一种改进的小波滤噪算法及其在NMRL中的应用

本文针对核磁共振测井（NMRL）回波信号微弱且信号带宽较大的特点,将小波变换良好的时频局部化分析性能以及优良的噪声抑制能力与相敏检波算法相结合,设计并实现了一种基于小波阈值降噪的数字相敏检波（DPSD）算法。仿真结果表明,与FIR式DPSD算法相比,小波式DPSD算法在低孔隙度（T2＊较小）的情况下具有较低的峰值误差和良好的噪声抑制能力,在一定程度上解决了滤除噪声与保留信号完整性之间的矛盾,适用于低孔隙度地层的核磁共振测井应用。

An Improved De-noising Arithmetic Based on Wavelet Transform and its Application to NMRL

The paper aimed at the echo signal of Nuclear Magnetic Resonance Logging（NMRL）,which has the character of weak and wide bandwidth,combined Phase Sensitive Detector with wavelet transform,which has the efficient time-frequency localization and excellent performance of de-noising,designed and implemented a Digital Phase Sensitive Detector（DPSD） based on wavelet threshold de-noising.Simulation results show that wavelet DPSP has lower peak relative error and better de-noising on low porosity（T2＊ less） compared with the FIR DPSD.The wavelet DPSP solves the conflict between de-noising and keeping the integrality of the signal to some extent,and is suitable for Nuclear Magnetic Resonance Logging in low porosity stratum.