Thumbnail
Access Restriction
Subscribed

Author Flinn, E. A.
Source ACM Digital Library
Content type Text
Publisher Association for Computing Machinery (ACM)
File Format PDF
Copyright Year ©1960
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science
Abstract Filon's method of numerical integration was developed to deal with integrals of the form $\textit{I}$ = $∫\textit{B}\textit{A}$ $ƒ(\textit{x})$ cos px dx (1) (Filon, 1928; Tranter, 1951). This method, most useful when $\textit{p}$ is large, is a modified Simpson's rule using an interval no larger than is required to integrate $∫\textit{B}\textit{A}$ $ƒ(\textit{x})$ $\textit{dx}$ alone to the desired accuracy. The derivation proceeds as follows: The range of integration is divided into panels of width $2\textit{h},$ and a second-order curve is fitted to the middle and end ordinates of one panel. After twice integrating by parts over the width of the panel and summing over all the panels, the result is $\textit{I}$ = $\textit{h}{\textit{α}}[ƒ(\textit{B})$ sin $\textit{pB}$ - $ƒ(\textit{A})$ sin $\textit{pA}]$ + $\textit{βCe}$ + $\textit{γCo}},$ (2) where $\textit{h}$ is the interval, $\textit{Ce}$ is the sum of all the even ordinates of $ƒ(\textit{x})$ cos $\textit{px}$ less half the end ordinates, $\textit{Co}$ is the sum of all the odd ordinates of $ƒ(\textit{x})$ cos $\textit{px}$ less half the end ordinates, $\textit{Co}$ is the sum of all the odd ordinates of $ƒ(\textit{x})$ cos $\textit{px},$ and $\textit{&thgr;}$ = $\textit{hp}$ $\textit{&thgr;}3\textit{α}$ = $\textit{&thgr;}2$ + $\textit{&thgr;}$ sin $\textit{&thgr;}$ cos $\textit{&thgr;}$ - 2 sin2 $\textit{&thgr;}$ $\textit{&thgr;}3\textit{β}$ = $2[\textit{&thgr;}(1$ + cos2) - 2 sin $\textit{&thgr;}$ cos $\textit{&thgr;}]$ $\textit{&thgr;}3\textit{γ}$ = 4[sin $\textit{&thgr;}$ - $\textit{&thgr;}$ cos $\textit{&thgr;}].$ (3) In the limit as $\textit{p}$ approaches zero, (2) reduces to Simpson's rule.The present modification was developed to evaluate functions of the form $\textit{F}(\textit{T})$ = $∫\textit{T}0$ $ƒ(\textit{x})$ cos px dx using a larger interval for a permissible error than is possible with Filon's formula. A fifth-order instead of a second-order curve may be fitted to the middle and end points of a panel. Substituting the first five terms of a Stirling approximation into (1), integrating over the width of the panel by parts five times, and summing over all the panels, we obtain $\textit{I}$ = $∫\textit{B}\textit{A}$ $ƒ(\textit{x})$ cos px dx = $\textit{h}{\textit{S}[ƒ(\textit{B})$ sin $\textit{pB}$ - $ƒ(\textit{A})$ sin $\textit{pA}]$ + $\textit{hP}[ƒ′(\textit{B})$ cos $\textit{pB}$ - $ƒ′(\textit{A})$ cos $\textit{pA}]$ + $\textit{RCce}$ + hQ $C}′\textit{se}$ + $\textit{NCco}$ + hM $C}′\textit{so}$ (4) where primes denote differentiation with respect to the argument, $\textit{Cco}$ = sum of odd ordinates of $ƒ(\textit{x})$ cos $\textit{px};$ $\textit{C}′\textit{so}$ = sum of odd ordinates of $ƒ′(\textit{x})$ sin $\textit{px};$ $\textit{Cce}$ = sum of even ordinates of $ƒ(\textit{x})$ cos $\textit{px},$ less half the end ordinates; $\textit{C}′\textit{se}$ = sum of even ordinates of $ƒ′(\textit{x})$ sin $\textit{px},$ less half the end ordinates; and $\textit{&thgr;}$ = $\textit{hp}$ $\textit{&thgr;}6\textit{M}$ = $16\textit{&thgr;}(15$ - $\textit{&thgr;}2)$ cos $\textit{&thgr;}$ + $48(2\textit{&thgr;}2$ - 5) sin $\textit{&thgr;}$ $\textit{&thgr;}6\textit{N}$ = $16\textit{&thgr;}(3$ - $\textit{&thgr;}2)$ sin $\textit{&thgr;}$ - $48\textit{&thgr;}2$ cos $\textit{&thgr;}$ $\textit{&thgr;}6\textit{P}$ = $2\textit{&thgr;}(\textit{&thgr;}2$ - 24) sin $\textit{&thgr;}$ cos $\textit{&thgr;}$ + $15(\textit{&thgr;}2$ - 4) cos2 $\textit{&thgr;}$ + $\textit{&thgr;}4$ - $27\textit{&thgr;}2$ + 60 $\textit{&thgr;}6\textit{Q}$ = $2[\textit{&thgr;}(12$ - $5\textit{&thgr;}2)$ + $15(\textit{thgr;}2$ - 4) sin $\textit{&thgr;}$ cos $\textit{&thgr;}$ + $2\textit{&thgr;}(24$ - $\textit{&thgr;}2)$ cos2 $\textit{&thgr;}]$ $\textit{&thgr;}2\textit{R}$ = $2[\textit{&thgr;}(156$ - $7\textit{&thgr;}2)$ sin $\textit{&thgr;}$ cos $\textit{&thgr;}$ + 3(60 - $17\textit{&thgr;}2)$ cos2 $\textit{&thgr;}$ - 15(12 - $5\textit{&thgr;}2)]$ $\textit{&thgr;}6\textit{S}$ = $\textit{&thgr;}$ $(\textit{&thgr;}4$ + $8\textit{&thgr;}2$ - 24) + $\textit{&thgr;}(7\textit{&thgr;}2$ - 156) cos2 $\textit{&thgr;}$ + 3(60 - $17\textit{&thgr;}2)$ sin $\textit{&thgr;}$ cos $\textit{&thgr;}.