this disclosure; Figure 10 This is a schematic diagram of a satellite network mode provided in an embodiment of this disclosure; Figure 11 This is a schematic diagram illustrating a privacy constraint mode provided in an embodiment of this disclosure; Figure 12 A schematic diagram illustrating another path acquisition method provided in this embodiment of the disclosure; Figure 13 A schematic diagram of a path acquisition device provided in an embodiment of this disclosure; Figure 14 A hardware block diagram of an electronic device provided in an embodiment of this disclosure; Figure 15 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of the present disclosure. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this disclosure more apparent, exemplary embodiments according to this disclosure will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments of this disclosure. It should be understood that this disclosure is not limited to the exemplary embodiments described herein.

[0020] In bearer networks designed for 5G, cloud computing, and multi-service convergence, Internet Protocol / Multi-Protocol Label Switching (IP / MPLS) technology has long been widely deployed as the mainstream end-to-end service bearer. MPLS technology achieves high-speed forwarding through label switching and can provide service quality assurance by combining traffic engineering mechanisms.

[0021] In cross-domain business scenarios, existing technologies typically rely on multi-domain traffic engineering, Path Computation Element (PCE) control architecture, or topology and state information distribution mechanisms based on Border Gateway Protocol-Link State (BGP-LS) to achieve cross-domain path calculation and distribution. In addition, some new solutions introduce MPLS-based segmented routing (SR-MPLS) or IPv6-based segmented routing (SRv6) technologies, leveraging the advantages of segmented routing in reducing protocol state maintenance and improving programming capabilities to enhance multi-domain path controllability and business deployment efficiency.

[0022] However, in the actual operation of cross-domain IP / MPLS networks, especially in scenarios involving multiple autonomous systems (i.e., a network composed of links from multiple autonomous systems), resource consumption and path state information synchronization between autonomous systems often experience delays due to factors such as control plane propagation latency, cross-domain interface bandwidth limitations, and information exchange strategies (e.g., only periodic summary exchanges). Here, an autonomous system can be understood as a network scope managed by a single organization with a unified routing policy. Examples include carrier domains and enterprise domains.

[0023] When a segment of a service path is simultaneously allocated to other service paths from different autonomous systems, a double occupancy misjudgment can occur if the occupancy information used by the source domain during path calculation has not yet reflected the latest allocation status of the target domain. This means that the source domain may consider the path segment available, while the target domain has already allocated it to another service. Furthermore, such conflicts typically surface after path establishment, easily triggering path backtracking and secondary convergence in the network, leading to a significant increase in service establishment latency and even short-term service interruptions. Existing solutions that rely on real-time synchronization of occupancy status by a global controller are extremely sensitive to occupancy information delays; delayed updates directly cause incorrect route selection, and there is a lack of effective prevention mechanisms.

[0024] On the other hand, in multi-path selection scenarios, when the occupancy values ​​of candidate paths are close or the path status fluctuates frequently within a short period, existing path selection mechanisms are prone to repeatedly switching between different candidate paths, resulting in routing oscillations. This not only wastes bandwidth resources but also triggers frequent reconfiguration of the forwarding plane, reducing overall routing stability. In cross-domain environments, if the occupancy perceptions of different domains are inconsistent, the frequent switching problem will be further exacerbated, affecting path stability and resource utilization efficiency.

[0025] To address the aforementioned issues, this disclosure predicts potentially conflicting links by analyzing link device information across various autonomous systems along the target path and proactively replaces them. This, to some extent, avoids multiple services conflicting on the same link. Please refer to [link / reference needed] for details. Figure 1 ,like Figure 1 As shown, the method includes: S101, Obtain link device information for each autonomous system (AS) along the first target path; the link device information is used to characterize the cross-domain occupancy status of the AS; Specifically, after identifying the sender and receiver, the first target path from the sender to the receiver can be determined. This first target path may pass through at least one autonomous system (AS). Based on the link device information of each AS, the traffic situation in each AS can be obtained. These link devices can be border devices; through their periodic status reports, the specific status of each cross-domain link can be obtained. In other words, based on the link device information, it is possible to infer whether cross-domain traffic occupancy may occur within an AS.

[0026] The link device information includes, but is not limited to, at least one of the following: link utilization, queue latency, packet loss rate, ratio of instantaneous peak to average traffic, and slice isolation status. Here, a slice can be understood as a network channel used to run a specific service, which may include different links from different autonomous systems.

[0027] Link utilization can be understood as the degree of congestion on a link.

[0028] Queue delay is used to characterize the busyness of a link and the forwarding pressure.

[0029] Packet loss rate is used to characterize whether a link is overloaded.

[0030] The ratio of instantaneous peak traffic to average traffic is used to characterize whether the link is stable.

[0031] Slice isolation status is used to characterize the degree of resource isolation on the link executing the service.

[0032] Optionally, after obtaining the link device information of different autonomous systems, it can be normalized and mapped to obtain a multi-dimensional cross-domain occupancy feature vector. This vector can represent the state of each autonomous system and can also eliminate the difference in dimensions to a certain extent, reducing the difficulty of subsequent processing.

[0033] Next, occupancy intensity information can be determined based on the link device information. This occupancy intensity information can be understood as data used to determine whether an autonomous system is congested and the intensity of congestion. This occupancy intensity information includes, but is not limited to, at least one of the following: occupancy rate change, cross-domain coupling degree, boundary conservatism, current occupancy level, and rating information for each autonomous system.

[0034] The current occupancy level is determined by the resource occupancy status of the autonomous region (e.g., Level 1: Idle, Level 2: Normal, Level 3: Busy, Level 4: Congested). The higher the level, the more congested the autonomous region is.

[0035] The occupancy change rate can be understood as the average transition speed and direction of occupancy levels within an autonomous system (ASI) per unit detection period. In other words, if a 10-second unit detection period is used, and the occupancy change rate is consistently increasing (e.g., a rapid increase in occupancy level within the period), or if the occupancy change rate is highly unstable (e.g., the occupancy level cannot be stabilized within the period), then the links in that ASI can be considered unstable, or about to become congested.

[0036] Stability refers to the strength of fluctuations in the number of segments occupied within a window. The larger the value and the closer it is to 1, the more stable the autonomous system is. The stability coefficient can be obtained by calculating the reciprocal of the dispersion of the segment values ​​within the same window.

[0037] Boundary conservatism refers to the extra safety factor added to key splicing positions such as cross-domain entry / exit points during decision-making, used to suppress cross-domain jitter caused by frequent switching.

[0038] Cross-domain coupling refers to the strength of whether related paths in adjacent domains of the same business path simultaneously experience an increase in occupancy. The higher the coupling, the more likely an increase in congestion in an autonomous system (Autonomous System) will trigger neighboring domains to follow suit. In other words, occupancy in one Autonomous System can quickly spread to adjacent domains, causing occupancy along the entire path, thus leading to a higher risk for the Autonomous Systems along that path.

[0039] The scoring information for each autonomous domain can be obtained based on actual conditions (such as one or more of the aforementioned information like occupancy change rate, stability, and cross-domain coupling degree). It should be noted that this disclosure does not specifically limit the method of obtaining the scoring information; it can be based on manual evaluation or calculated from actual data.

