0
$\begingroup$

I just had that question in my final exam

Solve in $\mathbb{C}$ : $|z-i| = |z-1|$

and I couldn't do it. I found a similar thread here : Showing that $\{z\in\mathbb{C}:|z-1|<|z+i|\}$ is an open set but it's too complicated for me to understand.

I tried 2 different ways to solve this

First, I developed the $z$ which gave me $$\begin{align} & (a+bi)-i=(a+bi)-1 \\ & = a-i+bi=a-1+bi \\ & = a-1(i+b)=a-1+bi \\ \end{align} $$ However we are looking at the module of this, so i have to do $\sqrt{((a^2)+(-1(i+b))^2} = \sqrt{((a-1)^2+(bi)^2)}$

Then I got some sort of $\sqrt{a^2-b^2-2ab}$ that equals to similar on the other side but couldn't find anything.

I tried the geometric way and the "logic" way (how can $i = 1$ when $i^2 = -1$), but I just don't understand how i'm suppose to solve this.

I also tried using the roots which say that for $z^n = w$

$w = r^{(1/n)} \text{Ei}((\angle m \cdot \frac{2k\pi}{n}))$

Every formula I knew failed

$\endgroup$
5
$\begingroup$

Hint: The geometric way is probably the simplest. The expression $|z-w|$ gives the distance between the points $z$ and $w$, so $z$ satisfying $|z-i|=|z-1|$ corresponds to a point which is equidistant from $1$ and $i$. (Or, in Cartesian coordinates, $(1,0)$ and $(0,1)$). But there's only one geometric figure corresponding to the set of points which are equidistant to two other points...

$\endgroup$
  • $\begingroup$ I'm not sure I understand, sorry. I tried to draw a cart to represent what you said : imgur.com/ZvXgX9l But I don't quite get your point...Do you mean I need to calculate the hypothenuse of a right triangle? $\endgroup$ – Maude Jun 15 '16 at 21:45
  • $\begingroup$ @Maude I really don't understand what's in that diagram. And this really doesn't have much to do with right triangles (though I can see how it might appear so). The idea is simpler: If I pick two points, then in general any other point in the plane will be closer to one than the other. How would you represent that idea in a drawing? $\endgroup$ – Semiclassical Jun 15 '16 at 21:49
  • $\begingroup$ I'm sorry i'm very bad in geometry. If you pick 2 points, then the third one will make a triangle, no? $\endgroup$ – Maude Jun 15 '16 at 21:57
  • 1
    $\begingroup$ Given that this guessing game doesn't seem to be helpful, I'll just give say what I'm getting at: If I pick any two points, the set of points which is equidistant from both of them is the straight line which lies halfway between them. If I had the points $(1,0)$ and $(-1,0)$, this would be the $y$-axis. In the complex plane, with points $z=1$ and $z=i$, this corresponds to the condition $\text{Re }z=\text{Im }z$ i.e. the line through the points $z=0$ and $z=1+i$ (both of which satisfy the equation under consideration.) $\endgroup$ – Semiclassical Jun 15 '16 at 22:05
1
$\begingroup$

Write $z = a + bi$. Then you are trying to find all $z$ for which $$\sqrt{a^2 + (b - 1)^2} = \sqrt{(a - 1)^2 + b^2}$$ Squaring both sides and expanding the squares, this becomes $$a^2 + b^2 - 2b + 1 = a^2 - 2a + 1 + b^2$$ Subtracting $a^2 + b^2 - 1$ from both sides, this becomes $$-2b = -2a$$ So we are reduced to $a = b$. All steps here are reversible, so we conclude that all complex numbers where $b = a$ satisfy the equality. Geometrically this is the line $y = x$, the points equidistant from $(1,0)$ and $(0,1)$.

$\endgroup$
  • $\begingroup$ Oh this is exactly what I did but I was stuck at square rooting a^2+b^2-2b+1..I didn't know how to remove the root without messing up the equation. I guess I was on the right track! $\endgroup$ – Maude Jun 15 '16 at 21:54
1
$\begingroup$

Your approach works, you just made a mistake in algebra. The condition $|z-i|=|z-1|$ is equivalent to $|z-i|^2=|z-1|^2$. Plugging in $z:=a+bi$, this is equivalent to $$ \begin{align} |a+bi-i|^2&=|a+bi-1|^2\\ \Longleftrightarrow\ |a+i(b-1)|^2&=|(a-1)+bi|^2.\\ \end{align} $$ Using the identity $|x+iy|^2=x^2+y^2$ we get $$ a^2+(b-1)^2=(a-1)^2+b^2.\tag1$$ Expanding (1) and simplifying gives you $$a=b$$

$\endgroup$
  • $\begingroup$ Thanks! i'm really happy to know I wasn't wrong all along =) $\endgroup$ – Maude Jun 15 '16 at 22:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.