Research on statistical characteristics of engine backward
RCS based on K-KDE algorithm

doi:10.19682/j.cnki.1005-8885.2023.2014

中国邮电高校学报(英文) ›› 2023, Vol. 30 ›› Issue (4): 33-42.doi: 10.19682/j.cnki.1005-8885.2023.2014

• Artificial Intelligence • 上一篇下一篇

Research on statistical characteristics of engine backward RCS based on K-KDE algorithm

Fu Li, Cui Zhe, Deng Hongwei

1. School of Automation, Shenyang Aerospace University, Shenyang 110136, China 2. Avic Shenyang Engine Research Institute, Shenyang 110015, China

收稿日期:2022-09-01 修回日期:2023-06-28 接受日期:2023-08-31 出版日期:2023-08-31 发布日期:2023-08-31
通讯作者: Fu Li, E-mail: ffulli@163.com E-mail:ffulli@163.com

Research on statistical characteristics of engine backward RCS based on K-KDE algorithm

Fu Li, Cui Zhe, Deng Hongwei

1. School of Automation, Shenyang Aerospace University, Shenyang 110136, China 2. Avic Shenyang Engine Research Institute, Shenyang 110015, China

Received:2022-09-01 Revised:2023-06-28 Accepted:2023-08-31 Online:2023-08-31 Published:2023-08-31
Contact: Fu Li, E-mail: ffulli@163.com E-mail:ffulli@163.com

摘要/Abstract

摘要： In order to solve the problem of low accuracy of traditional fixed window width kernel density estimation (KDE) in radar cross section (RCS) statistical characteristics analysis, an improved Epanechnikov KDE (K-KDE) algorithm was proposed to analyze the statistical characteristics of the engine's backward RCS. Firstly, the K-nearest neighbor method was used to calculate the dynamic window width of the K-KDE, and the Euclidean distance of each adjacent sample was used to judge the local density of the sample, and then the window width of the kernel function was adjusted by the distance between the sample point and the nearest neighbor to complete the KDE. Secondly, based on the K-KDE and the traditional KDE algorithm, the cumulative probability density function (CPDF) of four RCS random distribution sample points subject to fixed parameters was calculated. The results showed that the root mean square error of the K-KDE was reduced by 31.2%, 38.8%, 38.1% and 31.9% respectively compared with the KDE. Finally, the K-KDE combined with the second generation statistical analysis models were used to analyze the statistical characteristics of the engine backward RCS.

关键词: radar cross section, K-nearest neighbor method, kernel density estimation, fluctuation characteristics

Abstract: In order to solve the problem of low accuracy of traditional fixed window width kernel density estimation (KDE) in radar cross section (RCS) statistical characteristics analysis, an improved Epanechnikov KDE (K-KDE) algorithm was proposed to analyze the statistical characteristics of the engine's backward RCS. Firstly, the K-nearest neighbor method was used to calculate the dynamic window width of the K-KDE, and the Euclidean distance of each adjacent sample was used to judge the local density of the sample, and then the window width of the kernel function was adjusted by the distance between the sample point and the nearest neighbor to complete the KDE. Secondly, based on the K-KDE and the traditional KDE algorithm, the cumulative probability density function (CPDF) of four RCS random distribution sample points subject to fixed parameters was calculated. The results showed that the root mean square error of the K-KDE was reduced by 31.2%, 38.8%, 38.1% and 31.9% respectively compared with the KDE. Finally, the K-KDE combined with the second generation statistical analysis models were used to analyze the statistical characteristics of the engine backward RCS.

Key words: radar cross section, K-nearest neighbor method, kernel density estimation, fluctuation characteristics

Fu Li, Cui Zhe, Deng Hongwei. Research on statistical characteristics of engine backward RCS based on K-KDE algorithm[J]. The Journal of China Universities of Posts and Telecommunications, 2023, 30(4): 33-42.

参考文献

References

[1] ZHANG W, WANG G Y, ZENG Y H, et al. Research on dynamic RCS distribution characteristics of aircraft targets. Journal of Radio Science, 2010, 25(1): 117 - 121 (in Chinese).

[2] GAO Z H, WANG M L. An efficient algorithm for calculating aircraft RCS based on the geometrical characteristics. Chinese Journal of Aeronautics, 2008, 21(4): 296 - 303.

[3] SWERLING P. Radar probability of detection for some additional fluctuating target cases. IEEE Transactions on Aerospace and Electronic Systems, 2007, 33(2): 698 - 709.

[4] ZHUANG Y Q, WANG Y Z, LIU J L. Research on statistical characteristic of stealth UAV. Journal of Microwave, 2020, 36(S1): 60 - 62 (in Chinese).