[0040] For example, rating information can be specifically represented as: ; Where I represents the scoring information of the autonomous region; the first column vector This is the occupancy change rate vector for all cross-domain link segments, used to characterize the occupancy change trend of each segment. (Second column vector) : The occupancy stability coefficient vector of all cross-domain link segments, representing the degree of fluctuation in the occupancy of each segment. This represents the average rate of change of the occupancy characteristics of link i over the most recent N periods (positive values ​​indicate an increase, and negative values ​​indicate a decrease). This represents the stability of link i's occupancy characteristics over the most recent N periods. Here, N, M, n, m, and i are all integers greater than 0.

[0041] Next, based on the first preset condition, it can be determined whether there is cross-domain occupation in each autonomous domain; if so, the first link set can be determined based on the link device information of the autonomous domain with cross-domain occupation.

[0042] The first preset condition includes at least one of the following: the number of link segments in the autonomous system that satisfy the second preset condition meets a first quantity threshold; and the number of times a critical link segment in the autonomous system satisfies the second preset condition meets a second quantity threshold. Here, a critical link can be understood as a link that is indispensable in the first target path.

[0043] The second preset condition includes at least one of the following: the comprehensive score information of the autonomous region is greater than the first preset threshold, the occupancy change rate is greater than the second preset threshold, and the stability coefficient is less than the third preset threshold.

[0044] In other words, when the overall score of an autonomous system (AS) exceeds the first preset threshold, it indicates that the AS is unstable and congested. If the rate of change in occupancy exceeds the second preset threshold, it indicates that link occupancy is rapidly increasing. A low stability coefficient indicates significant fluctuations within the AS. This suggests that links within the AS may experience conflicts in the future, or that the links are unstable and require replacement.

[0045] It should be noted that the first quantity threshold, the second quantity threshold, the first preset threshold, the second preset threshold, and the third preset threshold involved in this disclosure can all be determined according to the actual situation, and this disclosure does not impose specific restrictions here.

[0046] To facilitate understanding, specific examples will be provided below; please refer to them for details. Figure 2 ,like Figure 2 As shown, after collecting and reporting the data (i.e., the aforementioned acquisition of link device information), the data can be normalized, and then the rate of change (i.e., the rate of change of occupancy) and stability can be calculated. Based on the preset conditions (i.e., the first preset condition), it is determined whether the above data meets the requirements. If the conditions are met, it is determined that there may be cross-domain occupancy in the future. If the conditions are not met, monitoring can continue.

[0047] S102, determine a first link set based on the link device information of each autonomous system; the first link set includes suspected conflict links of the first target path in each autonomous system; the suspected conflict links are links that may be occupied. Once it's determined that an autonomous system (AS) may experience conflicts, further analysis of link device information can be used to identify potentially conflicting links within the AS. It's important to note that these potentially conflicting links refer to links within the first target path, i.e., links that might be occupied within that path. The first set of links includes at least one potentially conflicting link; in other words, within an AS that may experience cross-domain conflicts, at least one link might be in use.

[0048] Based on the link device information of the autonomous system with cross-domain occupancy, the link occupancy strength of each link in the autonomous system is determined; and it is determined whether the link occupancy strength is greater than the dynamic conflict threshold; if so, it indicates that the link is about to have a conflict or congestion, and the link can be identified as a suspected conflict link, thus obtaining the first link set.

[0049] Optionally, suspected conflict segments can be categorized according to their occupancy change rate. Since cross-domain status is not a static mean deviation but a non-stationary, random process with sudden jumps, asynchronous sampling and clock skew in the reports from different autonomous regions can easily lead to the source domain observing time-displaced occupancy vectors. Therefore, cross-domain occupancy strength matrix and dynamic conflict thresholds can be used for categorization to underestimate the prior risk of double occupancy and Type II errors. This can, to some extent, avoid selecting frequently fluctuating paths and prevent routing oscillations, thereby reducing bandwidth waste and improving overall routing stability. Please refer to Table 1 below for details: Table 1

[0050] As shown in Table 1, when the occupancy change rate is less than or equal to 0.25, it is considered low, indicating that the link occupancy within the autonomous system is changing slowly. If the occupancy change rate is greater than 0.25 and less than or equal to 0.5, it indicates that the link occupancy within the autonomous system is also changing slowly. At these two levels, it means that the autonomous system is unlikely to experience conflicts or congestion quickly. If the occupancy change rate is greater than 0.5 and less than or equal to 0.75, it indicates high occupancy. A occupancy change rate greater than 0.75 indicates extremely high occupancy, meaning that conflicts and congestion are likely to occur at any time. It should be noted that this range is only an example for ease of understanding; specific values ​​can be set according to actual conditions, and this disclosure does not impose specific limitations.

[0051] For easier understanding, please refer to Figure 3 ,like Figure 3As shown, after continuously receiving cross-domain link data (i.e., the aforementioned link device information), the occupancy intensity of each link can be calculated. If the occupancy intensity is greater than the dynamic conflict threshold, the link is marked as a suspected conflict segment (i.e., the aforementioned suspected conflict link). Next, the suspected conflict segments can be classified according to the preset change rate range, and the classification results are output.

[0052] S103, based on the first link set, obtain at least one candidate replacement link corresponding to a suspected conflict link; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system; Next, within the same autonomous system (AS), candidate replacement links similar to those suspected of conflicting links can be identified. These candidate replacement links can be those within the same AS with latency not significantly higher than the suspected conflicting links, low cross-domain coupling, and stable historical occupancy change rates (e.g., low or medium). This allows for the lowest cost, fastest switching speed, and least risk during subsequent path switching.

[0053] S104, replace the suspected conflicting links in the first target path with the candidate replacement links to obtain the second target path.

[0054] Optionally, among the candidate replacement links that meet the conditions, the one with the performance closest to the suspected conflict link can be selected as the second target path, and the one with the lowest fluctuation can be selected as the steady-state backup link, forming a buffer pair. When the suspected conflict link approaches or reaches the conflict threshold, the system can quickly switch to the steady-state backup link to avoid service interruption. The conflict threshold here can be understood as the critical point that affects the normal operation of the service. This disclosure does not specifically limit its type; it can be a fixed value or obtained based on system values ​​(for example, it can be dynamically obtained by combining link fluctuation levels and historical congestion events).

[0055] Furthermore, the path cost formed after the buffer pair replaces the original suspected conflict link can be calculated. Here, path cost refers to the comprehensive cost of the path after replacement, including but not limited to at least one of the following: latency cost, bandwidth occupation cost, handover overhead, cross-domain coordination cost, etc.

[0056] To further control costs, the path cost can be compared with the cost of the global rerouting path to obtain a switching strategy determination value. The global rerouting path can be understood as an overall alternative path scheme that bypasses the domain of the suspected conflicting link, typically including a longer transmission path, more cross-domain links, and higher latency fluctuations. If the path cost is lower than the cost of the global rerouting path, the second target path is adopted; otherwise, the global detour scheme is adopted.

