N-sphere

Summary

In mathematics, an n-sphere or hypersphere is an -dimensional generalization of the -dimensional circle and -dimensional sphere to any non-negative integer . The -sphere is the setting for -dimensional spherical geometry.

2-sphere wireframe as an orthogonal projection
Just as a stereographic projection can project a sphere's surface to a plane, it can also project a 3-sphere into 3-space. This image shows three coordinate directions projected to 3-space: parallels (red), meridians (blue), and hypermeridians (green). Due to the conformal property of the stereographic projection, the curves intersect each other orthogonally (in the yellow points) as in 4D. All of the curves are circles: the curves that intersect ⟨0,0,0,1⟩ have an infinite radius (= straight line).

Considered extrinsically, as a hypersurface embedded in -dimensional Euclidean space, an -sphere is the locus of points at equal distance (the radius) from a given center point. Its interior, consisting of all points closer to the center than the radius, is an -dimensional ball. In particular:

  • The -sphere is the pair of points at the ends of a line segment (-ball).
  • The -sphere is a circle, the circumference of a disk (-ball) in the two-dimensional plane.
  • The -sphere, often simply called a sphere, is the boundary of a -ball in three-dimensional space.
  • The 3-sphere is the boundary of a -ball in four-dimensional space.
  • The -sphere is the boundary of an -ball.

Given a Cartesian coordinate system, the unit -sphere of radius can be defined as:

Considered intrinsically, when , the -sphere is a Riemannian manifold of positive constant curvature, and is orientable. The geodesics of the -sphere are called great circles.

The stereographic projection maps the -sphere onto -space with a single adjoined point at infinity; under the metric thereby defined, is a model for the -sphere.

In the more general setting of topology, any topological space that is homeomorphic to the unit -sphere is called an -sphere. Under inverse stereographic projection, the -sphere is the one-point compactification of -space. The -spheres admit several other topological descriptions: for example, they can be constructed by gluing two -dimensional spaces together, by identifying the boundary of an -cube with a point, or (inductively) by forming the suspension of an -sphere. When it is simply connected; the -sphere (circle) is not simply connected; the -sphere is not even connected, consisting of two discrete points.

Description

edit

For any natural number  , an  -sphere of radius   is defined as the set of points in  -dimensional Euclidean space that are at distance   from some fixed point  , where   may be any positive real number and where   may be any point in  -dimensional space. In particular:

  • a 0-sphere is a pair of points  , and is the boundary of a line segment ( -ball).
  • a 1-sphere is a circle of radius   centered at  , and is the boundary of a disk ( -ball).
  • a 2-sphere is an ordinary  -dimensional sphere in  -dimensional Euclidean space, and is the boundary of an ordinary ball ( -ball).
  • a 3-sphere is a  -dimensional sphere in  -dimensional Euclidean space.

Cartesian coordinates

edit

The set of points in  -space,  , that define an  -sphere,  , is represented by the equation:

 

where   is a center point, and   is the radius.

The above  -sphere exists in  -dimensional Euclidean space and is an example of an  -manifold. The volume form   of an  -sphere of radius   is given by

 

where   is the Hodge star operator; see Flanders (1989, §6.1) for a discussion and proof of this formula in the case  . As a result,

 

n-ball

edit

The space enclosed by an  -sphere is called an  -ball. An  -ball is closed if it includes the  -sphere, and it is open if it does not include the  -sphere.

Specifically:

  • A  -ball, a line segment, is the interior of a 0-sphere.
  • A  -ball, a disk, is the interior of a circle ( -sphere).
  • A  -ball, an ordinary ball, is the interior of a sphere ( -sphere).
  • A  -ball is the interior of a 3-sphere, etc.

Topological description

edit

Topologically, an  -sphere can be constructed as a one-point compactification of  -dimensional Euclidean space. Briefly, the  -sphere can be described as  , which is  -dimensional Euclidean space plus a single point representing infinity in all directions. In particular, if a single point is removed from an  -sphere, it becomes homeomorphic to  . This forms the basis for stereographic projection.[1]

Volume and area

edit

Let   be the surface area of the unit  -sphere of radius   embedded in  -dimensional Euclidean space, and let   be the volume of its interior, the unit  -ball. The surface area of an arbitrary  -sphere is proportional to the  st power of the radius, and the volume of an arbitrary  -ball is proportional to the  th power of the radius.

