摘要
本文成功地将小波变换的多尺度信号分解（ＭＲＳＤ）方法用于一维核磁共振（ＮＭＲ）信号的数据压缩。将一幅由３２７６８个数据点构成的ＮＭＲ谱图压缩至５１２个数据点，仍能在均方差很小（２．７７６９×１０－４）的情况下很好地重建原始谱图。
The multiresolution signal decomposition(MRSD) algorithm based on wavelet transform was applied to the compression of 1 D NMR spectra.A NMR spectrum which was composed of 32 768 data points was investigated and the spectrum can be completely reconstructed with the residual square deviation(RSD) being as small as 2 776 9×10 -4 when the spectrum was compressed to 512 data points.The relationship among the RSD and the different wavelet function,number of decomposition and the compression ratio was also discussed.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
1999年第2期139-141,共3页
Spectroscopy and Spectral Analysis
关键词
数据压缩
小波变换
核磁共振谱
Data compression, Wavelet transform, NMR spectra
