An application of expert system in recognition of radiated noise of underwater target

Author

Li, Qihu ; Wang, Jinlin ; Wei, Wei

Author_Institution

Inst. of Acoust., Acad. Sinica, Beijing, China

Volume

1

fYear

1995

fDate

9-12 Oct 1995

Firstpage

404

Abstract

An expert system, named “EXPLORE”, has previously been developed. It is used to identify radiated noise of underwater targets, including surface ships, submarines, high speed gunboats, cargo, ambient noise and fishing vessels. The theory of cluster analysis and fuzzy logic are applied in this system. A multiple feature extraction technique of the signal is derived, which can best describe the characteristics of the signature of underwater noise sources. A special metric is established in Hilbert space of the feature vector, by which the decision is deducted. The experimental test for this system, including several hundred samples, shows that the EXPLORE system is intelligent in recognition of underwater targets, even in relative small signal to noise ratios

Keywords

acoustic noise; decision theory; expert systems; feature extraction; fuzzy logic; military computing; ships; sonar target recognition; telecommunication computing; EXPLORE; Hilbert space; ambient noise; cargo; cluster analysis; expert system; feature vector; fishing vessels; fuzzy logic; high speed gunboats; multiple feature extraction technique; radiated noise; recognition; signal to noise ratios; submarines; surface ships; underwater target; Expert systems; Feature extraction; Fuzzy logic; Hilbert space; Intelligent systems; Marine vehicles; Signal to noise ratio; System testing; Target recognition; Underwater vehicles;

fLanguage

English

Publisher

ieee

Conference_Titel

OCEANS '95. MTS/IEEE. Challenges of Our Changing Global Environment. Conference Proceedings.

Conference_Location

San Diego, CA

Print_ISBN

0-933957-14-9

Type

conf

DOI

10.1109/OCEANS.1995.526801

Filename

526801