boost_1_45_0/libs/math/doc/sf_and_dist/ellint_introduction.qbk

[/
Copyright (c) 2006 Xiaogang Zhang
Use, modification and distribution are subject to the
Boost Software License, Version 1.0. (See accompanying file
LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
]

[section:ellint_intro Elliptic Integral Overview]

The main reference for the elliptic integrals is:

[:M. Abramowitz and I. A. Stegun (Eds.) (1964)
Handbook of Mathematical Functions with Formulas, Graphs, and
Mathematical Tables,      
National Bureau of Standards Applied Mathematics Series, U.S. Government Printing Office, Washington, D.C.]

Mathworld also contain a lot of useful background information:

[:[@http://mathworld.wolfram.com/EllipticIntegral.html Weisstein, Eric W. 
"Elliptic Integral." From MathWorld--A Wolfram Web Resource.]]

As does [@http://en.wikipedia.org/wiki/Elliptic_integral Wikipedia Elliptic integral].

[h4 Notation]

All variables are real numbers unless otherwise noted.

[h4 [#ellint_def]Definition]

[equation ellint1]

is called elliptic integral if ['R(t, s)] is a rational function
of ['t] and ['s], and ['s[super 2]] is a cubic or quartic polynomial
in ['t].

Elliptic integrals generally can not be expressed in terms of
elementary functions. However, Legendre showed that all elliptic
integrals can be reduced to the following three canonical forms:

Elliptic Integral of the First Kind (Legendre form)

[equation ellint2]

Elliptic Integral of the Second Kind (Legendre form)

[equation ellint3]

Elliptic Integral of the Third Kind (Legendre form)

[equation ellint4]

where

[equation ellint5]

[note ['[phi]] is called the amplitude.

['k] is called the modulus. 

['[alpha]] is called the modular angle. 

['n] is called the characteristic.]
      
[caution Perhaps more than any other special functions the elliptic
integrals are expressed in a variety of different ways.  In particular,
the final parameter /k/ (the modulus) may be expressed using a modular
angle [alpha], or a parameter /m/.  These are related by:

k = sin[alpha]

m = k[super 2] = sin[super 2][alpha]

So that the integral of the third kind (for example) may be expressed as
either:

[Pi](n, [phi], k)

[Pi](n, [phi] \\ [alpha])

[Pi](n, [phi]| m)

To further complicate matters, some texts refer to the ['complement
of the parameter m], or 1 - m, where:

1 - m = 1 - k[super 2] = cos[super 2][alpha]

This implementation uses /k/ throughout: this matches the requirements
of the [@http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2005/n1836.pdf
Technical Report on C++ Library Extensions].  However, you should
be extra careful when using these functions!]

When ['[phi]] = ['[pi]] / 2, the elliptic integrals are called ['complete].

Complete Elliptic Integral of the First Kind (Legendre form)

[equation ellint6]

Complete Elliptic Integral of the Second Kind (Legendre form)

[equation ellint7]

Complete Elliptic Integral of the Third Kind (Legendre form)

[equation ellint8]

Carlson [[link ellint_ref_carlson77 Carlson77]] [[link ellint_ref_carlson78  Carlson78]] gives an alternative definition of
elliptic integral's canonical forms:

Carlson's Elliptic Integral of the First Kind

[equation ellint9]

where ['x], ['y], ['z] are nonnegative and at most one of them
may be zero.

Carlson's Elliptic Integral of the Second Kind

[equation ellint10]

where ['x], ['y] are nonnegative, at most one of them may be zero,
and ['z] must be positive.

Carlson's Elliptic Integral of the Third Kind

[equation ellint11]

where ['x], ['y], ['z] are nonnegative, at most one of them may be
zero, and ['p] must be nonzero.

Carlson's Degenerate Elliptic Integral

[equation ellint12]

where ['x] is nonnegative and ['y] is nonzero.

