Conformal Mapping & Bilinear Transformation
Open in the Padhaiko app2.5 Transformations
For a complex function , there are two complex planes: all values of the independent variable are plotted in the -plane, and the corresponding functional values are plotted in the -plane. The correspondence between points in the -plane and points in the -plane is called a transformation or mapping, denoted briefly as the mapping . Corresponding points in the two planes are called images of each other.
We first examine the elementary transformations: translation, rotation, magnification, and inversion.
Translation:
Under the transformation , where is a complex constant, every point in the -plane is displaced (translated) in the direction of the vector through a distance :
Equating real and imaginary parts gives
Thus a point in the -plane is translated along the vector into the image point in the -plane. Any region in the -plane maps to a congruent, translated copy of that region in the -plane.
1
What is the region of the -plane into which the rectangular region in the -plane bounded by the lines , , , and is mapped under the transformation ?
Expressing the transformation in Cartesian components: Write the transformation in terms of real and imaginary components:
Equating real and imaginary parts:
Mapping the boundary lines of the rectangle: The region in the -plane is bounded by the four lines
Substituting these boundary values into equation (1):
- When : .
- When : .
- When : .
- When : .
Hence the image region in the -plane is the rectangle bounded by the lines , , , and .
The rectangle in the -plane bounded by the lines , , , and .
Rotation: ( real)
A pure rotation about the origin is defined by the rule
Setting and in (1) gives
Equating moduli and arguments:
If is a point in the -plane, its image is obtained by rotating the radius vector about the origin through an angle :
- The rotation is counter-clockwise (anti-clockwise) if .
- The rotation is clockwise if .
The modulus remains unchanged while the argument increases or decreases by . Consequently, every region in the -plane is rotated rigidly about the origin through the angle .
2
Determine the region of in the -plane corresponding to the triangular region bounded by the lines , , and in the -plane.
Expressing the transformation in Cartesian coordinates:
Equating real and imaginary parts:
Transforming each of the boundary lines , , and :
- For : Substituting into (1) gives
- For : Substituting into (1) gives
- For : Substituting into (1) gives
The triangular region in the -plane bounded by the lines , , and .
3
Show that the mapping rotates each non-zero point counter-clockwise through a right angle about the origin, and determine the region in the -plane corresponding to the triangular region bounded by , , and .
Writing the transformation in terms of and :
Equating real and imaginary parts gives
Since , multiplying by preserves and adds to , rotating every non-zero point counter-clockwise through a right angle.
Mapping the triangular region: The triangular region in the -plane is bounded by the lines
Substituting these boundaries into and :
- ,
- ,
- .
Hence the triangular region bounded by , , and in the -plane is mapped into the triangular region in the -plane bounded by , , and .
The triangular region in the -plane bounded by the lines , , and .
Magnification: ()
A magnification (also called a homothetic transformation) is defined by
Equating real and imaginary parts gives
Every point in the -plane is moved radially along the ray to such that :
- If , the region is stretched (expanded) away from the origin.
- If , the region is contracted toward the origin.
4
Determine the region of the transformation in the -plane where the region in the -plane is enclosed by the lines , , and .
Writing the transformation in Cartesian components:
Equating real and imaginary parts:
Mapping the boundary lines: The triangular region in the -plane is bounded by
- Putting and in (1) gives and .
- Adding and and using :
The triangular region in the -plane bounded by the lines , , and .
Rotation and Magnification: ( complex)
When is a non-zero complex constant (, ), the transformation
combines both magnification and rotation:
Thus is mapped to :
- Rotation: counter-clockwise if , clockwise if , and no rotation if .
- Magnification: expansion if , contraction if , and no magnification if .
5
Determine the image region under the transformation when the region in the -plane is bounded by , , , and .
Writing the transformation in Cartesian form:
Equating real and imaginary parts gives
Transforming each of the four boundary lines , , , and :
- For : and .
- For : and .
- For : and . Adding gives .
- For : and . Subtracting gives , or .
The rectangular region in the -plane bounded by the lines , , , and .
Inversion:
The inversion transformation is defined by
Equating moduli and arguments gives
Geometrically, is the composition of two steps:
- Inversion in the unit circle : , which sends a point at distance along the ray to distance along the same ray.
- Reflection across the real axis: .
In Cartesian coordinates, since :
and conversely,
6
Find the image of the infinite horizontal strips: (i) (ii) under the transformation .
Expressing in terms of :
Equating real and imaginary parts gives
Part (i): Strip :
- For the lower boundary , substitute into (1):
which is a circle in the -plane with center and radius .
- For the upper boundary , substitute into (1):
which is a circle in the -plane with center and radius .
Therefore, the infinite strip in the -plane maps to the region in the -plane between the circles and .
Part (ii): Strip :
- For : , and since we have (the lower half-plane).
- For : as above, . Since , we have .
Thus the strip is mapped into the region of the lower half-plane lying outside the circle .
(i) The region between the circles and ; (ii) the region in the lower half-plane bounded by .
Conformal Mapping
Definition 2.5.1: Conformal and Isogonal Mapping
Let two smooth curves and in the -plane intersect at a point at an angle , and let their image curves and under intersect at at an angle .
- If the angle of intersection is preserved in both magnitude and sense (direction) (), the mapping is called conformal at .
- If only the magnitude of the angle is preserved while the sense may be reversed (for example, ), the mapping is called isogonal.
Theorem 13: Analyticity Condition for Conformal Mapping
If is analytic at and in a region of the -plane, then the mapping is conformal at (and at every point of the region where ).
Proof
Let be a point in the -plane lying on a smooth curve , and let be its image on the corresponding curve in the -plane. Take a neighboring point on and its image on , and write the increments in polar form:
where , , , and . Then the difference quotient is
As along (), the chord angle approaches (the angle the tangent to at makes with the positive -axis) and approaches (the angle the tangent to at makes with the positive -axis). Taking the limit as :
Since , write , where and . Comparing moduli and arguments in (2) gives
Now let a second curve pass through with tangent angle , and let its image pass through with tangent angle . Since depends only on the point and not on the curve, we likewise have
Subtracting (3) from (4) yields
Thus the directed angle between the two curves at equals the directed angle between their image curves at in both magnitude and sense. Therefore is conformal at .
Key Properties of Conformal Mappings (Notes I–IV)
- Note I (Critical vs. Ordinary Points): A point at which is called a critical point of the mapping (conformality fails there; angles are multiplied). Points where are called ordinary points.
- Note II (Angle of Rotation): Any tangent vector at is rotated through the angle
- Note III (Preservation of Harmonicity): A harmonic function remains harmonic under a conformal mapping.
- Note IV (Coefficient of Magnification): Infinitesimal arc lengths passing through are magnified in the ratio , where
is called the coefficient of magnification (or linear scale factor) at .
8
For the conformal mapping , find the coefficient of magnification and the angle of rotation at .
Evaluating at : Differentiate with respect to and evaluate at :
Computing the coefficient of magnification and angle of rotation:
- Coefficient of magnification:
- Angle of rotation:
Coefficient of magnification ; angle of rotation .
Bilinear (Mobius) Transformation
Bilinear (Mobius / Linear Fractional / Homographic) Transformation
A transformation defined by the rule
where are complex constants, is called a bilinear transformation (because multiplying out gives , which is linear in both and ). It is also known as a Mobius transformation, linear fractional transformation, or homographic transformation. The quantity is called the determinant of the transformation.
Properties of Bilinear Transformations (Notes I–III)
- Note I (Analyticity): Since , is analytic everywhere in the finite complex plane except at the pole (when ). In the extended complex plane , and .
- Note II (Conformality): Differentiating with respect to gives
Since , never vanishes, so a bilinear transformation is conformal everywhere except at .
- Note III (Decomposition into Elementary Transformations): When , long division yields
Setting
shows that every bilinear transformation is a composition of translations, inversion, rotation, and magnification.
Theorem 14: Bilinear Transformations Map Circles into Circles
A linear fractional transformation () maps circles in the -plane into circles in the -plane (where straight lines are regarded as limiting cases of circles passing through ).
Proof
Let a circle in the -plane with center and radius be given by
Solving for in terms of :
Substitute into (1):
Assuming and , factor out :
where , , and . This is Apollonius's equation of a circle in the -plane (a proper circle when , and the perpendicular bisector line when ).