[0057] Optionally, the handover process can be completed in one step in the forwardable table using an atomic replacement mechanism, thus avoiding short-term service interruptions caused by step-by-step updates. The atomic replacement mechanism can be understood as an operation mode that replaces the entire segment list at once in the network device's data plane, ensuring path continuity during handover. Furthermore, it can significantly reduce handover jitter and momentary packet loss. Simultaneously, cross-slice redundancy sharing is introduced, using segment-level borrowing based on a priority order of latency, followed by bandwidth, and then fluctuation level. This improves reachability and resource utilization without exposing domain details, alleviating expansion pressure.

[0058] Optionally, when a suspected conflicting link is shared by multiple service paths, all service paths involving the suspected conflicting link can be identified and listed, and the above operations can be performed uniformly to avoid omissions. In other words, within the autonomous system where the suspected conflicting link is located, this disclosure selects a buffer segment pair composed of a link with equivalent performance and a steady-state backup link, and selects the optimal one by comparing the cost of the primary / stable hybrid path with the cost of a full-domain detour; when the occupancy approaches the threshold, the segment list is switched at once, which can reduce control plane chain updates and forwarding jitter, and suppress routing oscillations and bandwidth waste caused by repeated switching between candidate paths. At the same time, chain updates are converged into a one-time effect, significantly reducing switching jitter and instantaneous packet loss. Furthermore, cross-slice redundancy sharing is introduced, with segment-level borrowing based on a priority order of latency, bandwidth, and fluctuation level, improving reachability and resource utilization without exposing domain details, and alleviating expansion pressure.

[0059] For easier understanding, please refer to Figure 4 ,like Figure 4 As shown, after identifying a suspected conflict segment (i.e., the aforementioned suspected conflict link), an impact analysis is first performed, which involves retrieving the cross-domain path database (corresponding to the business paths involved in the suspected conflict link) and forming a list of affected paths. Next, candidate link segments within the domain can be searched for each affected path (i.e., obtaining candidate replacement links corresponding to at least one suspected conflict link). Then, buffer segment pairs can be constructed, i.e., the aforementioned buffer pair construction, selecting the link with performance closest to the remote end as the primary replacement segment, and selecting the one with the lowest fluctuation as the steady-state backup segment (i.e., the steady-state backup link). Next, the cost of the primary / stable hybrid path (i.e., the path cost) can be calculated and a cost table generated. Then, the cost of the global rerouting path is obtained, and the global rerouting path cost is compared with the hybrid path cost. When the global rerouting path cost is less than the hybrid path cost, global rerouting is used; otherwise, the buffer segment pair is used for replacement.

[0060] In summary, this disclosure obtains link device information for each autonomous system (AS) along a first target path; this link device information characterizes the cross-domain occupancy status of the AS; based on the link device information of each AS, a first link set is determined; the first link set includes suspected conflicting links of the first target path in each AS; the suspected conflicting links are links that may be occupied; based on the first link set, at least one candidate replacement link corresponding to a suspected conflicting link is obtained; the AS is the same as the AS corresponding to the candidate replacement link; the candidate replacement link is used to replace the suspected conflicting links in the first target path to obtain a second target path. In other words, this disclosure uses the obtained link device information to predict the status of each link in the first target path in advance, obtaining suspected conflicting links that may cause problems, thereby switching paths in advance. This avoids, to a certain extent, the problem of path status misjudgment caused by delayed link device information updates, efficiently completes path selection and switching, thereby avoiding the probability of occupancy conflicts, improving path stability, and reducing extreme sensitivity to delayed updates.

[0061] When a second target path is identified, but the number of failures along this path exceeds a fourth preset threshold, it indicates a serious problem with that path. Alternatively, if the number of candidate replacement links corresponding to the suspected conflicting links is less than a fifth preset threshold, it indicates that there are very few idle links in that autonomous system (AS). In this case, it means that resources within the AS corresponding to the second target path are extremely strained, making normal transmission impossible. In this situation, links from other ASs can be used to replace links in the original AS to form a new target path for transmission.

[0062] Specifically, it may include: S105, obtaining link occupancy information for each autonomous system and determining the set of idle links corresponding to each suspected conflicting link; First, link occupancy information for all autonomous systems (AS) can be obtained. It's important to note that these ASs are not limited to those involved in the first or second target path, but rather represent the current service path occupancy status of each slice. Links that are completely unused and temporarily available for allocation within the current period are selected – the set of idle links. In other words, this disclosure is not limited to the slice type of the idle links; the target slice types corresponding to their target paths can be the same or different. The target slice here can be understood as the network used to execute the service corresponding to the target path.

[0063] S106, matching the attribute information of each link in the idle link set with the attribute information of the target slice to determine at least one backhaul candidate link; the attribute information includes at least one of the following: bandwidth quota, isolation policy, and latency level; the target slice is used to execute the service corresponding to the target path; Next, the attribute information of each link in the idle link set can be matched with the attribute information of the target slice. That is, the return candidate links need to have the same or compatible characteristics as the target slice (e.g., have the same slice / network resource area attributes). The attribute information includes, but is not limited to, at least one of the following: bandwidth quota, isolation policy, and latency level.

[0064] Optionally, borrowing restriction parameters can be attached to each backhaul candidate link, including but not limited to at least one of the following: maximum borrowing duration, maximum concurrent borrowing quantity, and available domain range. This can, to some extent, prevent borrowed services from saturating the bandwidth of the original slice and ensure that the original slice is not affected by borrowed traffic. Once the conflict is resolved or the borrowing duration expires, the borrowed backhaul candidate link can be released, and the list of available borrowed path segments can be updated.

[0065] Optionally, the topology overlap of each candidate backhaul link can be compared with that of the suspected conflicting link. Candidate backhaul links that overlap with the current suspected conflicting link on a physical link or node level can be eliminated to obtain candidate backhaul links with low conflict risk. This avoids the continued existence of conflicts after subsequent handovers and reduces repeated handovers.

[0066] It should be noted that this disclosure does not specifically restrict the execution order of topological overlap comparison, additional limiting parameters, and sorting; these can be determined according to the actual situation.

[0067] S107, Sort the candidate backhaul links to obtain the first candidate link; In multi-path selection scenarios, when the occupancy of multiple candidate paths is close or fluctuates frequently, repeated switching and routing oscillations are likely to occur, resulting in bandwidth waste, frequent reconfiguration of the forwarding plane, and a decrease in overall routing stability. The problem is further aggravated when cross-domain state perception is inconsistent.

[0068] To address the aforementioned issues, the next step is to rank the candidate backhaul links, specifically identifying the best-performing link based on attribute matching, thus determining the first candidate link. This ranking logic can include, but is not limited to, at least one of the following: latency, bandwidth, fluctuation level, and cost. It could prioritize cost optimization to reduce switching costs. It could prioritize performance optimization to ensure that service performance does not degrade after replacement and that switching stability is maximized. It could also prioritize latency to improve reachability and resource utilization without exposing domain details, alleviating scaling pressure. Multiple ranking logics can also be combined; the specific choice depends on the actual situation, and this disclosure does not impose specific limitations. In this way, the perceived occupancy across different domains becomes consistent, meaning they share the same evaluation criteria, which helps avoid frequent switching to some extent and improves path stability and resource utilization efficiency.

[0069] S108, replace the suspected conflicting link in the second target path with the first candidate link to obtain the third target path.