[note ['R[sub C](x, y) = R[sub F](x, y, y)] 

['R[sub D](x, y, z) = R[sub J](x, y, z, z)]]


[h4 [#ellint_theorem]Duplication Theorem]

Carlson proved in [[link ellint_ref_carlson78  Carlson78]] that

[equation ellint13]

[h4 [#ellint_formula]Carlson's Formulas]

The Legendre form and Carlson form of elliptic integrals are related
by equations:

[equation ellint14]

In particular,

[equation ellint15]

[h4 Numerical Algorithms]

The conventional methods for computing elliptic integrals are Gauss
and Landen transformations, which converge quadratically and work
well for elliptic integrals of the first and second kinds.
Unfortunately they suffer from loss of significant digits for the
third kind. Carlson's algorithm [[link ellint_ref_carlson79  Carlson79]] [[link ellint_ref_carlson78  Carlson78]], by contrast,
provides a unified method for all three kinds of elliptic integrals
with satisfactory precisions.

[h4 [#ellint_refs]References]

Special mention goes to:

[:A. M. Legendre, ['Traitd des Fonctions Elliptiques et des Integrales 
Euleriennes], Vol. 1. Paris (1825).]

However the main references are:

# [#ellint_ref_AS]M. Abramowitz and I. A. Stegun (Eds.) (1964)
Handbook of Mathematical Functions with Formulas, Graphs, and
Mathematical Tables,      
National Bureau of Standards Applied Mathematics Series, U.S. Government Printing Office, Washington, D.C.
# [#ellint_ref_carlson79]B.C. Carlson, ['Computing elliptic integrals by duplication],
    Numerische Mathematik, vol 33, 1 (1979).
# [#ellint_ref_carlson77]B.C. Carlson, ['Elliptic Integrals of the First Kind],
    SIAM Journal on Mathematical Analysis, vol 8, 231 (1977).
# [#ellint_ref_carlson78]B.C. Carlson, ['Short Proofs of Three Theorems on Elliptic Integrals],
    SIAM Journal on Mathematical Analysis, vol 9, 524 (1978).
# [#ellint_ref_carlson81]B.C. Carlson and E.M. Notis, ['ALGORITHM 577: Algorithms for Incomplete
    Elliptic Integrals], ACM Transactions on Mathematmal Software,
    vol 7, 398 (1981).
# B. C. Carlson, ['On computing elliptic integrals and functions]. J. Math. and
Phys., 44 (1965), pp. 36-51.
# B. C. Carlson, ['A table of elliptic integrals of the second kind]. Math. Comp., 49
(1987), pp. 595-606. (Supplement, ibid., pp. S13-S17.)
# B. C. Carlson, ['A table of elliptic integrals of the third kind]. Math. Comp., 51 (1988),
pp. 267-280. (Supplement, ibid., pp. S1-S5.)
# B. C. Carlson, ['A table of elliptic integrals: cubic cases]. Math. Comp., 53 (1989), pp.
327-333.
# B. C. Carlson, ['A table of elliptic integrals: one quadratic factor]. Math. Comp., 56 (1991),
pp. 267-280.
# B. C. Carlson, ['A table of elliptic integrals: two quadratic factors]. Math. Comp., 59
(1992), pp. 165-180.
# B. C. Carlson, ['[@http://arxiv.org/abs/math.CA/9409227 
Numerical computation of real or complex elliptic integrals]]. Numerical Algorithms,
Volume 10, Number 1 / March, 1995, p13-26.
# B. C. Carlson and John L. Gustafson, ['[@http://arxiv.org/abs/math.CA/9310223
Asymptotic Approximations for Symmetric Elliptic Integrals]], 
SIAM Journal on Mathematical Analysis, Volume 25, Issue 2 (March 1994), 288-303.


The following references, while not directly relevent to our implementation,
may also be of interest:

# R. Burlisch, ['Numerical Compuation of Elliptic Integrals and Elliptic Functions.]
Numerical Mathematik 7, 78-90.
# R. Burlisch, ['An extension of the Bartky Transformation to Incomplete
Elliptic Integrals of the Third Kind]. Numerical Mathematik 13, 266-284.
# R. Burlisch, ['Numerical Compuation of Elliptic Integrals and Elliptic Functions. III]. 
Numerical Mathematik 13, 305-315.
# T. Fukushima and H. Ishizaki, ['[@http://adsabs.harvard.edu/abs/1994CeMDA..59..237F 
Numerical Computation of Incomplete Elliptic Integrals of a General Form.]]
Celestial Mechanics and Dynamical Astronomy, Volume 59, Number 3 / July, 1994,
237-251.

[endsect]