Please use this identifier to cite or link to this item:
Title: Prediction of underwater acoustic signals based on ESMD and ELM
Authors: Li, Guohui
Wang, Siliang
Yang, Hong
Keywords: Underwater acoustic signals;Extreme-point symmetric mode decomposition;Extreme learning machine;Prediction
Issue Date: Mar-2019
Publisher: NISCAIR-CSIR, India
Abstract: The local predictability of underwater acoustic signals plays an important role in underwater acoustic signal processing, as it is the basis for solving non-stationary signal detection. A prediction model of underwater acoustic signals based on extreme-point symmetric mode decomposition (ESMD) and extreme learning machine (ELM) is proposed. First, underwater acoustic signals are decomposed by ESMD to obtain a set of intrinsic model functions (IMFs). After IMFs are grouped, the training samples and forecast samples are obtained. Then, prediction model for training samples is established by using ELM to obtain the input layer, output layer weight vector and offset matrix. The trained ELM is used to predict the forecast sample to obtain component. Finally, the reconstructed IMFs and residuals are the final prediction results. The experimental results show that the proposed model is a good predictive model having better prediction accuracy and smaller error.
Page(s): 357-362
ISSN: 0975-1033 (Online); 0379-5136 (Print)
Appears in Collections:IJMS Vol.48(03) [March 2019]

Files in This Item:
File Description SizeFormat 
IJMS 48(3) 357-362.pdf440.95 kBAdobe PDFView/Open

Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.