2
$\begingroup$

First of all, I did some research on this question. I found the same question (but only for $\mathbb{R}^2$) here. I also found an article in Wikipedia about how to generalize spherical coordinates for n dimensions, which promptly leads to the answer.

My question is: Is there any other way of finding a homeomorphism that wouldn't rely on previous knowledge about this coordinate system?

$\endgroup$
  • $\begingroup$ I suppose you mean $S^n\times \mathbb{R}_{>0}$, right? Besides that I don't quite understand the question. Geometrically speaking this homeomorphism is natural, independently of the dimension $\endgroup$ – b00n heT Nov 10 '16 at 20:39
  • 2
    $\begingroup$ @b00nheT: $\Bbb R$ works just as well: after all, it’s homeomorphic to $\Bbb R_{>0}$. $\endgroup$ – Brian M. Scott Nov 10 '16 at 20:40
  • 1
    $\begingroup$ @b00nheT The Reals and the positive reals are homeomorphic, so it doesn't matter. $\endgroup$ – N. Owad Nov 10 '16 at 20:40
  • 1
    $\begingroup$ You are indeed right (Brian and Owad). But putting only positive reals makes the bijection more immediate, don't you think? $\endgroup$ – b00n heT Nov 10 '16 at 20:44
  • $\begingroup$ The question is about $S^n \times \mathbb{R}$, after all. Perhaps the solution of the problem will require a formula for a homeomorphism between $\mathbb{R}$ and $\mathbb{R}_{>0}$... $\endgroup$ – Lee Mosher Nov 10 '16 at 21:13
3
$\begingroup$

... And I myself have just found the answer I needed.

Let $x=(x_1,x_2,...,x_{n+1}) \in \mathbb{R}^{n+1}\backslash\{0\}$. Then, I could have simply made $f:\mathbb{R}^{n+1}\backslash\{0\}\rightarrow S^n\times\mathbb{R}$ as:

$$f(x_1,...,x_{n+1})=\left(\frac{x_1}{||x||},...,\frac{x_{n+1}}{||x||},\ln||x||\right)$$

And it's inverse $f^{n-1}:S^n\times\mathbb{R}\rightarrow \mathbb{R}^{n+1}\backslash\{0\}$ as

$$f^{-1}(x_1,...,x_{n+1},z)=\left(e^z x_1,...,e^z x_{n+1}\right)$$

$\endgroup$
1
$\begingroup$

You have mistaken between R^(n+1)-{0} and R^(n+1), but this answer is very good. If you learn Algebraic of Topology, you will have some solutions to find a homeomorphism.

$\endgroup$
  • $\begingroup$ Thanks for the tip, I edited it. $\endgroup$ – AspiringMathematician Nov 16 '16 at 1:02

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.