DocumentCode :
1972775
Title :
A novel method using GA-based Clustering and spectral features for modulation classification
Author :
Ebrahimzadeh, Ataollah ; Hossienzadeh, Mahdi
Author_Institution :
Fac. of Electr. & Comput. Eng., Babol Univ. of Technol., Babol, Iran
fYear :
2011
fDate :
16-18 Sept. 2011
Firstpage :
4705
Lastpage :
4708
Abstract :
Because of rapid growing of radio communication technology of late years, importance of monitoring of radio waves is rising increasingly. Automatic radio signal types recognition is an important topic for both the civil and military domains. This paper proposes a high efficient technique for recognition of seven digital modulations. This technique is a pattern recognition approach. In this technique we have used the spectral characteristics for extraction the efficient features. A reduced set of parameters is derived from these coefficients and used as input to a GA-Clustering technique. The simulation results show that the proposed algorithm has high recognition accuracy to discriminate the considered radio signals.
Keywords :
genetic algorithms; pattern clustering; signal classification; GA-based clustering; automatic radio signal types recognition; digital modulations; modulation classification; pattern recognition approach; radio communication technology; radio waves; spectral features; Biological cells; Classification algorithms; Clustering algorithms; Feature extraction; Genetic algorithms; Modulation; Signal to noise ratio; clustering; genetic algorithm; modulation classification; spectral features;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical and Control Engineering (ICECE), 2011 International Conference on
Conference_Location :
Yichang
Print_ISBN :
978-1-4244-8162-0
Type :
conf
DOI :
10.1109/ICECENG.2011.6057019
Filename :
6057019
Link To Document :
بازگشت