Online Streaming Feature Selection Based on Conditional Information Entropy

2017 ◽  
Author(s):  
Huaming Wang ◽  
Guoyin Wang ◽  
Xianhua Zeng ◽  
Siyuan Peng
Keyword(s):  
Feature Selection ◽  
Information Entropy ◽  
Conditional Information ◽  
Online Streaming

OFS-Density: A novel online streaming feature selection method

2019 ◽  
Vol 86 ◽  
pp. 48-61 ◽  
Author(s):  
Peng Zhou ◽  
Xuegang Hu ◽  
Peipei Li ◽  
Xindong Wu
Keyword(s):  
Feature Selection ◽  
Feature Selection Method ◽  
Selection Method ◽  
Online Streaming

Modulation Recognition of Digital Multimedia Signal Based on Data Feature Selection

2017 ◽  
Vol 8 (3) ◽  
pp. 90-111 ◽  
Author(s):  
Hui Wang ◽  
Li Li Guo ◽  
Yun Lin
Keyword(s):  
Feature Selection ◽  
Information Entropy ◽  
Feature Subset ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Modulation Recognition ◽  
Signal Modulation ◽  
Digital Multimedia ◽  
Optimal Feature Subset ◽  
Optimal Feature

Automatic modulation recognition is very important for the receiver design in the broadband multimedia communication system, and the reasonable signal feature extraction and selection algorithm is the key technology of Digital multimedia signal recognition. In this paper, the information entropy is used to extract the single feature, which are power spectrum entropy, wavelet energy spectrum entropy, singular spectrum entropy and Renyi entropy. And then, the feature selection algorithm of distance measurement and Sequential Feature Selection(SFS) are presented to select the optimal feature subset. Finally, the BP neural network is used to classify the signal modulation. The simulation result shows that the four-different information entropy can be used to classify different signal modulation, and the feature selection algorithm is successfully used to choose the optimal feature subset and get the best performance.

Sign in / Sign up

Export Citation Format

Share Document