Appendix B Complex Numbers and Complex Exponentials
B.1. Complex Numbers
By the way, read as “minus ” not “negative .” To say is “negative ” would imply that is positive, and it isn’t.
This is because we can not have an order on .
The complex conjugate of is .
We have
For the integral
We can also right
The magnitude of is
We also have
Polar form
For , we can also write it as
B.2. The Complex Exponential and Euler’s Formula
Define
This converges for all , but we won’t check that.
Also we have
To prove
Notice if then .
And if we fix , we have
From the knowledge of difference equation, we must have
Then plug , we have .
Euler’s formula
The left-hand side has only been defined via a series. The exponential function in the real case has nothing to do with the trig functions sine and cosine, and why it should have anything to do with them in the complex case is a true wonder.
Use some results for differential equations (the results themselves are difficult).
Note , so . Moreover
i.e. satisfy
On the other hand
So we have
Thus and satisfy the same differential equation with the same initial conditions; so and must be equal. Slick.
Plugging into Euler’s formula, we have
This is sometimes referred to as the most famous equation in mathematics. It expresses a simple relationship — and why should there be any at all? — between the fundamental numbers and , not to mention . We’ll probably never see this most famous equation again, but now we’ve seen it once.
Consequences of Euler’s formula.
Then the multiplication of 2 complex numbers is
We also have a dead easy way of deriving the addition formulas for the sine and cosine
On the other hand
So we have
Quote from the author:
I went through this derivation because it expresses in a simple way an extremely important principle in mathematics and its applications.
- If you can compute the same thing two different ways, chances are you’ve done something significant.
Take this seriously. This maxim appears throughout the course.
Symmetries of the sine and cosine: Even and odd functions.
Use this
We have
Note is an even function and is an odd function.
Why is this true? There are many ways of seeing it (Taylor series, differential equations), but here’s one you may not have thought of before, and it fits into a general framework of evenness and oddness that we’ll find useful when discussing symmetries of the Fourier transform.
Given any function ,
The top one is an even function and the bottom one is an odd function.
Algebra and geometry.
B.3. Further Applications of Euler’s Formula
For additional applications we’ll replace by and think of
we see that is a (complex-valued) parametrization of the circle: the circle is traced out exactly once in the counterclockwise direction as t goes from to .
For our efforts we prefer to work with
the complex exponential contains the sinusoids and , each of frequency 1 Hz.
With the , as goes from to the point traces out the unit circle exactly once (one cycle) in a counter clockwise direction.
The units in the exponential are (as they are in and )
Without the the units in are
change the amplitude and frequency and to include a phase shift
The general complex exponential that includes this information is then
- Then the point traces out the circle in the counterclockwise direction at a rate of cycles per second;
- The phase determines the starting point on the circle, for at the point is
In fact, we can write
We can think of this as the (initial) vector set rotating at a frequency Hz through multiplication by the time-varying phasor .
What happens when is negative? That simply reverses the direction of motion around the circle from counterclockwise to clockwise. The catch phrase is just so: * positive frequency means counterclockwise rotation * negative frequency means clockwise rotation.
Sums of sinusoids.
What one can say about the superposition of two signals?
Consider
whose imaginary part is the sum of sinusoids, above. Before messing with the algebra, think geometrically in terms of rotating vectors. At we have the two vectors from the origin to the starting points
Their sum is the starting point (or starting vector) for the sum of the two motions. But how do those two starting vectors move? They rotate together at the same rate, the motion of each described by and , respectively. Thus their sum also rotates at that rate — think of the whole parallelogram (vector sum) rotating rigidly about the vertex at the origin.
Now mess with the algebra and arrive at the same result:
Problems and Further Results
B.1.
If , show that the area of the triangle whose vertices are , , and is given by
Proof:
Assume is shown in the figure below.
Then if we rotate the triangle until is completely on -axis, then we can see the area is
Also note
So we have .
B.2.
Show that the equation of a line can be written as , where is a complex number and is real (and is the variable). What is the slope of the line?
Solution:
Let .
Then
I think the problem is asking, given a line like in the figure below, how to find complex number and real number , such that as long as satisfy
, then will be on that line.
Assume the distance between and the line is , and the angle between x-axis and the dash line is . From the figure, we can see, if we rotate clockwise by , to get , then is on the vertical line.
So we can have
The if satisfy , will be on the line.
Then slop of the line is .
B.3.
Recall the identity .
As a generalization of this, show that
Proof:
For the first part
For the second part