MA(q)模型参数的估计——一种混合相位子波反演算法
|
摘要
笔者在文中讨论了混合相位子波A的反演问题.利用高阶累积矩获得近似反子波B(0),再通过B(0)与地震记录S褶积生成数据S=B(0)*S,最后对模型S=C(0)*r采用统计递归算法SRA[14]得到介入子波C(0)=(C,C*).真实子波A的估计通过解方程C=B(0)*A得到.如果混合相位子波A的长度有限,我们的算法可获得相当满意的结果.
In the paper,an inverse problem on mixed-phase wavelet A is discussed.An approximate reverse wavelet B~((0)) is obtained by using higher-order accumulate moments,data S is generated by S=B~((0))*S,and a middle wavelet C~((0))=(C,C~*) is obtained through using SRA arithmetic for the model S=C~((0))*r.Estimation on the true wavelet A is obtained by solution for the equation C=B~((0))*A.If the length of mixed-phase wavelet A is finite,a quite good result can be obtained by the arithmetic in the paper.
In the paper,an inverse problem on mixed-phase wavelet A is discussed.An approximate reverse wavelet B~((0)) is obtained by using higher-order accumulate moments,data S is generated by S=B~((0))*S,and a middle wavelet C~((0))=(C,C~*) is obtained through using SRA arithmetic for the model S=C~((0))*r.Estimation on the true wavelet A is obtained by solution for the equation C=B~((0))*A.If the length of mixed-phase wavelet A is finite,a quite good result can be obtained by the arithmetic in the paper.
引文
[1]BROCKWELL P J and DAVIS R A.Time Series:Theory and Methods[M].Springer-Verlag worldPublishing Corporation,1987.
[2]TUGNAIT J K.Estimation of Linear ParametricModels Using Inverse Filter Criteria and Higher Or-der Statistics[J].IEEE Trans.Signals Proc,1993,41(11):3196-3199.
[3]TUGNAIT J K.Parameter Estimation for NoncausalARMA Models of Non-Gaussian Signals via CumulantMatching[J].IEEE Trans.On Signal Proc.,1995,43(4):886-893.
[4]SHALVI O.and WEINSTEIN E.New Criteria forBlind Deconvolution of nonminimum Phase Systems(Channels)[J].IEEE Trans.on Information Theory,1990,36(2):312-321.
[5]CADZOW J A.Blind Deconvolution via Cumulant Ex-trema[M].IEEE Signal Proc.Magazine,1996:24-42.
[6]KUNDUR D.and HATZINAKOS D.Blind ImageDecovolution[M].IEEE signal Proc.Magazine,1996:43-64.
[7]CHEN C H.New Cumulant-Based Inverse FilterCriteia for Deconvolution of Non-minimum PhaseSystem[J].IEEE Trans.Signal Proc.,1996,44(5):1292-1297.
[8]MENDEL J M.Tutorial on Higher_Order Statistics(Spectra)in Signal Processing and System Theory:Theoritical Results and Some Applications[J].Proc.IEEE,1991,79(3):278-305.
[9]LAZEAR G D.Mixed-phase wavelet estimation usingfourth-order cumulants[J].Geophysics,1993,58(7):1042-1049.
[10]GIANNAKJS G B,MENDEL J M.and XiaofengZhao,A Fast Prediction_Error Detecter for Estima-tion Sparse_Spike Sequences[J].IEEE Trans.ongeoscience and remote sensing,1989,27(3):344-351.
[11]WANG L X,etc.Adaptive minimum prediction-errordeconvolution and source wavelet estimation usingHopfield neural networks[J].Geophysics,1992,57(4):670-675.
[12]Li_Xin Wang and MENDEL J M.Adaptive minimumprediction_error deconvolution and source wavelet es-timation using Hopfield neural networks[J].Geo-physics,1992,57(4):670-679.
[13]HU M K.Visual Pattern Recognition by Moment In-variant[J].IRE Trans.Inform.Theory,1962,8:179-187.
[14]胡必锦,汪达成,雷鸣.褶积模型参数估计的递归算法及其收敛性[J].西南师范大学学报,2005,30(4):633-637.
[2]TUGNAIT J K.Estimation of Linear ParametricModels Using Inverse Filter Criteria and Higher Or-der Statistics[J].IEEE Trans.Signals Proc,1993,41(11):3196-3199.
[3]TUGNAIT J K.Parameter Estimation for NoncausalARMA Models of Non-Gaussian Signals via CumulantMatching[J].IEEE Trans.On Signal Proc.,1995,43(4):886-893.
[4]SHALVI O.and WEINSTEIN E.New Criteria forBlind Deconvolution of nonminimum Phase Systems(Channels)[J].IEEE Trans.on Information Theory,1990,36(2):312-321.
[5]CADZOW J A.Blind Deconvolution via Cumulant Ex-trema[M].IEEE Signal Proc.Magazine,1996:24-42.
[6]KUNDUR D.and HATZINAKOS D.Blind ImageDecovolution[M].IEEE signal Proc.Magazine,1996:43-64.
[7]CHEN C H.New Cumulant-Based Inverse FilterCriteia for Deconvolution of Non-minimum PhaseSystem[J].IEEE Trans.Signal Proc.,1996,44(5):1292-1297.
[8]MENDEL J M.Tutorial on Higher_Order Statistics(Spectra)in Signal Processing and System Theory:Theoritical Results and Some Applications[J].Proc.IEEE,1991,79(3):278-305.
[9]LAZEAR G D.Mixed-phase wavelet estimation usingfourth-order cumulants[J].Geophysics,1993,58(7):1042-1049.
[10]GIANNAKJS G B,MENDEL J M.and XiaofengZhao,A Fast Prediction_Error Detecter for Estima-tion Sparse_Spike Sequences[J].IEEE Trans.ongeoscience and remote sensing,1989,27(3):344-351.
[11]WANG L X,etc.Adaptive minimum prediction-errordeconvolution and source wavelet estimation usingHopfield neural networks[J].Geophysics,1992,57(4):670-675.
[12]Li_Xin Wang and MENDEL J M.Adaptive minimumprediction_error deconvolution and source wavelet es-timation using Hopfield neural networks[J].Geo-physics,1992,57(4):670-679.
[13]HU M K.Visual Pattern Recognition by Moment In-variant[J].IRE Trans.Inform.Theory,1962,8:179-187.
[14]胡必锦,汪达成,雷鸣.褶积模型参数估计的递归算法及其收敛性[J].西南师范大学学报,2005,30(4):633-637.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心