[0070] For ease of understanding, an example can be found by referring to... Figure 5 ,like Figure 5 As shown, we can first select link segments that are not involved in service forwarding to obtain a set of idle path segments (i.e., idle link set), which corresponds to step S105. Then, we can filter path segments with compatible attributes, which corresponds to step S106. Next, we can attach borrowing restriction parameters and then remove path segments that overlap with the topology of conflicting segments to obtain a set of path segments with ground conflict risk (i.e., the aforementioned low-conflict-risk backhaul candidate links). Next, we can sort them according to the priority of optimal latency, second-best bandwidth, and least favorable fluctuation to generate a list of borrowable path segments (i.e., the aforementioned first candidate links). Finally, after the conflict is resolved or the lease expires, we can release the first candidate links and update the list.

[0071] If, after the above replacements, the target path still fails to meet the service quality requirements of the business, then the conflicting links in the third target path can be highlighted. Subsequent link replacements will only replace these highlighted segments, retaining other normal links, thus improving replacement efficiency. Next, based on the conflicting links, a second candidate link corresponding to the conflicting link can be selected from the backflow candidate links. The autonomous system to which the second candidate link belongs is different from the autonomous system to which the first candidate link belongs. This second candidate link can be one or more. At this point, the conflicting link in the third target path can be replaced with the second candidate link, resulting in multiple candidate paths. These candidate paths are then filtered to obtain the fourth target path.

[0072] For details, please refer to Table 2 below. As shown in Table 2, the rows in the table are candidate path IDs (i.e., multiple fourth target path IDs); the columns are autonomous region IDs. Based on this table, you can obtain the link / node overlap ratio between the candidate link and the original path (i.e., the third target path) in the autonomous region.

[0073] Table 2

[0074] In the table, Ni represents the number of links between each fourth target path and the original path within the autonomous system. Mi represents the node overlap ratio between the second candidate link and the original path within the same autonomous system.

[0075] The fourth target path must meet at least the following constraints: non-overlapping links are introduced in at least one different autonomous system; the total delay is less than or equal to the preset maximum delay; and the path fluctuation level is ≤ medium level. It should be noted that the candidate paths obtained here can be one or more. That is, any path that meets the above conditions can be used as a candidate path, providing multiple possibilities for subsequent path switching. The specific path can be determined according to the actual situation.

[0076] Furthermore, when multiple candidate paths exist, they can be filtered. For example, the future volatility variance can be calculated as a stability index based on the occupancy level sequence of the past T detection cycles, using the component links of each path. The stability index can then be used as a scoring metric; a higher stability index indicates less volatility in the candidate path, making it more suitable as a switching path. It should be noted that this disclosure does not specifically limit the method of determining the scoring metric. For example, the stability index can be used alone as the scoring standard, or a weighted sum of the stability index and the value of Mi can be used to obtain the score for each candidate path. Next, the paths are sorted in descending order of score, and the top K paths are selected as the fourth target path, where K is an integer greater than 0.

[0077] Finally, the list of conflicting links, replacement locations, and execution policies in the fourth target path can be fixed to the controller configuration and distributed to the forwarding nodes of the relevant domains via PCEP or BGP-LS.

[0078] If the path still fails to meet service quality requirements after replacement, a rollback can be attempted sequentially according to the multiple candidate paths, replacement locations, and execution strategies provided: switch to the rollback path one by one according to the final rollback path sequence, quickly verifying latency and conflict issues through data analysis at each switch. If a path can be successfully mapped in the functional mapping table of this domain, switch immediately. If a conflict or failure occurs, continue trying the next rollback path.

[0079] For easier understanding, please refer to Figure 6 ,like Figure 6 As shown, we can first establish replacement location marking paths for the business paths that need to be rolled back (i.e., mark the conflicting links in the third target path). Next, we can establish a cross-domain diversity matrix, i.e., Ni / Mi. Further, we generate a preliminary candidate path set (i.e., multiple candidate paths). Then, we can calculate the stability index of each path segment based on the historical occupancy level sequence, combine the stability and diversity matrix to calculate the score of each candidate path, and sort them in descending order of score. The top K paths are taken as the final rollback path sequence (i.e., the fourth target path). Next, we attempt to replace the forwarding nodes. If the path still does not meet the service quality requirements after replacement, we try the next rollback path in sequence. If it meets the requirements, we switch to that rollback path.

[0080] When the proportion of attempted fallback paths exceeds the preset fallback proportion, such as 60% without finding a valid path, the conventional fallback mechanism can be considered to have failed. At this point, the fallback mechanism can be terminated, or a multi-branch emergency handling phase can be initiated. Specifically, this means identifying the reasons for the fallback mechanism's failure and implementing different response strategies based on those reasons. These reasons for failure include, but are not limited to, at least one of the following: domain boundary conflicts, insufficient quotas and cross-domain mutual exclusion, signaling channel blockage, Transparency and Control Summary Declaration (TSCD) reconciliation failure, and unauthorized slice attempts. Domain boundary conflicts can be understood as mutual exclusion of policies or resources during mapping of segments from different autonomous systems, preventing candidate paths from being implemented.

[0081] There are no restrictions on how to find the cause; it can be notified by the system or other devices, or you can make your own judgment based on the actual situation.

[0082] For example, please refer to the following for easier understanding: Figure 7 ,like Figure 7 As shown, the following conditions can be used to determine whether a cross-domain path switching failure is due to a domain boundary conflict: whether the most recent X rollback failures were caused by a domain boundary conflict, whether the last successful path switch was still within the same service slice, and whether the confirmation message order of the cross-domain signaling channel is normal. If all the above conditions are met, thus eliminating the two common causes of cross-domain path switch failures—slice inconsistency and abnormal signaling order—it can be inferred that the most recent rollback failure was indeed due to a domain boundary conflict. In this case, links from other domains can be invoked; links with equivalent functionality to the original link can be retrieved from the link pool, and the conflicting link can be replaced by splicing it, thereby bypassing the conflicting link. Finally, the validity check is completed in the local function mapping table, and the switch is performed.

[0083] For example, the following conditions can be used to determine whether a problem is due to insufficient quota or cross-domain exclusion: the last Y rollback failures were due to insufficient quota, cross-domain exclusion, or policy conflicts between domains (e.g., the policy label for this business slice is "prohibit topology exposure," while the policy label for the target domain is "require minimum latency"); and borrowing requests from at least two adjacent slices are already in a complementary state (e.g., one is abundant, and the other is scarce). In this case, complementary low-conflict links can be extracted from the link pools of different slices, and these low-conflict links can be temporarily spliced ​​into a virtual slice link; a new virtual ID is registered for it in the slice ID mapping table; business traffic is switched to be carried by this virtual slice link; when the conflict is resolved, the virtual ID is automatically deregistered, and the borrowed link is released. Here, a virtual slice link refers to a temporary logical channel spliced ​​from resource segments in multiple different slices, used for emergency detours.

