2

For the equation $$|z-a|+|z+a|=2|c|,$$ where $|a|\le|c|$, what are the smallest and largest values of $|z|$?

I am finding various bounds, but I don't know which bounds are the most precise ones. For example, I found that $|z| \le |c|$, but can this bound be improved?

Joffan
  • 39,627
jimm
  • 1,017
  • 5
  • 22
  • Go to this link as a little hint $\to$ https://math.stackexchange.com/questions/2314488/how-can-i-find-maximum-and-minimum-modulus-of-a-complex-number?rq=1 – George N. Missailidis Aug 17 '17 at 04:12
  • Picture the ellipse, then think at which are the points closest to and farthest from its center. – dxiv Aug 17 '17 at 04:24
  • You should change the title to "maximum and minimum of the absolute value of a complex number". "maximum and minimum value of a complex number" is, of course, nonsense. – fleablood Aug 17 '17 at 05:34

2 Answers2

2

Using arithmetic-quadratic mean inequality, $$|c| = \frac{|z-a|+|z+a|}{2} \le \sqrt{\frac{|z-a|^2+|z+a|^2}{2}}=\sqrt{|z|^2+|a|^2}.$$ Consequently, $|z| \ge \sqrt{|c|^2-|a|^2}$. The minimum value is achieved when $z = \pm i a \sqrt{\frac{|c|^2}{|a|^2}-1}$.

For the maximum value of $|z|$, observe that $$2|c|=|z-a|+|z+a| \ge |z+a+z-a|=2|z|.$$ Consequently, $|z|\le|c|$, and the maximum value is achieved when $z = \pm a \frac{|c|}{|a|}$.

Math Lover
  • 15,153
2

Geometric hint:   the locus $|z-a|+|z+a|=2|c|$ of points $z$ with a constant sum of distances to two fixed points $a,-a$ is an ellipse centered at the origin, with foci $-a, a$ and semi-major axis $|c|\,$.

The point farthest from the origin is either vertex $z=\lambda \cdot a\,$, which substituting into the equation gives $\lambda=|c|/|a|\,$, so the vertex is $z = |c|/|a| \cdot a\,$ and the maximum magnitude is $|z| = |c|\,$.

The point closest to the origin is either co-vertex $z=i\,\mu \cdot a\,$, which substituting into the equation gives $\mu=\sqrt{|c|^2/|a|^2-1}\,$, so the co-vertex is $z = i\,\sqrt{|c|^2/\,|a|^2-1} \cdot a\,$ and the minimum magnitude is $|z| = \sqrt{|c|^2 - |a|^2}\,$.

dxiv
  • 76,497
  • The only reason I did not give the proof of the minimum is that I found the minimum using your hint. I was waiting for your answer. – Math Lover Aug 17 '17 at 05:26
  • @MathLover And I was waiting for you to complete the proof before I decided to post this as an answer ;-) Seriously now, it is fair game to use hints/clues/etc posted in comments, as far as I am concerned. – dxiv Aug 17 '17 at 05:30
  • Posted a new solution based on the arithmetic-quadratic mean inequality. – Math Lover Aug 17 '17 at 17:52
  • @MathLover Right, that's a direct way to prove it, and a nice well rounded answer overall. – dxiv Aug 17 '17 at 18:03