Read · 1 min
Suppose that we have an isometry in -dimensional coordinate space. We would like to show that this isometry may expressed as an orthogonal linear transformation, up to translation. We would also like to generalize to higher dimensions, if possible.
Let us denote , where is our isometry, is a final translation that we will ignore for now, and is what we care about.
Since we are disregarding the translation, the origin obviously maps to itself. Thus, .
Now, let us take two other points and .
Both of these points must be mapped to points on the unit circle, since isometries preserve distances.
Thus, let us say that .
Since isometries preserve angles, we know that the vector from the origin to is perpendicular to the vector from the origin to . Thus, we have one of two possibilities:
Since we have , , and , the isometry is actually completely determined!
In fact, it is easy to see that the isometry must be one of the following, depending on the value of : These representations leave no doubt that the isometry is indeed an orthogonal linear transformation.
Thus, we have proven that in -dimensional space, an isometry is an orthogonal linear transformation, up to translation.
A similar but much more complicated process may be used in any number of dimensions. First observe that the origin maps to itself, and then try to calculate the value of the function for all of the principal unit vectors.