关键词: 滑坡变形监测　, 小波阈值　, MEMS 传感器　, 粒子群优化

Abstract: Aiming at the random noise problem of MEMS sensor signal in landslide deep deformation monitoring,an improved
wavelet threshold denoising method based on particle swarm optimization(PSO) algorithm is proposed. By introducing
PSO algorithm to optimize the key parameters in the improved soft and hard threshold compromise function,the effective denoising
of the monitoring signal is realized. The improved algorithm can suppress the random noise in the signal,and can better
retain the detailed information of the signal,improve the quality and reliability of the signal. Compared with the traditional hard
threshold and soft threshold denoising methods,the proposed method has a significant improvement in signal-to-noise ratio,
reaching more than twice the original signal-to-noise ratio,and the root mean square error has also been significantly reduced,
indicating that the method can recover the signal more accurately. Through the experimental analysis of the selected bior3. 3
wavelet basis function,the superiority and practical application value of this method in the deep deformation monitoring of landslide
are further verified. The research results show that this method not only has strong noise suppression ability,but also can
retain more signal details,which is suitable for the field of geological disaster monitoring such as landslide. It provides a new idea
for MEMS sensor signal processing,and has strong universality and promotion value.

Key words: landslide deformation monitoring,wavelet threshold,MEMS sensor,particle swarm optimization