A New Method of Combined Classifier Design Based on Fuzzy Integral and Support Vector Machines

摘要：

To make the modulation classification system more suitable for signals in a wide range of signal noise rate (SNR), a novel method of designing combined classifier based on fuzzy integral and multi-class support vector machines (MSVM) is presented in this paper. The method employs multi-class support vector machines classifiers and fuzzy integral to improve recognition reliability. Experimental results illustrate that the proposed combined classifier has high ecognition rate with large variation range of SNR (success rates are over 98.2％ when SNR is not lower than 5dB).

作者: Kexin Jia Youxin Lu

作者单位: University of Electronic Science and Technology of China Chengdu, China

会议类型: 国际会议

会议名称: 2007年通信、电路与系统国际会议(2007 International Conference on Communications,Circuits and Systems Proceedings)

会议地点: 日本福冈

会议语种:英文

在线出版日期: 2007-07-11（万方平台首次上网日期，不代表论文的发表时间）