 
Sn &minus 1) of n-balls of radius 1.

The  -ball is sometimes defined as a single point. The  -dimensional Hausdorff measure is the number of points in a set. So

 

A unit  -ball is a line segment whose points have a single coordinate in the interval   of length  , and the  -sphere consists of its two end-points, with coordinate  .

 

A unit  -sphere is the unit circle in the Euclidean plane, and its interior is the unit disk ( -ball).

 

The interior of a 2-sphere in three-dimensional space is the unit  -ball.

 

In general,   and   are given in closed form by the expressions

 

where   is the gamma function.

As   tends to infinity, the volume of the unit  -ball (ratio between the volume of an  -ball of radius   and an  -cube of side length  ) tends to zero.[2]

Recurrences

edit

The surface area, or properly the  -dimensional volume, of the  -sphere at the boundary of the  -ball of radius   is related to the volume of the ball by the differential equation

 

Equivalently, representing the unit  -ball as a union of concentric  -sphere shells,

 

We can also represent the unit  -sphere as a union of products of a circle ( -sphere) with an  -sphere. Then  . Since  , the equation

 

holds for all  . Along with the base cases  ,   from above, these recurrences can be used to compute the surface area of any sphere or volume of any ball.

Spherical coordinates

edit

We may define a coordinate system in an  -dimensional Euclidean space which is analogous to the spherical coordinate system defined for  -dimensional Euclidean space, in which the coordinates consist of a radial coordinate  , and   angular coordinates  , where the angles   range over   radians (or   degrees) and   ranges over   radians (or   degrees). If   are the Cartesian coordinates, then we may compute   from   with:[3][a]

 

Except in the special cases described below, the inverse transformation is unique:

 

where atan2 is the two-argument arctangent function.

There are some special cases where the inverse transform is not unique;   for any   will be ambiguous whenever all of   are zero; in this case   may be chosen to be zero. (For example, for the  -sphere, when the polar angle is   or   then the point is one of the poles, zenith or nadir, and the choice of azimuthal angle is arbitrary.)

Spherical volume and area elements

edit

To express the volume element of  -dimensional Euclidean space in terms of spherical coordinates, let   and   for concision, then observe that the Jacobian matrix of the transformation is:

 

The determinant of this matrix can be calculated by induction. When  , a straightforward computation shows that the determinant is  . For larger  , observe that   can be constructed from   as follows. Except in column  , rows   and   of   are the same as row   of  , but multiplied by an extra factor of   in row   and an extra factor of   in row  . In column  , rows   and   of   are the same as column   of row   of  , but multiplied by extra factors of   in row   and   in row  , respectively. The determinant of   can be calculated by Laplace expansion in the final column. By the recursive description of  , the submatrix formed by deleting the entry at   and its row and column almost equals  , except that its last row is multiplied by  . Similarly, the submatrix formed by deleting the entry at   and its row and column almost equals  , except that its last row is multiplied by  . Therefore the determinant of   is

 

Induction then gives a closed-form expression for the volume element in spherical coordinates

 

The formula for the volume of the  -ball can be derived from this by integration.

Similarly the surface area element of the  -sphere of radius  , which generalizes the area element of the  -sphere, is given by

 

The natural choice of an orthogonal basis over the angular coordinates is a product of ultraspherical polynomials,

 

for  , and the   for the angle   in concordance with the spherical harmonics.

Polyspherical coordinates

edit

The standard spherical coordinate system arises from writing   as the product  . These two factors may be related using polar coordinates. For each point   of  , the standard Cartesian coordinates

 

can be transformed into a mixed polar–Cartesian coordinate system:

 

This says that points in   may be expressed by taking the ray starting at the origin and passing through  , rotating it towards   by  , and traveling a distance   along the ray. Repeating this decomposition eventually leads to the standard spherical coordinate system.