[0084] If at least one of the following occurs: signaling channel congestion, consecutive Z rounds of TSCD reconciliation failures, or unauthorized attempts to rescind a slice, it indicates a serious anomaly or untrustworthiness of the control plane. In such cases, the shortest reachable path can be invoked within the current slice, maintaining the slice ID and creating a temporary protection channel. Traffic is switched to this path to ensure uninterrupted service. The backend control plane simultaneously marks the service as temporarily protected and continuously recalculates new compliant paths; once a suitable path is found, a smooth migration back is immediately implemented, releasing the emergency path. The temporary protection channel refers to the shortest path forwarding activated temporarily in extreme circumstances; it does not guarantee an SLA but ensures uninterrupted service. This allows for emergency rollback using the shortest path to gain survival time during signaling congestion.

[0085] It should be noted that X, Y, and Z are all integers greater than 0, and the specific numbers can be determined according to the actual situation. This disclosure does not impose specific restrictions.

[0086] After determining the target path, its legality can be verified to ensure that the path itself is legally sound and that resources are available. Legality verification includes, but is not limited to, at least one of the following: protocol / encapsulation verification and cross-domain consistency verification. Protocol / encapsulation verification includes, but is not limited to, at least one of the following: verifying whether the Multi-Protocol Label Switching (MPLS) label stack depth and SRv6SID list length are within the device's capabilities; verifying whether there are conflicts in the segment function sequence (including SRv6End.X, End.DT6, and MPLSPOP / SWAP order); and verifying whether the encapsulation budget, Maximum Transmission Unit (MTU), and latency budget meet the lower limit of the corresponding slice's Service Level Agreement (SLA). Cross-domain consistency verification verifies the cross-domain splicing point, checking whether the routing target and Slice / NRP identifier are consistent, thereby avoiding loops or data drops caused by inconsistent policies between autonomous systems.

[0087] Optionally, after passing the above verification, a path delivery mode is selected based on the actual situation, i.e., how to send relevant information to nodes in each autonomous system. This allows for the selection of a more suitable and secure delivery mode to address frequently fluctuating real-world network conditions, thereby improving system security. Specifically, this step may include: determining the path delivery mode based on the path information of the target path; the path delivery mode includes at least one of the following: general mode, satellite network mode, and privacy-constrained mode; the path information includes, but is not limited to, at least one of the following: the number of autonomous systems in the target path, link type, packet header length, device processing capacity, service information, number of links, control channel bandwidth, link latency, link occupancy fluctuations, and service requirements.

[0088] The general mode is selected if the following conditions are met: the available length of the packet header is greater than or equal to the length of the complete segment list required for the path (MPLS stack depth or SRv6SID sequence is within the hardware allowable range); the total number of links in the cross-domain path is less than or equal to the device / protocol carrying limit; and the control plane can complete the installation of the complete segment list within a single delivery delay without causing cross-domain chain traffic sway.

[0089] Satellite network mode is selected if the following conditions are met: high latency (one-way ≥25ms) satellite or satellite-terrestrial hybrid link with bandwidth less than or equal to 2Mbps; available header length less than or equal to 48 bytes; device MPLS stack or SRv6SID processing capacity less than the number of links required for the path; total number of links in the cross-domain path greater than or equal to 50 and involving more than or equal to 3 autonomous systems, while maintaining slice QoS isolation and synchronous handover. Optionally, the following conditions may also be included: the service is extremely sensitive to traffic fluctuations during the handover process; cross-domain delivery delay is close to / exceeds the service's tolerable threshold; link occupancy fluctuations exceed the threshold multiple times within the past 5 minutes.

[0090] Privacy constraint mode is selected if all of the following conditions are met: multi-carrier / MPLS cross-domain scenario, and there are legal / commercial agreement restrictions that prevent the exposure of intra-domain topology or any reversible numerical performance indicators; only encrypted control channels of less than 100kbps are available for coordination between domains; temporary detours / segment replacements are required during sudden peak periods, while ensuring the minimum guarantee of slice QoS; the number of participating domains is greater than or equal to 3.

[0091] For easier understanding, please refer to the detailed process. Figure 8 ,like Figure 8 As shown, after performing protocol / encapsulation constraints (i.e., the above protocol / encapsulation verification) and verifying cross-domain consistency routing target / Slice identifier (i.e., cross-domain consistency verification) in sequence, the path distribution mode can be selected. Depending on different conditions, the general mode, satellite network mode, and privacy constraint mode can be selected.

[0092] Specifically, when distributing via the general mode, please refer to the detailed steps. Figure 9 ,like Figure 9As shown, based on the slice identifier, the MPLS label and SRv6 SID can be bound and written into the data plane to carry a composite header structure that simultaneously carries the MPLS label and SRv6 segment identifier. If the final path segment spans multiple domains, a BindingSID is pre-configured for the next domain at the domain boundary to reduce the overhead of cross-domain distribution. Here, BindingSID refers to a special SRv6 segment identifier used to represent a sub-path. When a packet arrives at the node corresponding to this identifier, it is automatically expanded into a pre-configured link list. Next, a path version number and a complete path checksum are generated for the path. The path version number is a sequence number used to identify the path configuration version, ensuring that it is not mixed when switching between old and new paths; the checksum is calculated using a hash function to ensure the consistency and integrity of the path data during distribution. Therefore, the path distribution message includes a set of key fields: slice identifier, path version number, path checksum, and segment list. Subsequently, the path is distributed to nodes within the relevant domains via the control plane protocol. Upon receiving the path message, the node performs the following operations in sequence: First, it establishes a slice context based on the slice identifier; then, it generates forwarding table entries based on the link list and writes them to the data plane; once all relevant nodes have completed writing the table entries, it triggers path activation via a consistency confirmation message. The inbound node sets the new path as the active path while retaining the previous version as a fallback path for rapid switching in case of current path failure or performance degradation.

[0093] To illustrate the synergistic benefits and reduced control plane overhead resulting from dynamic risk tiering, buffering, and atomic replacement mechanisms, this disclosure compares its key performance indicators with the closest prior art under the same topology and traffic conditions. These indicators include path establishment latency (P95), first-collision backoff rate, number of routing oscillations, packet loss rate (P99) during handover, resource utilization, control plane message volume per change, and SLA compliance rate. Please refer to Table 3 below for details. Table 3

[0094] As shown in Table 3, this disclosure reduces P95 latency from 380ms to 240ms, first-collision backoff rate from 6.5% to 1.8%, oscillation from 4.2 times / hour to 1.1 times / hour, and P99 packet loss during handover from 0.45% to 0.18%. Resource utilization is improved to 0.74 (an increase of 19.4%), control plane message volume is reduced to 46 messages / change, and SLA compliance rate is improved to 99.1%. Dynamic risk tiering suppresses false access and secondary convergence, and buffer pair and atomic replacement convergence is single-step strong consistency, reducing the probability of inconsistency within the handover window; the increase in resource utilization and the decrease in message volume occur simultaneously, and the structured pre-arrangement reduces ineffective adjustments instead of the traditional approach of sacrificing more control plane overhead for stability.

