Journal of Terahertz Science and Electronic Information Technology , Volume. 22, Issue 7, 703(2024)
Information feature association analysis based on historical data of off-line electronic countermeasure
In view of the high requirements of electronic countermeasures signal analysis, the large amount of intelligence processing data, and the difficulty of multi-dimensional information analysis and extraction in complex electromagnetic environment, the big data processing method of cross-industry data mining standard process, namely Cross-Industry Standard Process for Data Mining(CRISP-DM), is studied, and an processing and analysis platform for historical data of offline electronic countermeasures electromagnetic signal is designed based on the characteristics of the massive historical reconnaissance data of the electronic countermeasures intelligence system and the existing electromagnetic target knowledge base. By exploring the technical path of mining electromagnetic target parameters, time rules of the target, multi-target association rules and other intelligence analysis, the application of analysis methods such as clustering, the mining of time series and association rules in the processing of electronic countermeasure electromagnetic signal historical data is realized. The information features of unknown electronic target such as clustering, target quantity and scale prediction, multi-target association and cooccurrence rule are analyzed and obtained. The results show that the characteristics and correlation laws of electromagnetic target parameters are obvious, and the fitting correlation degree of target time characteristics reaches 0.825. This work lays a foundation for further research and practical application.
Get Citation
Copy Citation Text
LU Yaxiao, ZHOU Changlin, WANG Haisong, WANG Yicheng, LIU Guangyi, YU Daojie. Information feature association analysis based on historical data of off-line electronic countermeasure[J]. Journal of Terahertz Science and Electronic Information Technology , 2024, 22(7): 703
Category:
Received: Mar. 24, 2023
Accepted: --
Published Online: Aug. 22, 2024
The Author Email: Yaxiao LU (378033213@qq.com)