随钻核磁共振自旋回波数据联合编码压缩方法

CPMG spin echo data compression method based on joint encoding for NMR-LWD

一维随钻核磁共振CPMG(Carr purcell meiboom gill)自旋回波数据量大,而井下仪器存储容量较小,为了在有限的存储空间内保存更多的数据信息,结合CPMG自旋回波数据呈多指数变化的特点,提出了离散余弦变换(discrete cosine transform,DCT)、游程编码(run-length encoding,RLE)和霍夫曼编码相结合的联合编码压缩方法。首先采用连续光栅扫描法将一维CPMG自旋回波数据重组为二维矩阵,增强数据间的相关性;然后采用DCT变换去除数据冗余;DCT系数经均匀量化后进行游程编码,增强连续重复数据的压缩效率;最后由霍夫曼编码器进一步去除编码冗余。利用仿真数据进行了压缩实验,结果表明,在相对误差不超过5%的前提下,压缩比可以达到15∶1,表明提出的方法能够在较好地保留信号特征的前提下有效压缩信号。与现有方法相比,联合编码方法更适合井下CPMG自旋回波数据的压缩应用。

One-dimensional nuclear magnetic resonance logging while drilling( NMR-LWD) will produce a gigantic quantity of CPMG spin echo data,but underground storage capability is relatively limited. In order to store more information under limited memory space,a joint encoding compression method is proposed,which integrates DCT transform,RLE and entropy coding together based on the multiexponential characteristic of CPMG spin echo data. Firstly,a one-dimensional CPMG spin echo vector is reorganized into a twodimensional matrix by continuous raster scanning to enhance data correlation. Then,DCT transform is adopted to remove the data redundancy. The quantified DCT coefficients are encoded by RLE for improving the compression efficiency of continuous repetitive data.Finally,further encoder redundancy is removed through Huffman coding. The experimental results with simulation data show that the compression ratio of the proposed method can achieve up to 15 ∶ 1 with relative error less than 5%,which can compress CPMG data efficiently with important features well preserved. Compared with existing compression methods,the overall performance of joint encoding is more suitable for compressing CPMG spin echo data.