Skip to content

This repository demonstrates how to accurately recognize the different types of acoustic PD signals in noisy environments. The results are very interesting; the proposed method achieves classification accuracies in the range 99.33%-100.00% for noise-corrupted data of SNR -40:30dB.

Notifications You must be signed in to change notification settings

ramyh/Robust_PD_Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Robust_PD_Detection

RR stands for Robust Regression Here, we have three Matlab codes:- (1) RR_NoisyPD_PosSNR: Applies Robust Regression to the spectrum of the noisy PD signals of Positive SNRs. (2) RR_NoisyPD_0SNR: Applies Robust Regression to the spectrum of the noisy PD signals of 0dB SNR. (3) RR_NoisyPD_NegSNR: Applies Robust Regression to the spectrum of the noisy PD signals of Negative SNRs.

The Matlab matrices named "SpectrumNPD_AWGN_Pos_20", "SpectrumNPD_AWGN_Pos_10", "SpectrumNPD_AWGN_0", "SpectrumNPD_AWGN_Neg_10". ...., and "SpectrumNPD_AWGN_Neg_40" include the frequency spectrum of the noisy PD signals corrupted with Additive White Gaussian Noise of the SNRs: 20dB, 10dB, 0dB, -10dB, ....., and -40dB, respectively.

The CSV files named"SpectrumNPD_AWGN_Pos_20_Sparse", "SpectrumNPD_AWGN_Pos_10_Sparse", "SpectrumNPD_AWGN_0_Sparse", "SpectrumNPD_AWGN_Neg_10_Sparse". ...., and "SpectrumNPD_AWGN_Neg_40_Sparse" include the EXTRACTED SPECTRAL FEATURES of the noisy PD signals of the

About

This repository demonstrates how to accurately recognize the different types of acoustic PD signals in noisy environments. The results are very interesting; the proposed method achieves classification accuracies in the range 99.33%-100.00% for noise-corrupted data of SNR -40:30dB.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages