Optical Fiber Vibration Signal Denoising based on Improved EEMD Wavelet Threshold

Optical fiber vibration sensing systems have been widely applied in complex scenarios such as structural health monitoring and environmental detection, due to their high sensitivity, fast response, and wide detection range. However, in realworld applications, vibration signals are often heavily affected by strong environmental noise. This issue becomes particularly pronounced under harsh conditions where the signal-to-noise ratio (SNR) falls below 10 dB, leading to a significant degradation in signal quality. Moreover, due to the diverse and complex nature of vibration signals, different types of signals exhibit varying degrees of sensitivity to noise interference. To address these challenges, this paper proposed a joint denoising algorithm based on an improved Ensemble Empirical Mode Decomposition (EEMD) and wavelet threshold denoising. The proposed method first employs EEMD to adaptively decompose the signal into multiple Intrinsic Mode Functions (IMFs) and then reconstructs the useful IMFs using a new designed reconstruction strategy. Finally, wavelet thresholding is applied to the reconstructed signal for secondary denoising. Experimental results demonstrate that the proposed method achieves superior denoising performance on four typical vibration signals, providing a better balance between noise suppression and feature preservation.