Polyspherical coordinate systems arise from a generalization of this construction.[4] The space   is split as the product of two Euclidean spaces of smaller dimension, but neither space is required to be a line. Specifically, suppose that   and   are positive integers such that  . Then  . Using this decomposition, a point   may be written as

 

This can be transformed into a mixed polar–Cartesian coordinate system by writing:

 

Here   and   are the unit vectors associated to   and  . This expresses   in terms of  ,  ,  , and an angle  . It can be shown that the domain of   is   if  ,   if exactly one of   and   is  , and   if neither   nor   are  . The inverse transformation is

 

These splittings may be repeated as long as one of the factors involved has dimension two or greater. A polyspherical coordinate system is the result of repeating these splittings until there are no Cartesian coordinates left. Splittings after the first do not require a radial coordinate because the domains of   and   are spheres, so the coordinates of a polyspherical coordinate system are a non-negative radius and   angles. The possible polyspherical coordinate systems correspond to binary trees with   leaves. Each non-leaf node in the tree corresponds to a splitting and determines an angular coordinate. For instance, the root of the tree represents  , and its immediate children represent the first splitting into   and  . Leaf nodes correspond to Cartesian coordinates for  . The formulas for converting from polyspherical coordinates to Cartesian coordinates may be determined by finding the paths from the root to the leaf nodes. These formulas are products with one factor for each branch taken by the path. For a node whose corresponding angular coordinate is  , taking the left branch introduces a factor of   and taking the right branch introduces a factor of  . The inverse transformation, from polyspherical coordinates to Cartesian coordinates, is determined by grouping nodes. Every pair of nodes having a common parent can be converted from a mixed polar–Cartesian coordinate system to a Cartesian coordinate system using the above formulas for a splitting.

Polyspherical coordinates also have an interpretation in terms of the special orthogonal group. A splitting   determines a subgroup

 

This is the subgroup that leaves each of the two factors   fixed. Choosing a set of coset representatives for the quotient is the same as choosing representative angles for this step of the polyspherical coordinate decomposition.

In polyspherical coordinates, the volume measure on   and the area measure on   are products. There is one factor for each angle, and the volume measure on   also has a factor for the radial coordinate. The area measure has the form:

 

where the factors   are determined by the tree. Similarly, the volume measure is

 

Suppose we have a node of the tree that corresponds to the decomposition   and that has angular coordinate  . The corresponding factor   depends on the values of   and  . When the area measure is normalized so that the area of the sphere is  , these factors are as follows. If  , then

 

If   and  , and if   denotes the beta function, then

 

If   and  , then

 

Finally, if both   and   are greater than one, then

 

Stereographic projection

edit

Just as a two-dimensional sphere embedded in three dimensions can be mapped onto a two-dimensional plane by a stereographic projection, an  -sphere can be mapped onto an  -dimensional hyperplane by the  -dimensional version of the stereographic projection. For example, the point   on a two-dimensional sphere of radius   maps to the point   on the  -plane. In other words,

 

Likewise, the stereographic projection of an  -sphere   of radius   will map to the  -dimensional hyperplane   perpendicular to the  -axis as

 

Probability distributions

edit

Uniformly at random on the (n − 1)-sphere

edit
 
A set of points drawn from a uniform distribution on the surface of a unit 2-sphere, generated using Marsaglia's algorithm.

To generate uniformly distributed random points on the unit  -sphere (that is, the surface of the unit  -ball), Marsaglia (1972) gives the following algorithm.

Generate an  -dimensional vector of normal deviates (it suffices to use  , although in fact the choice of the variance is arbitrary),  . Now calculate the "radius" of this point:

 

The vector   is uniformly distributed over the surface of the unit  -ball.

