Research Article
Wireless Network Topology Discovery Based on Spectrum Data by Convolutional Neural Network
@INPROCEEDINGS{10.1007/978-3-031-65126-7_34, author={Xinfeng Deng and Zhihui Xie and Li Zhou}, title={Wireless Network Topology Discovery Based on Spectrum Data by Convolutional Neural Network}, proceedings={Quality, Reliability, Security and Robustness in Heterogeneous Systems. 19th EAI International Conference, QShine 2023, Shenzhen, China, October 8 -- 9, 2023, Proceedings, Part I}, proceedings_a={QSHINE}, year={2024}, month={8}, keywords={Spectrum Data Communication Relationship Network Topology}, doi={10.1007/978-3-031-65126-7_34} }- Xinfeng Deng
Zhihui Xie
Li Zhou
Year: 2024
Wireless Network Topology Discovery Based on Spectrum Data by Convolutional Neural Network
QSHINE
Springer
DOI: 10.1007/978-3-031-65126-7_34
Abstract
Wireless network topology can reflect the communication relationships among network node. Since there are significant challenges and difficulties in deciphering the communication contents, spectrum data is adopted to discover communication relationships and network topology of a wireless network. In this paper, we propose a wireless network topology discovery method based on spectrum data to determine the communication relationships of nodes. Since the spectrum data features of nodes are correlated during the communication process, we construct the wireless network topology by mining the communication behaviors of nodes from the spectrum data features based on maximum similarity and hierarchical clustering. Simulation results demonstrate that the proposed method can achieve a better performance of hierarchical clustering than the existing methods.