[0095] When transmitting via satellite network, please refer to the following for details. Figure 10 ,like Figure 10 As shown, the path can be first divided into multiple intra-domain sub-paths according to the cross-domain boundary, with each group containing 5–12 segments. The specific number of 5 or 12 segments is mainly determined by the intra-domain controller based on the buffer capacity and latency budget of the devices in this domain. Each sub-path is compressed into a compressed link group table, which occupies 1–2 tag bits as the link group name. The expanded link group table of the compressed link group table is pre-stored at the domain boundary nodes and key satellite nodes: {complete segment list, execution order, validity period, conflict fallback entry}. A relative timing window [t0+Δ1, t0+Δ2] is allocated to each compressed link group table to ensure that the windows of adjacent domains do not overlap but are closely connected to achieve cross-domain synchronous switching. The timing window length can be determined based on the link round-trip latency, slice latency budget, intra-domain processing latency, and congestion history. The {compressed link group table, window} is encoded as a new data plane field and embedded in the packet header. Multiple sets of {compressed link group tables, windows} are concatenated and written to the header in path order, ensuring the total header length is ≤48 bytes. The header includes a slice ID, path version number, and expansion counter to facilitate version identification and expansion success rate statistics for each domain. When a data plane node receives a packet, it checks the first unexpanded compressed link group table and determines whether the current local clock falls within the corresponding timing window. If it does, it looks up the compressed link group table in the local expanded table, replaces it with the complete segment list, and writes it to the forwarding table. The packet is then forwarded according to the expanded path. Simultaneously, within this window, the expanded table of the next compressed link group table is preheated at the next domain boundary to improve the hit rate of the next window. Multiple domain boundary nodes expand in parallel within their respective windows, achieving cross-domain synchronous switching without requiring segment-by-segment updates from the control plane. If a local clock deviation or window congestion results in a miss, a backup compressed link group table is immediately activated, or the shortest path is returned to the next domain boundary. In detail, if the expanded table is missing, a minimum route fallback is triggered and the controller is notified to supplement the expanded table; if the link is congested, execution is automatically postponed to the next window, and the congestion event is written into the statistics to optimize subsequent window parameters. The expanded table of each compressed link group retains a keep-alive time of T after the last trigger. The new version path takes effect uniformly at the start of the next window, and the old compressed link group table automatically becomes invalid.

[0096] Furthermore, to illustrate the effectiveness of this disclosure in satellite scenarios, Table 4 below compares the performance of this disclosure and traditional solutions in terms of P95 latency, backoff rate, oscillation, P99 packet loss during handover, control plane message volume, 50-hop header overhead, and SLA.

[0097] Table 4

[0098] As shown in Table 4, this disclosure reduces P95 latency from 1200ms to 680ms, the first-collision backoff rate from 14.2% to 3.7%, oscillations from 7.8 times / hour to 1.9 times / hour, and P99 packet loss during handover from 1.2% to 0.35%. Resource utilization is improved to 78 (an increase of 74.8%), control plane message volume is reduced to 44 messages / change, and SLA compliance rate is improved to 97.8%. Thus, this disclosure significantly reduces latency, backoff, and oscillations in extreme satellite network environments, while also substantially reducing header overhead and control plane message volume.

[0099] When publishing using privacy-constrained modes, please refer to [reference needed]. Figure 11 ,like Figure 11As shown, a time-slice completion monitor can be deployed in each operator domain to monitor data plane forwarding according to the domain's time-slotted scheduling table. The monitoring parameters include at least one of the following: completion rate C-ratio, jitter window ΔJ, packet loss rate PLR, and utilization rate U. The completion rate C-ratio can be understood as the number of packets actually successfully forwarded in a time slot divided by the expected number of packets, indicating whether the domain has fulfilled its obligations; the jitter window ΔJ is the difference between the maximum and minimum latency in the time slot, reflecting latency stability; the packet loss rate PLR ​​is the number of lost packets divided by the expected number of packets, reflecting link reliability; and the utilization rate U is the actual bandwidth occupied in the time slot divided by the total available bandwidth, reflecting resource saturation. Next, the four monitored parameters can be combined into a performance vector V=[C-ratio,ΔJ,PLR,U] by the domain controller. Using the performance vector as input, a vector commitment C = VC.Commit(v;r) is constructed, where r is a random number, satisfying the following conditions: C locks the performance state of this domain, preventing modification of any sample point value without detection; and no specific value can be recovered from C externally. Then, a TSCD tuple is generated for each domain: TSCD = {TSCD tuple ID, current time slot period number, slice or service class identifier, vector commitment, zero-knowledge proof of the domain, signature using the private key of this domain}. Next, each domain sends its TSCD to the coordinator via an encrypted channel. The coordinator treats all qualified TSCDs as available edges, constructing a virtual cross-domain time slot graph, with each operator domain as a node; cross-domain links between adjacent domains are edges of the virtual cross-domain time slot graph; and the corresponding TSCD commitment value C is the edge weight of the virtual cross-domain time slot graph. The coordinator runs a commitment-driven path assembly algorithm on this virtual graph, outputting a TSCD sequence, such as TSCDA, TSCDB, TSCDC, etc., representing the end-to-end path from the source domain to the destination domain. The coordinator verifies the signatures of the TSCDs submitted by each domain, checking whether the domain signature is bound to the current time slot period number, slice, or service class identifier. If the verification passes, the domain is marked as having fulfilled its commitment in the corresponding slice of the current time slot period; otherwise, it is marked as having failed the commitment. Based on the fulfillment / failure status of multiple domains, the global cross-domain scheduling priority table is updated. Domains that fulfill their commitments are prioritized for splicing in the next time slot period, with their priority increased. Domains that fail to fulfill their commitments have their splicing priority reduced in the next time slot period, and the number of failures is recorded. Domains that fail to fulfill their commitments for a consecutive period reach a threshold are marked as low-trust and will only participate in splicing if there are no other available domains. The coordinator distributes the TSCD sequence to the boundaries of each relevant domain. Each relevant domain boundary performs a local reverse lookup, that is, it locates the original commitment vector V and the corresponding link through the TSCD_ID; this is mapped to the Rv6SID sequence or MPLSLabelStack within the domain. Each domain installs the SegmentList obtained from the reverse lookup into its forwarding table.For SRv6, the SID list needs to be written to the final hardware entry set used for packet forwarding; for MPLS, LabelStack needs to be loaded and ingress / egress forwarding rules configured. Installation must be completed before the start of the time slot period to which the TSCD belongs; otherwise, the TSCD will be invalid. After installation, each domain sends a very small reconciliation digest to the coordinator. If the reconciliation is inconsistent with the aforementioned commitment, the coordinator confirms it as a breach of commitment and immediately broadcasts a breach announcement. Other domains will automatically lower the priority of this domain in the next time slot period accordingly. If consecutive breaches reach a threshold, the coordinator removes this domain from the candidate domain order and triggers a multi-branch rollback.

[0100] In particular, for scenarios involving multi-carrier interconnection and subject to privacy constraints, this disclosure tests the differences in general performance indicators with traditional technologies and adds sensitive field exposure to measure the degree of exposure of private topology / performance parameters in collaborative scheduling.

