Multi-stage network reconstruction recovery method based on structure perception

Abstract

The invention discloses a multi-stage network reconstruction recovery method based on structure perception, which comprises the following steps of: identifying a path set influenced by a broken link in a network topological graph; generating a candidate path set for each influenced path in the network topological graph after the broken link is removed; for each candidate path, calculating the comprehensive structure similarity of the candidate path and the corresponding original path; calculating an end-to-end task success rate of each candidate path based on the real-time failure rate and path transmission delay of nodes and edges by considering a communication degradation factor; for each affected path, the candidate path with the maximum optimization objective function value is selected from the corresponding candidate path set to serve as a reconstruction path, a reconstruction result is output, and the optimization objective function comprises the comprehensive structure similarity. The method does not need training, can be completely explained and audited, and has remarkable engineering landing advantages in combat or key infrastructure scenes with extremely high requirements on safety and reliability.

Description

Technical Field

[0001] This invention relates to the field of complex network analysis and intelligent decision-making technology, and in particular to a multi-stage network reconstruction and recovery method based on structure awareness. Background Technology

[0002] With the expansion of information and networked systems, the stability and vulnerability of complex networks have attracted widespread attention from academia and industry. Identifying the "critical nodes" and "weak links" in complex networks is not only related to the overall robustness and reliability of the system, but also directly affects the continuity and efficiency of the task chain.

[0003] The current state of research and application is mainly reflected in the following directions: (1) Single index centrality method: Complex network analysis often uses metrics such as degree centrality, betweenness centrality, and proximity centrality as important criteria for identifying critical nodes. In 2001, Brandes proposed a fast betweenness centrality algorithm, reducing computational complexity to O(|V||E|), making large-scale network analysis possible. This type of method is widely used in communication networks, transportation networks, social networks, and other scenarios.

[0004] (2) Network robustness and percolation theory perspective: Complex network theory reveals that scale-free networks are robust to random failures but extremely vulnerable to targeted attacks (such as removing highly central nodes). This conclusion forms an important theoretical basis for critical node identification, and related research results have been widely disseminated on platforms such as Nature and arXiv.

[0005] (3) Critical Node Detection Problem (CNDP): CNDP models the problem of "finding the set of nodes that most severely degrades network connectivity" as a combinatorial optimization problem. It has various algorithmic solutions, such as integer programming and heuristics, and has been applied in the protection of critical infrastructures such as power, transportation, and communications.

[0006] (4) Network efficiency indicators: Latora and Marchori proposed a network efficiency metric that uses the reciprocal of the average shortest path to characterize the overall information transmission efficiency. This method provides a reference for evaluating network performance at the macro level and has been adopted in multiple network performance analysis scenarios.

[0007] While the above methods provide important insights for studying the vulnerability of complex networks, they still have the following limitations: 1. Structure-oriented, lacking task-oriented: Both single-index centrality and percolation theory focus on the importance of network topology or global connectivity, but fail to incorporate the probability of task achievability in specific information chains, making it difficult to reflect the role of nodes in actual task execution.

[0008] 2. Lack of node functionality and health status modeling: Most existing methods assume that nodes are in an ideal working state and do not consider the real-time performance degradation or failure probability of nodes, which leads to differences between the evaluation results and the actual performance.

[0009] 3. Ignoring the impact of environmental factors: Environmental conditions such as distance, latency, and interference have a significant impact on link reliability, but most methods do not directly model these factors.

[0010] 4. Insufficient real-time performance: CNDP and global efficiency metrics are typically computationally complex and suitable for static analysis, making them unsuitable for real-time vulnerability identification in dynamic environments. Summary of the Invention

[0011] To address the technical problems existing in the prior art, this invention proposes a structure-aware multi-stage network reconstruction and recovery method that reconstructs the communication chain from the information source to the task terminal in real time while maintaining the semantic and structural features of the original chain.

[0012] To achieve the above objectives, the present invention provides a structure-aware multi-stage network reconstruction and recovery method, comprising: In the network topology graph, identify the set of paths affected by broken links; In the network topology graph after removing broken links, a set of candidate paths is generated for each affected path. Each candidate path must satisfy length constraints, type quota constraints, and stage constraints based on the current scenario. For each candidate path, calculate the comprehensive structural similarity with the corresponding original path; Considering communication degradation factors, the end-to-end task success rate of each candidate path is calculated based on the real-time failure rate of nodes and edges and the path transmission delay. For each affected path, the candidate path with the largest optimization objective function value is selected from the corresponding candidate path set as the reconstruction path, and the reconstruction result is output. The optimization objective function includes the comprehensive structural similarity.

