The Electric Field Controlling Method Based on GM (1,1) RBF Neural Network

Author

Hao, Tan ; Lu, Xiong ; Shenguang, Gong

Author_Institution

Dept. of Weaponry Eng., Naval Univ. of Eng., Wuhan, China

fYear

2012

fDate

11-13 Aug. 2012

Firstpage

2177

Lastpage

2180

Abstract

To avoid detection or attack, electric field signal characteristic of ships should be controlled. The signal data is decomposed into low frequency component and high frequency component first. The high frequency component is predicted by RBF neural network, the low frequency component is predicted by GM(1,1) model and add up both of the values predicted. Then countercurrent is exported and then the electric signal is weakened. Based on the data got from sea, the simulations show that 70% of shaft-rate (SR) signal amplitude would be weakened by this method, and static electric field signal is almost eliminated at all.

Keywords

electric fields; radial basis function networks; ships; signal detection; signal processing; GM (1,1)_RBF neural network; SR signal amplitude; attack avoidance; countercurrent; detection avoidance; electric field controlling method; electric field signal characteristic; electric signal; high frequency component; low frequency component; shaft-rate signal amplitude; ships; signal data; static electric field signal; Corrosion; Electric fields; Marine vehicles; Mathematical model; Neural networks; Predictive models; Training; 1) model; Electric field; GM (1; RBF neural network; Signal character controlling; Wavelet Decomposition;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science & Service System (CSSS), 2012 International Conference on

Conference_Location

Nanjing

Print_ISBN

978-1-4673-0721-5

Type

conf

DOI

10.1109/CSSS.2012.541

Filename

6394859