0
$\begingroup$

For a problem that I am currently working on, I want to demonstrate that the following set, with $d \gt 0$, is of infinite size (whether or not the set is countable or not is unimportant):

$$S=\left\{(x',y') \in \mathbb R \times \mathbb R : \sqrt{(x'-x_0)^2+(y'-y_0)^2}=d \gt 0\right\}$$

The above equation should be recognizable as the formula for a circle in a 2D Euclidean plane that has its origin at $(x_0,y_0)$.

To show that this $S$ has infinite members, I created the following function: $f: \mathbb N \to [0,\frac{\pi}{2})$, where $f(n)=\frac{\pi}{2n}\cdot(n-1)$. Note that $f$ is a strictly increasing function, which means that $f$ is injective. For notation convenience, let $\theta_n:=f(n)$.

Next, to each unique $\theta_n$, assign the ordered pair $\left(\cos(\theta_n))\cdot d+x_0,\sin(\theta_n)\cdot d+ y_0 \right)$. Because $\cos$ and $\sin$ are functions, clearly the assignment of $\theta_n$ to $\left(\cos(\theta_n))\cdot d+x_0,\sin(\theta_n)\cdot d+ y_0 \right)$ is unambiguous. This means that we can consider the function $g: [0,\frac{\pi}{2}) \to \mathbb R \times \mathbb R$, where $g(\theta_n)=\left(\cos(\theta_n))\cdot d+x_0,\sin(\theta_n)\cdot d+y_0 \right)$. It should be straightforward to see that setting $x'=\cos(\theta_n)\cdot d+x_0$ and $y'=\sin(\theta_n)\cdot d+y_0$ will solve the desired equation of $S$.

Note that over the interval $[0,\frac{\pi}{2})$, $\cos$ is injective (because it is strictly decreasing). This means that for $n_1 \neq n_2 \in [0,\frac{\pi}{2})$: $\cos(\theta_{n_1}) \neq \cos(\theta_{n_2})$ ...which means that $\left(\cos(\theta_{n_1}))\cdot d+x_0,\sin(\theta_{n_1})\cdot d+y_0 \right) \neq \left(\cos(\theta_{n_2}))\cdot d+x_0,\sin(\theta_{n_2})\cdot d+y_0 \right)$. We can then say that $g$ is injective.

Because $f$ is injective and $g$ is injective, we must have that their composition is injective. That is to say that the function $h = g \circ f : \mathbb N \to \mathbb R \times \mathbb R$ is injective. Given that $\mathbb N$ is an infinite set, this necessarily means that the $\text{image}(h)$ must also be a set of infinite size. Further, $\text{image}(h) \subseteq S$, by construction. This means that $S$ must be an infinite set, as well.

I think the logic for this argument is fine, but I was wondering if there were more standard approaches to show that a set has an infinite size.

$\endgroup$
1
  • $\begingroup$ That is not how to specify that $d>0$. "I want to demonstrate that if $d>0$, then the following set is of infinite size". $\endgroup$
    – TonyK
    Commented Mar 19, 2022 at 19:13

1 Answer 1

-1
$\begingroup$

For any $x\in[x_0-d,x_0+d]$, let $y=y_0+\sqrt{d^2-(x-x_0)^2}$. Then $(x,y)\in S$. Assuming $d>0$ (which you forgot to specify), there are an infinite (indeed, uncountable) number of such $x$. Hence $S$ is infinite (indeed, uncountable).

$\endgroup$
3
  • $\begingroup$ Could you please explain why $(x,y) \in S$? Plugging in your values for $x'$ and $y'$, I get: $$\sqrt{(x-x_0)^2+\left(\sqrt{(x-x_0)^2}\right)^2}$$, which equals $$\sqrt{(x-x_0)^2+(x-x_0)^2}$$, equaling $$(x-x_0)\cdot \sqrt{2}$$ Why should this equal $d$? Thanks~ $\endgroup$ Commented Mar 19, 2022 at 19:25
  • $\begingroup$ You are right. I hope it's correct now. $\endgroup$
    – TonyK
    Commented Mar 19, 2022 at 19:47
  • $\begingroup$ Took a few to derive, but, yes, I agree with your claim. Note that we assumed $d \gt 0$, $|x-x_0| \lt d$, and $y \gt y_0$ in order to arrive at your result...starting from the equation: $$d^2=(x-x_0)^2+(y-y_0)^2$$ $\endgroup$ Commented Mar 19, 2022 at 20:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .