Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I want to know whether a function $f$ with the following properties can exist. $f$ should be periodic with period $2\pi$, $f$ should be smooth everywhere except at exactly one point per period, such that the function $f$ itself or one of its derivatives (of different order) should jump or blow-up at this point. Equally spaced sample points are chosen such that there are at least 3 points in one period. Now the values of this function $f$ at these sample points should be equal to those of a given sinusoidal function of period $2\pi$ i.e, $\sin(x)$. What I'd like to know is, can such a function $f$ exist ?

share|cite|improve this question
For a "jump discontinuity", just take your sinusoidal function and alter it at that one point. – uncookedfalcon Jun 2 '12 at 11:26
@uncookedfalcon : Thanks for the comment. I have made an attempt at an answer. Please comment any mistakes. – Rajesh Dachiraju Jun 2 '12 at 11:58
up vote 4 down vote accepted

The answer is yes, for any ordered derivative jumping. Just consider a bump function having a support $(0,l)$, now modify it by replacing the second half of the bump function with a negative of the original. $b_m(x) = bump(x), x\in[0,l/2)$ and $b_m(x) = -bump(x), x \in [l/2,l]$. Now Let $$g = b_m\ast b_m\ast ...(k~times)..\ast b_m$$, where $\ast$ is a convolution operation. Now $g$ is smooth except at $x = l/2$ where $g^{(k)}(x)$ jumps. The support of $g(x)$ is $(0,(k+1)l)$.

Choose $l$ such that $(k+1)l \leq d$ where $d$ is the spacing between the samples where you want functions to be equal. Now form the desired function $f(x)$ by adding a shifted version of $g(x)$ i.e., $g(x-\tau)$ to $\sin(x)$ to form $f(x) = g(x-\tau) + \sin(x)$. Choose $\tau$ such that the support of $g(x-\tau)$ sits exactly between two sample points. Now $f(x)$ agrees with $\sin(x)$ at all the required sample points.

PS : Please point out any errors/misconceptions.

share|cite|improve this answer
+1, elegant complete answer. :-) – Babak S. Jun 8 '12 at 15:03

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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