[5] ZHANG B, ZHANG Y, LU W W, et al. A study on fully polarimetric RCS statistical characteristics of fixed-wing UAV in Ku Band. Modern Radar, 2020, 42(6): 41 - 47 (in Chinese).

[6] FU L, JIANG G W, HUANG Q J. Statistical characteristics analysis based on F/A-XX fighter using adapative kernel density estimation algorithm. Journal of Shanghai Jiaotong University (Science), 2022, DOI: 10.1007/s12204-022-2467-9.

[7] CHEN S C, HUANG P L, JI J Q. Radar cross section fluctuation characteristics of typical stealth aircraft. Acta Aeronautica et Astronautica Sinica, 2014, 35(12): 3304 - 3314.

[8] DENG H W, SHANG S T, JIN H, et al. Analysis and discussion of aeroengine stealth technology. Aeronautical Science and Technology, 2017, 28(10): 1 - 7 (in Chinese).

[9] LAPKO A V. Modified fast algorithm for the bandwidth selection of the kernel density estimation. Optoelectronics, Instrumentation and Data Processing, 2020, 56(6): 566 - 572.

[10] SHI M, YIN R, WANG Y F, et al. Photovoltaic power interval forecasting method based on kernel density estimation algorithm. IOP Conference Series: Earth and Environmental Science, Vol 615: Proceedings of the 2020 International Conference on Green Development and Environmental Science and Technology (ICGDE'20), 2020, Sept 18 - 20, Changsha, China. Bristol, UK: IOP Publishing, 2020: Article 012062.

[11] Kotlarski I. On characterizing the Chi square distribution by the student law. Journal of the American Statistical Association, 1966, 61(316): 976 - 981.

[12] GEORGE K, DIMITRIOS G, KOUDAS N, et al. Efficient biased sampling for approximate clustering and outlier detection in large data sets. IEEE Transactions on Knowledge and Data Engineering, 2003, 15(5): 1170 - 1187.

[13] SWERLING P. Radar probability of detection for some additional fluctuating target cases. IEEE Transactions on Aerospace and Electronic Systems, 1997, 33(2): 698 - 709.

[14] JENSSEN R. Indefinite Parzen window for spectral clustering. Proceedings of the 2007 IEEE Workshop on Machine Learning for Signal Processing, 2007, Aug 27 - 29, Thessaloniki, Greece. Piscataway, NJ, USA: IEEE, 2007: 390 - 395.

[15] YANDELL B S. Kernel smoothing. Technometrics, 1996, 38(1): 75 - 76.

[16] ZHUANG Y Q, ZHANG C X, ZHANG X K, et al. Statistical analysis and simulation of typical stealth aircraft dynamic RCS. Journal of Microwave, 2014, 30(5): 17 - 21 (in Chinese).

[17] SUSMITA P, PRADIPTA K N. Kernel density estimation and correntropy based background modeling and camera model parameter estimation for underwater video object detection. Soft Computing—A Fusion of Foundations, Methodologies and Applications, 2021(15): 10477 - 10496.

[18] ZHANG Q L, WANG S T. A novel rule-centered fuzzy model by Epanechnikov mixtures. Journal of Computer Research and Development, 2010, 47(5): 940 - 947 (in Chinese).

[19] LI X C, HOU Y. Indoor positioning technology based on improved AP selection and KNN algorithm. Journal of Computer Applications, 2017, 37(11): 3276 - 3280 (in Chinese).

[20] CHEN S Y, CHEN T T. Dockless shared bicycle flow control by using kernel density estimation based clustering. Advances in Technology Innovation, 2021, 6(3): 146 - 156.

[21] TIAN S, ZENG L L. Fast charging behavior of electric taxi based on improved kernel density estimation. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(4): 221 - 229 (in Chinese).

[22] FU Y M. Communication signal detection and recognition technology based on goodness of fit test. Ph D Thesis. Changsha, China: National University of Defense Technology, 2015 (in Chinese).

[23] MA X. Anomaly detection method for viewing datasets based on improved DBSCAN algorithm and K-S test. Master Thesis. Beijing, China: Beijing University of Posts and Telecommunications, 2021 (in Chinese).

[24] ZHAO H. Research on edge distribution in copula distribution estimation algorithm based on K-S test. Master Thesis. Taiyuan, China: Taiyuan University of Science and Technology, 2013 (in Chinese).

[25] XIE X, XU Z G. Blind watermark detection based on K-S test in radio-frequency signals. Electronics Letters, 2020, 56(1): 30 - 32.

Research on statistical characteristics of engine backward RCS based on K-KDE algorithm

Research on statistical characteristics of engine backward RCS based on K-KDE algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价