circle (original) (raw)

Another way of defining the circle is thus: Given A and P as two points and O as another point, the circle with center O is the locus of points X with O⁢X congruent to A⁢P. (Hilbert, 1927)

Notice there is no definition of distance needed to make that definition and so it works in many geometries, even ones with no distance function. Hilbert uses it in his Foundations of Geometry book Also used by Forder in his Foundations of Euclidean geometry book, c 1927

A circle determines a closed curve in the plane, and this curve is called the perimeter or _circumference_of the circle. If the radius of a circle is r, then the length of the perimeter is 2⁢π⁢r. Also, the area of the circle is π⁢r2. More precisely, the interior of the perimeter has area π⁢r2. The diameter of a circle is defined as d=2⁢r.

The circle is a special case of an ellipse. Also, in three dimensions, the analogous geometric object to a circle is a sphere.

1

Let us next derive an analytic equation for a circle in Cartesian coordinates (x,y). If the circle has center (a,b) and radius r>0, we obtain the following condition for the points of the sphere,

In other words, the circle is the set of all points (x,y) that satisfy the above equation. In the special case thata=b=0, the equation is simply x2+y2=r2. The unit circle is the circle x2+y2=1.

It is clear that equation 1 can always be reduced to the form

where D,E,F are real numbers. Conversely, suppose that we are given an equation of the above form where D,E,F are arbitrary real numbers. Next we derive conditions for these constants, so that equation (2) determines a circle [2]. Completing the squares yields

whence

There are three cases:

1. 1. If D2-4⁢F+E2>0, then equation (2) determines a circle with center (-D2,-E2) and radius12⁢D2-4⁢F+E2.
2. 1. If D2-4⁢F+E2=0, then equation (2) determines the point (-D2,-E2).
3. 1. If D2-4⁢F+E2<0, then equation (2) has no (real) solution in the (x,y) - plane.

2 The circle in polar coordinates

Using polar coordinates for the plane, we can parameterize the circle. Consider the circle with center (a,b) and radius r>0 in the plane ℝ2. It is then natural to introduce polar coordinates (ρ,ϕ) for ℝ2∖{(a,b)} by

with ρ>0 and ϕ∈[0,2⁢π). Since we wish to parameterize the circle, the point (a,b) does not pose a problem; it is not part of the circle. Plugging these expressions for x,y into equation (1) yields the condition ρ=r. The given circle is thus parameterization by ϕ↦(a+ρ⁢cos⁡ϕ,b+ρ⁢sin⁡ϕ), ϕ∈[0,2⁢π). It follows that a circle is a closed curve in the plane.

3 Three point formula for the circle

Suppose we are given three points on a circle, say (x1,y1), (x2,y2), (x3,y3). We next derive expressions for the parameters D,E,F in terms of these points. We also derive equation (3), which gives an equation for a circle in terms of a determinant.

First, from equation (2), we have

These equations form a linear set of equations for D,E,F, i.e.,

Let us denote the matrix on the left hand side by Λ. Also, let us assume that det⁡Λ≠0. Then, using Cramer’s rule, we obtain

These equations give the parameters D,E,F as functions of the three given points. Substituting these equations into equation (2) yields

Using the cofactor expansion, we can now write the equation for the circle passing through (x1,y1),(x2,y2),(x3,y3) as [3, 4]

References

• 1 D. Hilbert, Foundations of Geometry Chicago: The Open Court Publishing Co. (1921): 163
• 2 J. H. Kindle,Schaum’s Outline Series: Theory and problems of plane of Solid Analytic Geometry, Schaum Publishing Co., 1950.
• 3 E. Weisstein, Eric W. Weisstein’s world of mathematics,http://mathworld.wolfram.com/Circle.htmlentry on the circle.
• 4 L. Råde, B. Westergren,Mathematics Handbook for Science and Engineering, Studentlitteratur, 1995.