An alternative given by Marsaglia is to uniformly randomly select a point   in the unit n-cube by sampling each   independently from the uniform distribution over  , computing   as above, and rejecting the point and resampling if   (i.e., if the point is not in the  -ball), and when a point in the ball is obtained scaling it up to the spherical surface by the factor  ; then again   is uniformly distributed over the surface of the unit  -ball. This method becomes very inefficient for higher dimensions, as a vanishingly small fraction of the unit cube is contained in the sphere. In ten dimensions, less than 2% of the cube is filled by the sphere, so that typically more than 50 attempts will be needed. In seventy dimensions, less than   of the cube is filled, meaning typically a trillion quadrillion trials will be needed, far more than a computer could ever carry out.

Uniformly at random within the n-ball

edit

With a point selected uniformly at random from the surface of the unit  -sphere (e.g., by using Marsaglia's algorithm), one needs only a radius to obtain a point uniformly at random from within the unit  -ball. If   is a number generated uniformly at random from the interval   and   is a point selected uniformly at random from the unit  -sphere, then   is uniformly distributed within the unit  -ball.

Alternatively, points may be sampled uniformly from within the unit  -ball by a reduction from the unit  -sphere. In particular, if   is a point selected uniformly from the unit  -sphere, then   is uniformly distributed within the unit  -ball (i.e., by simply discarding two coordinates).[5]

If   is sufficiently large, most of the volume of the  -ball will be contained in the region very close to its surface, so a point selected from that volume will also probably be close to the surface. This is one of the phenomena leading to the so-called curse of dimensionality that arises in some numerical and other applications.

Distribution of the first coordinate

edit

Let   be the square of the first coordinate of a point sampled uniformly at random from the  -sphere, then its probability density function, for  , is

 

Let   be the appropriately scaled version, then at the   limit, the probability density function of   converges to  . This is sometimes called the Porter–Thomas distribution.[6]

Specific spheres

edit
0-sphere
The pair of points   with the discrete topology for some  . The only sphere that is not path-connected. Parallelizable.
1-sphere
Commonly called a circle. Has a nontrivial fundamental group. Abelian Lie group structure U(1); the circle group. Homeomorphic to the real projective line.
2-sphere
Commonly simply called a sphere. For its complex structure, see Riemann sphere. Homeomorphic to the complex projective line
3-sphere
Parallelizable, principal U(1)-bundle over the  -sphere, Lie group structure Sp(1).
4-sphere
Homeomorphic to the quaternionic projective line,  .  .
5-sphere
Principal U(1)-bundle over the complex projective space  .  . It is undecidable whether a given  -dimensional manifold is homeomorphic to   for  .[7]
6-sphere
Possesses an almost complex structure coming from the set of pure unit octonions.  . The question of whether it has a complex structure is known as the Hopf problem, after Heinz Hopf.[8]
7-sphere
Topological quasigroup structure as the set of unit octonions. Principal  -bundle over S^4. Parallelizable.  . The  -sphere is of particular interest since it was in this dimension that the first exotic spheres were discovered.
8-sphere
Homeomorphic to the octonionic projective line  .
23-sphere
A highly dense sphere-packing is possible in  -dimensional space, which is related to the unique qualities of the Leech lattice.

Octahedral sphere

edit

The octahedral  -sphere is defined similarly to the  -sphere but using the 1-norm

 

In general, it takes the shape of a cross-polytope.

The octahedral  -sphere is a square (without its interior). The octahedral  -sphere is a regular octahedron; hence the name. The octahedral  -sphere is the topological join of   pairs of isolated points.[9] Intuitively, the topological join of two pairs is generated by drawing a segment between each point in one pair and each point in the other pair; this yields a square. To join this with a third pair, draw a segment between each point on the square and each point in the third pair; this gives a octahedron.