[0013] Preferably, generating the candidate path set includes: The candidate path set is generated by employing a depth-first search or iterative deepening search, and setting upper limits on the path enumeration depth and number; wherein, a prefix pruning strategy based on path length, node type and scenario constraints is applied during the search process.

[0014] Preferably, the comprehensive structural similarity with the corresponding original path is calculated as follows: ; In the formula, Candidate paths Compared to the original path The overall structural similarity; This is the original path; Candidate paths; is the node-based Jaccard similarity coefficient, representing the degree of overlap between nodes in the path; The edge-based Jaccard similarity coefficient represents the degree of overlap between links in the path; Similarity in path length; Similarity of node type distribution; These are the weighting coefficients for the four similarity indicators.

[0015] Preferably, when calculating the optimization objective, a penalty term is introduced to maintain specific key links, and the corrected similarity score is: ; In the formula, For the original comprehensive structural similarity, This represents the penalty coefficient for the critical link. The set of links for the original path. For the key strike link set, This is the set of links for candidate paths.

[0016] Preferably, the end-to-end task success rate for each candidate path is calculated: ; In the formula, The end-to-end task success rate for each candidate path. For node failure rate, For edge failure rate, For delay efficiency factor, v For nodes on the path, For links on the path.

[0017] Preferably, the optimization objective function is a multi-objective weighted function, specifically: ; In the formula, To allow for the use of regularization terms based on weak values ​​or congestion, For the original comprehensive structural similarity, For structural similarity terms, For task success rate, The weight coefficient for the regularization penalty term. This represents the end-to-end task success rate of the candidate paths.

[0018] Preferably, the reconstruction result includes: The selected alternative path, the overall structural similarity between the alternative path and the original path, the path success rate, and the link difference set between the old and new paths, wherein the difference set identifies the added, removed, and maintained links.

[0019] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a structure-aware multi-stage network reconstruction and recovery method.

[0020] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a structure-aware, multi-stage network reconstruction and recovery method.

[0021] Compared with the prior art, the present invention has the following advantages and technical effects: (1) This invention takes the comprehensive structural similarity index as its core objective, comprehensively considering the overlap between nodes and edges, the similarity of path lengths, and the consistency of node type distribution to ensure that the alternative path closely resembles the original path in both form and essence. Compared to simply taking the shortest path or a local heuristic, by adding a penalty term for attack link protection, the model actively avoids the destruction of key operational semantics, and the output results are more reusable and more in line with tactical rules. The system also provides the difference between the old and new links (addition / removal / preservation), making the degree of semantic preservation quantifiable, visual, and auditable.

[0022] (2) In complex electromagnetic environments or adversarial situations, distance and latency significantly reduce the success rate of communication. To address this, this invention employs a segmented failure rate model to characterize the effects of distance attenuation and latency accumulation, and uses a latency efficiency factor to attenuate the path success rate, thus reflecting a preference for "reachable and stable" paths during the candidate path evaluation stage. The optional multi-objective form (balancing similarity and success rate) achieves a practical balance between semantic preservation and reliability, making it suitable for strategy switching and parameter tuning in different scenarios.

[0023] (3) This invention uses scenario / stage constraints as the constraint surface, supporting both strong constraints (elimination if not met) and soft constraints (imposing costs through penalty items), and provides a unified expression for critical attack links, path length limits, node type quotas, etc. The output includes not only alternative paths, but also constraint satisfaction rate, penalty trigger rate, and critical link consistency report, which facilitates compliance auditing and retrospective tracking from the engineering and management levels.

[0024] (4) This invention is implemented in a modular / configurable manner. The type set, quota, stage constraints and weights can all be quickly adjusted, and the failure rate model can also be replaced according to industry scenarios (such as wireless, satellite, shipborne, industrial IoT, etc.). In terms of engineering combination, it can be seamlessly integrated with the k-shortest path algorithm (Yen / Eppstein). The other is responsible for generating diverse candidates, while this invention is responsible for unified constraint filtering and structural similarity scoring, which not only retains the advantages of the algorithm, but also reduces access costs and migration costs. Attached Figure Description

[0025] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a structure-aware multi-stage network reconstruction and recovery method according to an embodiment of the present invention. Detailed Implementation

[0026] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0027] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0028] This embodiment relates to robust communication and self-healing reconstruction of task chains in complex networks, especially for military / industrial command and control networks. Under adverse conditions such as link failure and communication degradation (distance + delay), it reconstructs the communication chain from the information source to the task terminal in real time while maintaining the semantic and structural characteristics of the original chain.