Theorem 15: Invariance of the Cross-Ratio
A linear fractional transformation () preserves the cross-ratio of four distinct points: if map to respectively, then
Proof
For any and (), compute the difference :
Divide (1) by (2) and (3) by (4):
Dividing (5) by (6), the factors and cancel completely, leaving
Fixed Points and Normal Forms of a Bilinear Transformation
Fixed (Invariant) Points of a Bilinear Transformation
Points that coincide with their own images () under a bilinear transformation are called fixed points (or invariant points) of the transformation. Setting in gives the fixed-point equation:
Theorem 16: Number of Fixed Points
A linear fractional transformation that is not the identity mapping has at most two fixed points in the finite plane; if it has three or more fixed points, it must be the identity mapping .
Proof
Fixed points satisfy the quadratic equation
- If , (1) is a quadratic equation in . It has two distinct fixed points when , and one repeated fixed point when .
- If and , (1) is linear and has a single finite fixed point .
- If , , and (a pure translation ), there is no finite fixed point.
If (1) is satisfied by more than two distinct values of , every coefficient of the quadratic polynomial must vanish:
Substituting these into yields , which is the identity mapping.
Normal Forms and Classification
- Two distinct finite fixed points : The bilinear transformation can be written in the normal (canonical) form
where the multiplier determines the geometric nature of the transformation:
- Elliptic if (, ),
- Hyperbolic if (),
- Loxodromic if with and .
- One repeated finite fixed point (): The transformation is called parabolic and its normal form is
Worked Out Examples (Bilinear Transformations)
1
Find the linear transformation which maps the points into the points respectively. Find also the fixed points of the transformation.
Applying the cross-ratio formula with : Here , , and , , . By invariance of the cross-ratio:
Substitute the given values:
Solving for and finding the fixed points:
Setting to find the fixed points:
; fixed points .
2
Find the linear transformation that maps into the points .
Setting up the cross-ratio equation: Substitute and into the cross-ratio formula:
Solving for : Cross-multiply and collect terms in :
.
3
Find the linear transformation that maps the points into the points respectively.
Setting up the cross-ratio equation: Substitute the given points into the cross-ratio formula:
Since , this simplifies to
Solving for : Cross-multiply and collect terms in :
Factor the coefficient of as :
.
4
Find the linear transformation which maps the points into the points .
Setting up the cross-ratio equation: Substitute and into the cross-ratio formula:
Since , we have
Solving for : Cross-multiply and solve for :
Since and :
.
5
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation:
Substitute and into the cross-ratio formula:
Solving for : Cross-multiply and solve for :
(printed book solution duplicates Example 4: ).
6
Find the fixed points and the normal form of the following bilinear transformations, and determine the nature of each transformation: (i) (ii)
Part (i): : Set to find the fixed points:
Using the two distinct fixed points and , compute and :
Dividing the first equation by the second gives the normal form:
Here the multiplier is . Since (), the transformation is elliptic.
Part (ii): : Set to find the fixed points:
Thus is a single repeated fixed point. Subtracting from both sides of :
Taking reciprocals gives the normal form:
Because there is a single repeated fixed point , the transformation is parabolic.
(i) Fixed points , normal form , elliptic; (ii) fixed point (repeated), normal form , parabolic.
Exercise - 7
Exercise - 7
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . By the cross-ratio invariance of bilinear transformations:
Substitute the given points:
Simplifying both sides: Simplifying the left-hand side:
Simplifying the right-hand side:
Equating the two sides:
Solving for : Cross-multiplying:
Find the linear fractional transformation that maps into the points respectively.
Applying the cross-ratio formula with : Let and . Taking the limit as in the cross-ratio formula:
Substituting the values:
Simplifying the cross-ratio equation: Evaluating both sides:
Solving for : Cross-multiplying and grouping terms in :
(The printed answer omits an overall minus sign in the numerator; the corrected transformation mapping is ).
Printed: (corrected for overall sign: ).
Find the linear fractional transformation that maps into the points respectively. Also find the fixed points of the transformation.
Setting up the cross-ratio equation: Let and . Using cross-ratio invariance:
Substitute the given points:
Simplifying the constants: Notice that:
Thus:
Dividing by (that is, multiplying by ) and using :
Solving for : If , then componendo and dividendo give . Here and :
Dividing numerator and denominator by , using , and :
Finding the fixed points: Setting :
Dividing through by :
Therefore, the fixed points are .