$ (5) For $\textit{&thgr;}$ less than about 0.9, it is better to expand equations (5) in powers of $\textit{&thgr;}:$ $\textit{M}$ = $-16/105\textit{&thgr;}$ + $8/945\textit{&thgr;}3$ - $2/10395\textit{&thgr;}5$ + $1/405405\textit{&thgr;}7$ - $1/48648600\textit{\textit{&thgr;}9$ $\textit{N}$ = 16/15 - $8/105\textit{&thgr;}2$ + $2/945\textit{&thgr;}4$ - $1/31185\textit{&thgr;}6$ + $1/3243240\textit{&thgr;}8$ $\textit{P}$ = -1/15 + $2/105\textit{&thgr;}2$ - $1/315\textit{&thgr;}4$ + $2/7425\textit{&thgr;}6$ - $62/4729725\textit{&thgr;}8$ $\textit{Q}$ = $-8/105\textit{&thgr;}$ + $16/945\textit{&thgr;}3$ - $104/51975\textit{&thgr;}5$ + $256/2027025\textit{&thgr;}7$ - $16/3274425\textit{&thgr;}9$ $\textit{R}$ = 14/15 - $16/105\textit{&thgr;}2$ + $22/945\textit{&thgr;}4$ - $608/311850\textit{&thgr;}6$ + $268/2837835\textit{&thgr;}8$ $\textit{S}$ = $19/105\textit{&thgr;}$ - $2/63\textit{&thgr;}3$ + $1/275\textit{&thgr;}5$ - $2/8775\textit{&thgr;}7$ + $34/3869775\textit{&thgr;}9.$ (6) For sin $\textit{px}$ instead of cos $\textit{px}$ in the integrand, the result is $∫\textit{B}\textit{A}$ $ƒ(\textit{x})$ sin px dx= $\textit{h}{\textit{S}[ƒ(\textit{A})$ cos $\textit{pA}$ - ƒ $(\textit{B})$ cos $\textit{pB}]$ + $\textit{hP}[ƒ′(\textit{B})$ sin $\textit{pB}$ - $ƒ′}(\textit{A})$ sin $\textit{pA}]$ + $\textit{RCse}$ - $\textit{hQC}′\textit{co}$ + $\textit{NCso}$ - $\textit{hMC}′\textit{co}}$ with an obvious change of notation for the $\textit{C}'s.In$ the limit as $\textit{p}$ approaches zero, equation (4) reduces to the modified Simpson's rule described by Lanczos (1957).Analytical estimation of the error involved in this method has not been carried out. As an empirical check on the error—and on the expansions (6)—the integral ∫1.50.5 $\textit{ex}$ cos πx dx was evaluated to nine decimal places by Filon's method and by the modification. The errors, compared with the true value of -1.7718441, were: Interval Number of points Error, Filon's method Error, Modification 0.1 11 .00000141 < 10-8 0.25 5 .00070660 .00000016 0.5 3 .00051522 .00008785 As Lanczos (1957) points out, the validity of these integration procedures depends on the convergence of the Stirling approximation of $ƒ(\textit{x}).$ For Filon's method this necessitates the smooth behaviour of differences up to the fourth order, and for the modification presented here, of differences up to the sixth order.Used with Hitchcock's (1957) approximations, Filon's method, or this modification of it, is also useful with integrals containing $\textit{J}0(\textit{px})$ or $\textit{J}1(\textit{px})$ instead of the trigonometric functions.Luke (1954) considered the fitting of an $\textit{n}th$ order curve to the middle and end points of a panel, and gave a detailed discussion of the error involved—the results obtained here are special cases of his equations. Luke's formulae are, however, suitable primarily for hand computation using tabulated functions, whereas the results given here are intended for use with an electronic computer.The author wishes to thank the referee for drawing his attention to the paper by Luke.
ISSN 00045411
Age Range 18 to 22 years ♦ above 22 year
Educational Use Research
Education Level UG and PG
Learning Resource Type Article
Publisher Date 1960-04-01
Publisher Place New York
e-ISSN 1557735X
Journal Journal of the ACM (JACM)
Volume Number 7
Issue Number 2
Page Count 4
Starting Page 181
Ending Page 184


Open content in new tab

   Open content in new tab
Source: ACM Digital Library