[0029] A structure-aware, multi-stage network reconstruction and recovery method, such as Figure 1 ,include: In the network topology graph, identify the set of paths affected by broken links; In the network topology graph after removing broken links, a set of candidate paths is generated for each affected path. Each candidate path must satisfy length constraints, type quota constraints, and stage constraints based on the current scenario. For each candidate path, calculate the comprehensive structural similarity with the corresponding original path; Considering communication degradation factors, the end-to-end task success rate of each candidate path is calculated based on the real-time failure rate of nodes and edges and the path transmission delay. For each affected path, the candidate path with the largest optimization objective function value is selected from the corresponding candidate path set as the reconstruction path, and the reconstruction result is output. The optimization objective function includes the comprehensive structural similarity.

[0030] Specifically, this embodiment proposes a link reconstruction method driven by structural similarity, aware of communication degradation, and consistent with task constraints: identifying the set of affected original paths; enumerating restricted candidates on the edge-de-edge graph; selecting alternative paths based on structural similarity scores; introducing communication degradation to evaluate the success rate; and outputting the reconstructed path, similarity, and statistics.

[0031] Furthermore, generating the candidate path set includes: The candidate path set is generated by employing a depth-first search or iterative deepening search, and setting upper limits on the path enumeration depth and number; wherein, a prefix pruning strategy based on path length, node type and scenario constraints is applied during the search process.

[0032] Specifically, the original set of valid paths (restricted simple paths): ; in, Indicates type quota constraints (e.g., the count for any type does not exceed 3); This is the original set of valid paths. This is the upper limit of the path length. As the source node, For the destination node set, This refers to the specific destination node in the set of destination nodes.

[0033] Affected path set ; In the formula, For broken links in the network, For path The set of links.

[0034] Network topology diagram after removing broken links The above is aimed at start and end points Generate candidates: ; in, For scenario / stage constraints, For the candidate path set, Candidate paths, The allowable path length deviation threshold, Candidate paths Type quota constraints.

[0035] Furthermore, let's denote the set of nodes and edges: , Path length .

[0036] definition: ; ; ; ; In the formula, The Jaccard similarity coefficient is based on nodes. The Jaccard similarity coefficient is based on edges. For path length similarity, Similarity of node type distribution For type In the path The number of times it appears in the text.

[0037] The overall structural similarity with the corresponding original path is calculated as follows: ; In the formula, Candidate paths Compared to the original path The overall structural similarity; This is the original path; Candidate paths; is the node-based Jaccard similarity coefficient, representing the degree of overlap between nodes in the path; The edge-based Jaccard similarity coefficient represents the degree of overlap between links in the path; Similarity in path length; Similarity of node type distribution; These are the weighting coefficients for the four similarity indicators.

[0038] Furthermore, when calculating the optimization objective, a penalty term is introduced to maintain specific key links, and the corrected similarity score is: ; In the formula, For the original comprehensive structural similarity, This represents the penalty coefficient for the critical link. The set of links for the original path. For the key strike link set, This is the set of links for candidate paths.

[0039] Furthermore, the node-based failure rate Minimum distance between the missile and the nearest aircraft carrier extremely low radius Constrained radius ,Delay (Second).

[0040] failure rate : ; ; Delay efficiency factor : ; In the formula, For the node's basic failure rate, The increase in failure rate due to distance. The increase in failure rate due to delay, For the restricted radius, For the key distance parameter, It has an extremely low radius.

[0041] Calculate the end-to-end task success rate for each candidate path: ; In the formula, The end-to-end task success rate for each candidate path. For node failure rate, For edge failure rate, v For nodes on the path, For links on the path.

[0042] Furthermore, optimization of a single affected path: ; Multiple objectives: ; In the formula, To allow for the use of regularization terms based on weak values ​​or congestion, For the original comprehensive structural similarity, For structural similarity terms, For task success rate, The weight coefficient for the regularization penalty term. This represents the end-to-end task success rate of the candidate paths.

[0043] Furthermore, in Top Enumerate simple paths and their depth limits. Maximum quantity ; Prioritize using restricted DFS / iterative deepening, and backtrack to the shortest path as a safety net when necessary; Prior caching: coordinates, failure rate, reachability, and partial path prefix scores.

[0044] Specifically, constraint checks include: Chain breakage elimination: ; Scenario / Phase: Phase 5 Maintain constant; Type Quota: ; Maximum length: ; Accessibility: for A simple and effective path.

