Weak signal detection in chaotic noise background - based on VMD-EEMD and self-attention mechanisms DOI Open Access
Shengli Zhao,

Yuanyuan Wu,

Xingxing Jiang

et al.

Journal of Physics Conference Series, Journal Year: 2024, Volume and Issue: 2851(1), P. 012015 - 012015

Published: Sept. 1, 2024

Abstract In this paper, a method for detection and estimation of weak harmonic signals in chaotic noise background is proposed. Firstly, the observed signal smoothed by local weighted regression model to reduce influence noise. Then, are decomposed into intrinsic mode functions Variational Mode Decomposition Ensemble Empirical (VMD-EEMD). The components reconstructed phase space, one-step prediction carried out adding layer self-attention mechanism Long Short-Term Memory (LSTM). error obtained summing up results will be detected from error. Finally, periodogram used detect whether contain or not. If signals, amplitude estimated least squares method. Simulation experiments show that proposed paper has good performance. single signal, Mean Absolute Error (MAE) Square (MSE) 0.0212 0.00068 respectively; MAE MSE 0.001977 0.0007 respectively. multiple 0. 0248 0006 0035 1.681×10 −5 able efficiently backgrounds estimate signals.

Language: Английский

Weak signal detection in chaotic noise background - based on VMD-EEMD and self-attention mechanisms DOI Open Access
Shengli Zhao,

Yuanyuan Wu,

Xingxing Jiang

et al.

Journal of Physics Conference Series, Journal Year: 2024, Volume and Issue: 2851(1), P. 012015 - 012015

Published: Sept. 1, 2024

Abstract In this paper, a method for detection and estimation of weak harmonic signals in chaotic noise background is proposed. Firstly, the observed signal smoothed by local weighted regression model to reduce influence noise. Then, are decomposed into intrinsic mode functions Variational Mode Decomposition Ensemble Empirical (VMD-EEMD). The components reconstructed phase space, one-step prediction carried out adding layer self-attention mechanism Long Short-Term Memory (LSTM). error obtained summing up results will be detected from error. Finally, periodogram used detect whether contain or not. If signals, amplitude estimated least squares method. Simulation experiments show that proposed paper has good performance. single signal, Mean Absolute Error (MAE) Square (MSE) 0.0212 0.00068 respectively; MAE MSE 0.001977 0.0007 respectively. multiple 0. 0248 0006 0035 1.681×10 −5 able efficiently backgrounds estimate signals.

Language: Английский

Citations

0