Research on Optimization of Large-Scale Heterogeneous Combat Network Based on Graph Embedding
In the era of information warfare, the combat system-of-systems consists of interconnected entities that can be abstracted as heterogeneous combat network (HCN). Developing HCNs with exceptional performance is crucial to building effective combat system-of-systems. Currently, large-scale research on...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10829954/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In the era of information warfare, the combat system-of-systems consists of interconnected entities that can be abstracted as heterogeneous combat network (HCN). Developing HCNs with exceptional performance is crucial to building effective combat system-of-systems. Currently, large-scale research on HCN suffers from significant time costs and poor convergence results. To address these problems, a graph embedding-based Heterogeneous Combat Network Optimization Framework (HCNOF) is proposed, which optimizes HCN features and reduces time overhead. In this paper, we design crossover operators and mutation operators for HCNs to iteratively search for the optimal HCN. Additionally, we developed a heterogeneous graph embedding method for HCNs to transform the high-dimensional HCN representation matrix into a low-dimensional embedding matrix. Finally, we designed experiments to optimize the functional robustness of HCN, verifying the effectiveness of HCNOF. The results demonstrate that, in the graph embedding task, the MAP value of HCNOF is higher than several conventional graph embedding methods, indicating better preservation of network structure. In the optimization task, HCNOF demonstrates superior performance, particularly in terms of convergence speed, which is a measure of optimization efficiency. It has shown a 17.23% improvement in this aspect compared to HCNOF0 and a 6.89% enhancement when compared to the GE-SU-EANet algorithm, underscoring its advantages in both optimization performance and cost reduction. Overall, our research findings motivate the establishment of stronger combat systems and provide an innovative approach for large-scale HCN optimization. |
|---|---|
| ISSN: | 2169-3536 |