TY - JOUR
T1 - Inversion of Abel's integral equation for experimental data
AU - Deutsch, Moshe
AU - Beniaminy, Israel
PY - 1983
Y1 - 1983
N2 - A stable high-accuracy method for calculating the inverse of Abel's integral equation for experimentally derived data is presented. The method employs a piece-wise cubic spline function, least-squares fitted to the data, to represent the function as inverted. Three formulas, two of which are based on the well-known analytic inverses of Abel's equation and a new one, which was recently developed by the authors and does not contain an explicit derivative, are given for calculating the inverse numerically. The results of numerical tests performed using these formulas on simulated data are presented and compared with the results obtained using other published methods. The comparative study indicates that our method, employing piece-wise least-squares cubic splines, accurately reproduces the inverse function, regardless of the inversion formula employed. It yields markedly better results than do all the other methods compared in this study, when highly error free, sparse, or very noisy data are inverted. The errors at the ends of the data interval ("termination errors") are also shown to be smaller than those of the other methods compared in this study.
AB - A stable high-accuracy method for calculating the inverse of Abel's integral equation for experimentally derived data is presented. The method employs a piece-wise cubic spline function, least-squares fitted to the data, to represent the function as inverted. Three formulas, two of which are based on the well-known analytic inverses of Abel's equation and a new one, which was recently developed by the authors and does not contain an explicit derivative, are given for calculating the inverse numerically. The results of numerical tests performed using these formulas on simulated data are presented and compared with the results obtained using other published methods. The comparative study indicates that our method, employing piece-wise least-squares cubic splines, accurately reproduces the inverse function, regardless of the inversion formula employed. It yields markedly better results than do all the other methods compared in this study, when highly error free, sparse, or very noisy data are inverted. The errors at the ends of the data interval ("termination errors") are also shown to be smaller than those of the other methods compared in this study.
UR - http://www.scopus.com/inward/record.url?scp=0020542636&partnerID=8YFLogxK
U2 - 10.1063/1.331739
DO - 10.1063/1.331739
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:0020542636
SN - 0021-8979
VL - 54
SP - 137
EP - 143
JO - Journal of Applied Physics
JF - Journal of Applied Physics
IS - 1
ER -