[0101] The tests were conducted in a simulated multi-carrier cross-domain network. The core environmental parameters were as follows: Network topology, comprising three Autonomous Systems (AS) belonging to different carriers, each consisting of 10 core node routers, interconnected via links of specific bandwidth. Basic conditions included a basic inter-domain link latency of 20-40ms and a bandwidth load of 50% to simulate real network congestion. Stress testing included: the first-collision backoff rate test under high-concurrency service requests (1000 path establishment requests / second); and the route oscillation count test, which introduced 5% random link performance fluctuations (simulating occupancy changes) into the network. SLA definition: SLA compliance was measured based on three key performance indicators: latency (<100ms), jitter (<5ms), and bandwidth (100M), with a test duration of 24 hours. Privacy assessment: The sensitive field exposure metric measured the degree to which private information (such as topology details and precise resource values) needed to be exposed to other domains during path coordination, with 1 indicating complete exposure and 0 indicating no exposure.

[0102] The test results are shown in Table 5 below: Table 5

[0103] As shown in Table 5, this disclosure was conducted in a typical complex environment simulating multi-carrier cross-domain interconnection. This environment included three autonomous systems, each with 10 core nodes, and a base latency of 20-40ms and a bandwidth load of 50% to simulate real network pressure. In this test environment, this disclosure demonstrated several significant benefits: First, in terms of performance, it reduced path establishment P95 latency by 38.9%, and significantly reduced conflict backoff rate and route oscillation count by 70.1% and 72.5%, respectively, effectively improving service response speed and network stability. Second, in terms of security and efficiency, it reduced control plane message overhead by 42.5%, and most importantly, reduced sensitive field exposure by 90%, achieving cross-domain secure collaboration without disclosing almost any private topology and resource details. Finally, in terms of service assurance, these improvements collectively increased the SLA compliance rate by 5.5 percentage points, reaching 98.5%, proving that it can provide efficient and secure network connections for critical services without sacrificing performance or availability for privacy and security.

[0104] In summary, this disclosure constructs a virtual cross-domain time slot graph using TSCD vector commitment and zero-knowledge verification, enabling commitment-driven path assembly and priority orchestration. This achieves end-to-end path delivery and reconciliation penalties without leakage of numerical values ​​and topology, breaking through the dependence of traditional cross-domain TE / PCE on visible topology and high-bandwidth control channels. Furthermore, this disclosure significantly enhances the inherent resilience of IP / MPLS networks through multi-dimensional detection of cross-domain occupancy status, fluctuation level classification, and a rapid buffer segment switching mechanism. It proactively avoids path conflicts and oscillations caused by dual resource occupancy and state delays, greatly enhancing the network's self-stabilization and rapid convergence capabilities under unstable conditions. Simultaneously, for special scenarios such as high latency in satellite networks and privacy constraints between operators, compressed concatenated timing headers and zero-knowledge verification technology are adopted to ensure reliable synchronization of control signaling under extreme conditions and achieve cross-domain path collaboration without leaking topology details, thereby constructing a more stable, reliable, and secure underlying network infrastructure. At the business security level, this disclosure directly empowers the continuity of business operations and data privacy security. Its collaborative processing mechanism and multi-branch emergency rollback capability provide a highly available channel for critical businesses, significantly reducing the risk of interruption and convergence latency, and ensuring business continuity. More importantly, through zero-knowledge batch verification technology, it can strictly adhere to data compliance requirements in multi-carrier collaborations, achieving end-to-end secure path scheduling that prevents data from leaving the domain. This perfectly solves the challenges of privacy protection and secure transmission of sensitive business data in cross-border and cross-carrier scenarios, providing a solid guarantee for the secure and stable operation of core businesses in complex network environments.

[0105] It should be noted that the above examples can be combined according to the actual situation, and this disclosure does not impose specific restrictions.

[0106] For example, please refer to Figure 12 ,like Figure 12 As shown, the process first checks for cross-domain occupancy in each domain. If found, suspected conflicting links are identified. If the occupancy rate of a suspected conflicting link remains high or very high for two consecutive periods, or if the occupancy rate jumps from low to very high within a single period, the cross-domain occupancy conflict avoidance process is triggered. Next, based on the list of suspected conflicting links, the list of avoided paths, and the list of remaining paths with high-risk links, it is determined whether the number of available links for the remaining paths is below a preset threshold, or whether the number of repeated failures using the same segment has exceeded two. If so, the pre-established list of cross-slice borrowable path links is searched, and external slice path links that do not overlap with the current conflicting links in the domain and have compatibility attributes are selected and replaced. For paths that still do not meet the service quality requirements, K fallback paths are generated to attempt a fallback. Finally, the final path (i.e., the aforementioned fourth target path) undergoes legality verification, and the path distribution method is selected.

[0107] This disclosure also provides a path acquisition device. Figure 13 This is a schematic diagram of a path acquisition device provided in an embodiment of the present disclosure, such as... Figure 13 As shown, the path acquisition device 1300 includes: The information acquisition module 1301 is configured to acquire link device information of each autonomous system in the first target path; the link device information is used to characterize the cross-domain occupancy of the autonomous system. The set determination module 1302 is configured to determine a first set of links based on the link device information of each autonomous system; the first set of links includes suspected conflicting links of the first target path in each autonomous system; the suspected conflicting links are links that may be occupied. The link acquisition module 1303 is configured to acquire at least one candidate replacement link corresponding to a suspected conflict link based on the first link set; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system. The path acquisition module 1304 is configured to replace the suspected conflicting links in the first target path with the candidate replacement links to obtain the second target path.

[0108] In one illustrative example, the path acquisition device 1300 is further configured such that the link device information includes at least one of the following: link utilization, queue latency, packet loss rate, ratio of instantaneous peak to average traffic, and slice isolation status; before determining the first link set based on the link device information of each autonomous system, the method further includes: determining occupancy intensity information based on the link device information; the occupancy intensity information includes at least one of the following: occupancy rate change, cross-domain coupling, boundary conservatism, current occupancy level, and autonomous system rating information; determining whether there is cross-domain occupancy in each autonomous system based on a first preset condition; if so, determining the first link set based on the link device information of the autonomous systems with cross-domain occupancy; wherein, the first preset condition includes at least one of the following: the number of link segments in the autonomous system that meet the second preset condition meets a first quantity threshold, and the number of times the critical link segments in the autonomous system meet the second preset condition meets a second quantity threshold; the second preset condition includes at least one of the following: the comprehensive rating information of the autonomous system is greater than the first preset threshold, the occupancy change rate is greater than the second preset threshold, and the stability coefficient is less than the third preset threshold.

[0109] In one illustrative example, the path acquisition device 1300 is further configured to determine the first link set based on the link device information of the autonomous system with cross-domain occupancy, including: determining the link occupancy strength of each link in the autonomous system based on the link device information of the autonomous system with cross-domain occupancy; determining whether the link occupancy strength is greater than the dynamic conflict threshold; if so, determining the link as a suspected conflict link to obtain the first link set.

[0110] In one illustrative example, the path acquisition device 1300 is further configured to: acquire link occupancy information of each autonomous system when the number of failures of the second target path exceeds a fourth preset threshold, or when the number of candidate replacement links corresponding to the suspected conflicting links is less than a fifth preset threshold; determine the set of idle links corresponding to each suspected conflicting link; match the attribute information of each link in the set of idle links with the attribute information of the target slice to determine at least one backhaul candidate link; the attribute information includes at least one of the following: bandwidth quota, isolation policy, latency level; the target slice is used to execute the service corresponding to the target path; sort the candidate backhaul links to obtain a first candidate link; and replace the suspected conflicting links in the second target path with the first candidate link to obtain a third target path.

