Real Affine Plane
Definition: Real affine plane
The Real Affine Plane is an incidence structure satisfying the following axioms:
- Point is an ordered pair \((x,y)\) where \(x, y \in \mathbb{R}\).
- Line is the set of points \((x,y)\) satisfying \(ax + by + c = 0\), where \(a, b, c \in \mathbb{R}\), and not both \(a\) and \(b\) are zero.
- A point \((x_0, y_0)\) lies on the line \(ax + by + c = 0\) if and only if \(ax_0 + by_0 + c = 0\).
We denote the Real Affine Plane (also known as the Euclidean Plane) by \(\alpha_R\).
- Two points determine a unique line:
Let \((x_1, y_1)\), \((x_2, y_2)\) be two points. The line through them is: \[ (y_2 - y_1)x + (x_1 - x_2)y + (x_2 y_1 - x_1 y_2) = 0 \quad \text{(A)} \] Suppose another line \(ax + by + c = 0\) passes through both. Then: \[ ax_1 + by_1 + c = 0, \quad ax_2 + by_2 + c = 0 \] Solving gives \(a\) and \(b\) in terms of \(c\), leading back to (A). So the line is unique. - Parallel line through a point:
Given line \(L: ax + by + c = 0\) and point \(P = (x_0, y_0)\) not on \(L\), the line: \[ ax + by = ax_0 + by_0 \] passes through \(P\) and is parallel to \(L\). Uniqueness follows from slope and constant comparison. - Three non-collinear points exist:
\((0,0)\), \((1,0)\), \((0,1)\) are non-collinear.
Real Projective Plane
Definition: Real projective plane
In the projective plane, points are homogeneous coordinates: \([x_1, x_2, x_3]\), where scaling by \(k \ne 0\) gives the same point: \([x_1,x_2,x_3] = [kx_1,kx_2,kx_3]\).
If \(x_3 \ne 0\), we identify \([x_1,x_2,x_3]\) with \((x_1/x_3, x_2/x_3)\) in the affine plane.
Example: \([1,2,3] = [2,4,6]\), but as triples they are different.
- Point: Proportionality class \([x_1,x_2,x_3]\), not all zero.
- Line: Proportionality class \(\langle l_1,l_2,l_3 \rangle\), not all zero.
- Incidence: \([x_i]\) lies on \(\langle l_j \rangle\) iff \(l_1x_1 + l_2x_2 + l_3x_3 = 0\).
-
Two points in \(\pi_R\) determine a line:
Let \(x=[x_1,x_2,x_3]\) and \(y=[y_1,y_2,y_3]\) be two points, then \(\begin{vmatrix}x_1 & x_2\\y_1 & y_2\end{vmatrix} \neq 0\).
Now, there exists a line: \[ \left\langle \begin{vmatrix}x_2 & x_3\\y_2 & y_3\end{vmatrix}, \begin{vmatrix}x_3 & x_1\\y_3 & y_1\end{vmatrix}, \begin{vmatrix}x_1 & x_2\\y_1 & y_2\end{vmatrix} \right\rangle \tag{A} \] Suppose there exists another line \(\langle l_1,l_2,l_3\rangle\) (i) passing through \(x\) and \(y\), then \[ l_1 x_1 + l_2 x_2 + l_3 x_3 = 0 \quad (ii), \quad l_1 y_1 + l_2 y_2 + l_3 y_3 = 0 \quad (iii) \] Solving (ii) and (iii), we get \[ l_1 = \frac{\begin{vmatrix}x_2 & x_3\\y_2 & y_3\end{vmatrix}}{\begin{vmatrix}x_1 & x_2\\y_1 & y_2\end{vmatrix}} l_3, \quad l_2 = \frac{\begin{vmatrix}x_3 & x_1\\y_3 & y_1\end{vmatrix}}{\begin{vmatrix}x_1 & x_2\\y_1 & y_2\end{vmatrix}} l_3 \] Hence, (i) and (A) are the same. Therefore, two points in \(\pi_R\) determine a line. -
Two lines always meet:
Let \(L=\langle l_1,l_2,l_3\rangle\) and \(M=\langle m_1,m_2,m_3\rangle\) be two lines. Then \[ \left[\begin{vmatrix}l_2 & l_3\\m_2 & m_3\end{vmatrix}, \begin{vmatrix}l_3 & l_1\\m_3 & m_1\end{vmatrix}, \begin{vmatrix}l_1 & l_2\\m_1 & m_2\end{vmatrix}\right] \] is a point on both. -
There is a four-point:
By inspection, \([1,1,1],[1,0,0],[0,1,0],[0,0,1]\) is a four-point.