[0045] Furthermore, rating and selection: calculate Jointly Weighted; When the highest equal division is achieved, Use a two- or three-order criterion based on path length; Optimal output And similarity, differential link sets (add / remove / keep).

[0046] Furthermore, the right is to be spatially approximated. ,pass control; score ; Caching significantly reduces redundant calculations.

[0047] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the structure-aware multi-stage network reconstruction and recovery method described above.

[0048] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the structure-aware multi-stage network reconstruction and recovery method described above.

[0049] This embodiment employs restricted k-path generation and prefix pruning (length / type / scenario), with its time complexity controlled to O(b{}{K{}}), while the effective branch coefficient is significantly reduced through the pruning coefficient p. Combined with caching of 3D coordinates, failure rate, reachability, and prefix scores, redundant calculations and ineffective expansions are avoided. Even on networks with thousands of nodes or larger scales, it maintains a reconstruction decision latency of seconds to sub-seconds, meeting the requirements for online monitoring, emergency response, and real-time drills.

[0050] This embodiment uses a standardized evaluation protocol: semantic preservation output. Mean / variance; reliability output The performance curve follows {r,R,I,Rr}; in terms of compliance, the statistical constraint satisfaction rate and the consistency rate of the attack chain are measured; in terms of efficiency, the number of candidates, time consumption, cache hit and pruning rate are reported; and the independent contribution of components such as penalty items, type quotas and length constraints are quantified through ablation experiments to form an evidence chain that can be directly used for review and production.

[0051] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A structure-aware, multi-stage network reconstruction and recovery method, characterized in that, include: In the network topology graph, identify the set of paths affected by broken links; In the network topology graph after removing broken links, a set of candidate paths is generated for each affected path. Each candidate path must satisfy length constraints, type quota constraints, and stage constraints based on the current scenario. For each candidate path, calculate the comprehensive structural similarity with the corresponding original path; Considering communication degradation factors, the end-to-end task success rate of each candidate path is calculated based on the real-time failure rate of nodes and edges and the path transmission delay. For each affected path, the candidate path with the largest optimization objective function value is selected from the corresponding candidate path set as the reconstruction path, and the reconstruction result is output. The optimization objective function includes the comprehensive structural similarity.

2. The structure-aware multi-stage network reconstruction and recovery method according to claim 1, characterized in that, Generating the candidate path set includes: The candidate path set is generated by employing a depth-first search or iterative deepening search, and setting upper limits on the path enumeration depth and number; wherein, a prefix pruning strategy based on path length, node type and scenario constraints is applied during the search process.

3. The structure-aware multi-stage network reconstruction and recovery method according to claim 1, characterized in that, The overall structural similarity with the corresponding original path is calculated as follows: ； In the formula, Candidate paths Compared to the original path The overall structural similarity; This is the original path; Candidate paths; is the node-based Jaccard similarity coefficient, representing the degree of overlap between nodes in the path; The edge-based Jaccard similarity coefficient represents the degree of overlap between links in the path; Similarity in path length; Similarity of node type distribution; These are the weighting coefficients for the four similarity indicators.

4. The structure-aware multi-stage network reconstruction and recovery method according to claim 3, characterized in that, When calculating the optimization objective, a penalty term is introduced to maintain specific key links. The corrected similarity score is: ； In the formula, For the original comprehensive structural similarity, This represents the penalty coefficient for the critical link. The set of links for the original path. For the key strike link set, This is the set of links for candidate paths.

5. The structure-aware multi-stage network reconstruction and recovery method according to claim 1, characterized in that, Calculate the end-to-end task success rate for each candidate path: ； In the formula, The end-to-end task success rate for each candidate path. For node failure rate, For edge failure rate, For delay efficiency factor, v For nodes on the path, For links on the path.

6. The structure-aware multi-stage network reconstruction and recovery method according to claim 1, characterized in that, The optimization objective function is a multi-objective weighted function, specifically: ； In the formula, To allow for the use of regularization terms based on weak values ​​or congestion, For the original comprehensive structural similarity, For structural similarity terms, For task success rate, The weight coefficient for the regularization penalty term. This represents the end-to-end task success rate of the candidate paths.

7. The structure-aware multi-stage network reconstruction and recovery method according to claim 1, characterized in that, The reconstruction results include: The selected alternative path, the overall structural similarity between the alternative path and the original path, the path success rate, and the link difference set between the old and new paths, wherein the difference set identifies the added, removed, and maintained links.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the structure-aware multi-stage network reconstruction and recovery method as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the structure-aware multi-stage network reconstruction and recovery method as described in any one of claims 1 to 7.

Images