A Modulation Type Recognition Method Using Wavelet Support Vector Machines

Author

Liu, Hai-Yuan ; Sun, Jian-Cheng

Author_Institution

Coll. of Inf. Tech. Sci., Nankai niversity, Tianjing, China

fYear

2009

fDate

17-19 Oct. 2009

Firstpage

1

Lastpage

4

Abstract

Soft defined radio is again the research issue because of cognitive radio, Modulation type recognition (MTR) is the key issue of the soft defined radio, in this paper, a new MTR method based on the wavelet support vector machine (WSVM) has been proposed, we derive the WSVM kernel function and utilize it to classify the modulation type, the results show when the SNR threshold for the modulation scheme classifier is approximately 10 dB. The correct recognizing rate exceeds 91%. It is shows that this method is better than the tradition method.

Keywords

cognitive radio; modulation; pattern classification; signal classification; software radio; support vector machines; telecommunication computing; wavelet transforms; SNR threshold; Soft defined radio; WSVM kernel function; cognitive radio; modulation scheme classifier; modulation type recognition; wavelet support vector machine; Arithmetic; Cognitive radio; Educational institutions; Finance; Kernel; Lagrangian functions; Neural networks; Support vector machine classification; Support vector machines; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing, 2009. CISP '09. 2nd International Congress on

Conference_Location

Tianjin

Print_ISBN

978-1-4244-4129-7

Electronic_ISBN

978-1-4244-4131-0

Type

conf

DOI

10.1109/CISP.2009.5301396

Filename

5301396