See also

edit

Notes

edit
  1. ^ Formally, this formula is only correct for  . For  , the line beginning with   must be omitted, and for  , the formula for polar coordinates must be used. The case   reduces to  . Using capital-pi notation and the usual convention for the empty product, a formula valid for   is given by   and   for  .
  1. ^ James W. Vick (1994). Homology theory, p. 60. Springer
  2. ^ Smith, David J.; Vamanamurthy, Mavina K. (1989). "How Small Is a Unit Ball?". Mathematics Magazine. 62 (2): 101–107. doi:10.1080/0025570X.1989.11977419. JSTOR 2690391.
  3. ^ Blumenson, L. E. (1960). "A Derivation of n-Dimensional Spherical Coordinates". The American Mathematical Monthly. 67 (1): 63–66. doi:10.2307/2308932. JSTOR 2308932.
  4. ^ N. Ja. Vilenkin and A. U. Klimyk, Representation of Lie groups and special functions, Vol. 2: Class I representations, special functions, and integral transforms, translated from the Russian by V. A. Groza and A. A. Groza, Math. Appl., vol. 74, Kluwer Acad. Publ., Dordrecht, 1992, ISBN 0-7923-1492-1, pp. 223–226.
  5. ^ Voelker, Aaron R.; Gosmann, Jan; Stewart, Terrence C. (2017). Efficiently sampling vectors and coordinates from the n-sphere and n-ball (Report). Centre for Theoretical Neuroscience. doi:10.13140/RG.2.2.15829.01767/1.
  6. ^ Livan, Giacomo; Novaes, Marcel; Vivo, Pierpaolo (2018), Livan, Giacomo; Novaes, Marcel; Vivo, Pierpaolo (eds.), "One Pager on Eigenvectors", Introduction to Random Matrices: Theory and Practice, SpringerBriefs in Mathematical Physics, Cham: Springer International Publishing, pp. 65–66, doi:10.1007/978-3-319-70885-0_9, ISBN 978-3-319-70885-0, retrieved 2023-05-19
  7. ^ Stillwell, John (1993), Classical Topology and Combinatorial Group Theory, Graduate Texts in Mathematics, vol. 72, Springer, p. 247, ISBN 9780387979700.
  8. ^ Agricola, Ilka; Bazzoni, Giovanni; Goertsches, Oliver; Konstantis, Panagiotis; Rollenske, Sönke (2018). "On the history of the Hopf problem". Differential Geometry and Its Applications. 57: 1–9. arXiv:1708.01068. doi:10.1016/j.difgeo.2017.10.014. S2CID 119297359.
  9. ^ Meshulam, Roy (2001-01-01). "The Clique Complex and Hypergraph Matching". Combinatorica. 21 (1): 89–94. doi:10.1007/s004930170006. ISSN 1439-6912. S2CID 207006642.

References

edit
  • Marsaglia, G. (1972). "Choosing a Point from the Surface of a Sphere". Annals of Mathematical Statistics. 43 (2): 645–646. doi:10.1214/aoms/1177692644.
  • Huber, Greg (1982). "Gamma function derivation of n-sphere volumes". Amer. Math. Monthly. 89 (5): 301–302. doi:10.2307/2321716. JSTOR 2321716. MR 1539933.
  • Weeks, Jeffrey R. (1985). The Shape of Space: how to visualize surfaces and three-dimensional manifolds. Marcel Dekker. ISBN 978-0-8247-7437-0(Chapter 14: The Hypersphere).{{cite book}}: CS1 maint: postscript (link)
  • Kalnins, E. G.; Miller, W. (1986). "Separation of variables on n-dimensionsional Riemannian manifolds. I. the n-sphere S_n and Euclidean n-sparce R_n". J. Math. Phys. 27: 1721–1746. doi:10.1063/1.527088. hdl:10289/1219.
  • Flanders, Harley (1989). Differential forms with applications to the physical sciences. New York: Dover Publications. ISBN 978-0-486-66169-8.
  • Moura, Eduarda; Henderson, David G. (1996). Experiencing geometry: on plane and sphere. Prentice Hall. ISBN 978-0-13-373770-7(Chapter 20: 3-spheres and hyperbolic 3-spaces).{{cite book}}: CS1 maint: postscript (link)
  • Barnea, Nir (1999). "Hyperspherical functions with arbitrary permutational symmetry: Reverse construction". Phys. Rev. A. 59 (2): 1135–1146. Bibcode:1999PhRvA..59.1135B. doi:10.1103/PhysRevA.59.1135.
edit