Summary: From Affine to Projective
- \((x,y) \in \alpha_R \mapsto [x,y,1] \in \pi_R\)
- Line \(ax+by+c=0 \mapsto \langle a,b,c \rangle\)
- Points \([a,b,0]\) are points at infinity
- Line \(\langle 0,0,1 \rangle\) is the line at infinity
Corollaries
- Lines \(\langle l_i \rangle\), \(\langle m_i \rangle\), \(\langle n_i \rangle\) concurrent iff determinant of coefficients is zero.
- If \(x, y\) lie on a line, then \(z\) lies on it iff \(z_i = \lambda x_i + \mu y_i\) for some \(\lambda,\mu\).
- Three points collinear iff there exist \(\alpha,\beta,\gamma\) not all zero such that \(\alpha x_i + \beta y_i + \gamma z_i = 0\).
Projective Plane over a Division Ring
Let \(D\) be a division ring. Define:
- \(\mathscr{P}_D = \{[x_1,x_2,x_3] \mid (x_i) \in D^3 \setminus \{0\}\}\)
- \(\mathscr{L}_D = \{\langle l_1,l_2,l_3 \rangle \mid (l_i) \in D^3 \setminus \{0\}\}\)
- Incidence: \(l_1x_1 + l_2x_2 + l_3x_3 = 0\)
- We define \(\mathscr{P}_D= \{[x_1,x_2,x_3] : (x_1,x_2,x_3) \in T \}\), where each point is a proportionality class of triples, not all zero.
- We define \(\mathscr{L}_D= \{ \langle l_1,l_2,l_3\rangle : (l_1,l_2,l_3) \in T \}\), where each line is a proportionality class of triples, not all zero.
- We define that a point \([x_1,x_2,x_3]\) lies on a line \(\langle l_1,l_2,l_3 \rangle\) iff \(l_1 x_1 + l_2 x_2 + l_3 x_3 = 0\).
Proof. The proof of the theorem is similar to the theorem for \(\pi_D\), so left for the reader.
If \(q\) is a power of a prime, then there exists a projective plane of order \(q\). If \(q\) is congruent to 1 or 2 modulo 4 and is not the sum of the squares of two integers, then there exists no projective plane of order \(q\).
Isomorphism of Planes
Planes \(\sigma = (\mathscr{P},\mathscr{L},\mathscr{I})\) and \(\sigma'\) are isomorphic (\(\sigma \sim \sigma'\)) if there are bijections: \[ f: \mathscr{P} \to \mathscr{P'}, \quad F: \mathscr{L} \to \mathscr{L'} \] preserving incidence: \((P,L) \in \mathscr{I} \iff (f(P), F(L)) \in \mathscr{I'}\).
-
\(F\) is well defined and one-one:
Let \(L_1 \in \mathscr{L}\), and \( L_2 \in \mathscr{L'}\), then \( x_1,y_1 \in L_1 \) and \( x_2,y_2 \in L_2\). Suppose \(L_1=L_2 \), then \(x_1,y_1, x_2,y_2 \) are collinear. This implies that \(f(x_1),f(y_1),f(x_2),f(y_2)\) are collinear, so \(F(L_1)=F(L_2)\). Hence, \( F:\mathscr{L} \to \mathscr{L}'\) is well defined and one-one. -
\(F\) is onto:
For each \( L' \in \mathscr{L'} \) there exists \( x',y' \in L' \) such that \(f(x)=x', f(y)=y'\). Since \(f(x),f(y)\) are collinear, therefore \(x,y\) are collinear. Thus, there exists a line \( L \in \mathscr{L}\) such that \(F(L)=L'\), where \(L=xy\). Hence, \(F\) is onto.
No comments:
Post a Comment