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio formula with : Let and . Taking the limit as :
Substitute the given points:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating and cancelling :
Solving for : Cross-multiplying:
Find the linear transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . By invariance of the cross-ratio:
Substitute the values:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating and cancelling :
Solving for : Cross-multiplying:
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . Invariance of the cross-ratio gives:
Substitute the points:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating both sides:
Solving for : Cross-multiplying:
Dividing both sides by :
Thus:
(The printed answer inverts the sign of the mapping; the corrected mapping is ).
Printed: (corrected: ).
Find the linear fractional transformation that maps into the points respectively.
Applying the cross-ratio formula with : Let and . Taking the limit as :
Substitute the given points:
Simplifying to find : Evaluate the right-hand side:
(The printed answer contains an extraneous in the denominator; the corrected transformation is ).
Printed: (corrected: ).
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . Invariance of the cross-ratio gives:
Substitute the values:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating gives:
Solving for : Cross-multiplying:
Dividing numerator and denominator by :
Multiplying numerator and denominator by :
Thus:
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . By invariance of the cross-ratio:
Substitute the points:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating and cancelling :
Solving for : Cross-multiplying:
Find the linear fractional transformation that maps into the points respectively.
Applying the cross-ratio formula with : Let and . Taking the limit as :
Substitute the given points:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating:
Solving for : Taking reciprocals on both sides:
(The printed answer maps and ; the unique transformation mapping the given points is ).
Printed: (corrected for : ).
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . By invariance of the cross-ratio:
Substitute the given points:
Simplifying both sides: Left-hand side, using :
Right-hand side:
Equating:
Solving for : Cross-multiplying:
Dividing by , using :
Find the linear fractional transformation that maps into the points respectively.
Setting up the cross-ratio equation: Let and . Using cross-ratio invariance:
Substitute the points:
Simplifying both sides: Left-hand side:
Right-hand side:
Equating:
Solving for : Applying componendo and dividendo:
Factoring from both numerator and denominator:
(The printed answer omits the factor before in the denominator; the corrected mapping is ).
Printed: (corrected: ).
Find the transformation which maps the semi-infinite strip bounded by and into the upper half of the -plane at and .
Relating the geometry to the sine transformation: The standard transformation maps the semi-infinite vertical strip , in the -plane conformally onto the upper half of the -plane , sending the boundary vertices and to and respectively.
Rotating and scaling the strip: The given strip is bounded by , , and (with ), which is a horizontal semi-infinite strip of width oriented along the positive -axis. To rotate the vertical strip of width into a horizontal strip of width :
- Scale by the factor to adjust the strip width from to .
- Multiply by to rotate the strip through an angle of (mapping the vertical boundary lines to horizontal lines ).
Thus:
Expressing in terms of : Since :
Find the transformation which maps the semi-infinite strip between the lines in the -plane for which into the upper half of the -plane at and .
Relating to the standard conformal map : Under the mapping , the semi-infinite strip in the -plane is mapped conformally onto the upper half-plane . The boundary points and correspond to and on the real axis of the -plane.
Scaling the strip width from to : The given strip in the -plane lies between and for . To scale the strip of width in the -plane to width in the -plane, let:
Writing the transformation: Inverting gives . Substituting this into the relation between and :
Find the transformation which maps the semi-infinite strip between the lines in the -plane for which into the upper half of the -plane at and .
Analyzing the mapping : Let and . Under the mapping :
Equating real and imaginary parts:
Mapping the boundary of the strip: For and :
- Along the base (): and , which is the segment on the real axis.
- Along the right boundary (): and .
- Along the left boundary (): and .
- In the interior (): and , so , which corresponds to the upper half-plane .
Inverting the relation: The boundary points map to . Inverting :
Show that the map of the circle under the transformation is an ellipse, and find its center and lengths of axes.
Parametrizing the circle : In polar form, any point on the circle can be represented as:
Then:
Evaluating : Adding and :
Separating into components in the -plane: Let . The given transformation is:
Equating real and imaginary parts:
Eliminating to find the curve: Using the trigonometric identity :
This is the standard equation of an ellipse:
- Center: .
- Semi-major axis: .
- Semi-minor axis: .
Thus the image is an ellipse with center and lengths of axes and .
Center: ; lengths of axes: and .