[0111] In one illustrative example, the path acquisition device 1300 is further configured to, when the third target path does not meet the quality of service requirements, also include: marking the links in the third target path that are in conflict; Based on the conflicting link, a second candidate link corresponding to the conflicting link is selected from the backflow candidate links; the autonomous system to which the second candidate link belongs is different from the autonomous system to which the first candidate link belongs; the conflicting link in the third target path is replaced by the second candidate link to obtain the fourth target path.

[0112] In one illustrative example, the path acquisition device 1300 is further configured to determine a path delivery mode based on the path information of the target path; the path delivery mode includes at least one of the following: general mode, satellite network mode, and privacy constraint mode; The path information includes at least one of the following: number of autonomous systems in the target path, link type, header length, device processing capacity, service information, number of links, control channel bandwidth, link latency, link occupancy fluctuation, and service requirements.

[0113] Figure 14 This is a hardware block diagram of an electronic device provided according to an embodiment of the present disclosure. The electronic device 1400 according to an embodiment of the present disclosure includes at least a processor; and a memory for storing computer-readable instructions. When the computer-readable instructions are loaded and executed by the processor, the processor performs the path acquisition method described in any of the preceding embodiments of the present disclosure.

[0114] Figure 14 The illustrated electronic device 1400 specifically includes a central processing unit (CPU) 1401, a graphics processing unit (GPU) 1402, and a memory 1403. These units are interconnected via a bus 1404. The CPU 1401 and / or GPU 1402 can function as the aforementioned processor, and the memory 1403 can function as the aforementioned memory storing computer-readable instructions. Furthermore, the electronic device 1400 may also include a communication unit 1405, a storage unit 1406, an output unit 1407, an input unit 1408, and an external device 1409, all of which are also connected to the bus 1404.

[0115] Figure 15 This is a schematic diagram of a computer-readable storage medium provided according to an embodiment of this disclosure. (As shown) Figure 15As shown, a computer-readable storage medium 1500 according to an embodiment of the present disclosure stores computer-readable instructions 1501 thereon. When the computer-readable instructions 1501 are executed by a processor, the path acquisition method described with reference to the above figures according to any embodiment of the present disclosure is performed. The computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0116] This disclosure further provides a computer program product, including a computer program that, when executed by a processor, implements the path acquisition method described in any of the preceding embodiments of this disclosure.

[0117] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0118] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0119] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are intended to require or imply that connections, arrangements, or cards must be made in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, or carded in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0120] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0121] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0122] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0123] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is intended to be limited to the aspects shown herein, but rather to be distributed within the widest scope consistent with the principles and novel features disclosed herein.

[0124] The above description has been given for purposes of illustration and description. Furthermore, this description is intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A path acquisition method, characterized in that, The method includes: Obtain link device information for each autonomous system (AS) along the first target path; the link device information is used to characterize the cross-domain occupancy status of the AS; Based on the link device information of each autonomous system, a first link set is determined; the first link set includes suspected conflict links of the first target path in each autonomous system; the suspected conflict links are links that may be occupied. Based on the first link set, at least one candidate replacement link corresponding to a suspected conflict link is obtained; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system. The suspected conflicting links in the first target path are replaced with the candidate replacement links to obtain the second target path.

2. The method according to claim 1, characterized in that, The link device information includes at least one of the following: link utilization, queue latency, packet loss rate, ratio of instantaneous peak to average traffic, and slice isolation status; before determining the first link set based on the link device information of each autonomous system, the method further includes: Based on the link device information, occupancy intensity information is determined; the occupancy intensity information includes at least one of the following: occupancy rate change, cross-domain coupling degree, boundary conservatism, current occupancy level, and autonomous system rating information; Based on the first preset condition, determine whether there is cross-domain occupation in each autonomous domain; If so, the first set of links is determined based on the link device information of the autonomous domain that has cross-domain occupation; The first preset condition includes at least one of the following: the number of link segments in the autonomous system that meet the second preset condition meets a first quantity threshold, and the number of times the critical link segments in the autonomous system meet the second preset condition meets a second quantity threshold. The second preset condition includes at least one of the following: the comprehensive score information of the autonomous region is greater than the first preset threshold, the occupancy change rate is greater than the second preset threshold, and the stability coefficient is less than the third preset threshold.

3. The method according to claim 2, characterized in that, The step of determining the first set of links based on link device information of autonomous systems with cross-domain occupancy includes: Based on the link device information of the autonomous system with cross-domain occupancy, determine the link occupancy strength of each link in the autonomous system; Determine whether the link occupancy strength is greater than the dynamic conflict threshold; If so, the link is identified as a suspected conflict link, thus obtaining the first link set.

4. The method according to claim 1, characterized in that, When the number of failures of the second target path exceeds a fourth preset threshold, or when the number of candidate replacement links corresponding to the suspected conflicting link is less than a fifth preset threshold, the method further includes: Obtain link occupancy information for each autonomous system and determine the set of idle links corresponding to each suspected conflicting link; The attribute information of each link in the idle link set is matched with the attribute information of the target slice to determine at least one backhaul candidate link; the attribute information includes at least one of the following: bandwidth quota, isolation policy, and latency level; the target slice is used to execute the service corresponding to the target path; The candidate backhaul links are sorted to obtain the first candidate link; The first candidate link is used to replace the suspected conflicting link in the second target path to obtain the third target path.

5. The method according to claim 4, characterized in that, When the third target path does not meet the quality of service requirements, the method further includes: Mark the links in the third target path that are in conflict; Based on the conflicting link, a second candidate link corresponding to the conflicting link is selected from the backflow candidate links; the autonomous system to which the second candidate link belongs is different from the autonomous system to which the first candidate link belongs. The fourth target path is obtained by replacing the conflicting links in the third target path with the second candidate links.

6. The method according to claim 1, characterized in that, The method further includes: Based on the path information of the target path, the path delivery mode is determined; the path delivery mode includes at least one of the following: general mode, satellite network mode, and privacy constraint mode; The path information includes at least one of the following: number of autonomous systems in the target path, link type, header length, device processing capacity, service information, number of links, control channel bandwidth, link latency, link occupancy fluctuation, and service requirements.

7. A path acquisition device, characterized in that, The device includes: The information acquisition module is configured to acquire link device information of each autonomous system (AS) along the first target path; the link device information is used to characterize the cross-domain occupancy of the AS. The set determination module is configured to determine a first set of links based on the link device information of each autonomous system; the first set of links includes suspected conflicting links of the first target path in each autonomous system; the suspected conflicting links are links that may be occupied. The link acquisition module is configured to acquire at least one candidate replacement link corresponding to a suspected conflict link based on the first link set; the suspected conflict link and the corresponding candidate replacement link are located in the same autonomous system. The path acquisition module is configured to replace the suspected conflicting links in the first target path with the candidate replacement links to obtain the second target path.

8. An electronic device, characterized in that, include: Memory, used to store computer-readable instructions; as well as A processor for executing the computer-readable instructions, causing the electronic device to perform the method as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium for storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by a processor, the processor performs the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-6.

Images