How to calculate primitive functions on maximal intervals for periodic functions?

Lotto
Messages
251
Reaction score
16
Homework Statement
My task is to find all primitive functions to ##f(x)= \frac{\sin^2 x}{1+\sin^2 x}## on maximal intervals.
Relevant Equations
For intervals ##\left(-\frac{\pi}{2}+k\pi, \frac{\pi}{2}+k\pi \right)## it is quite easy: ##F(x)=x-\frac{1}{\sqrt 2} \arctan{\left(\sqrt 2 \tan x\right)}+C_k##. (##C_k## corresponds to a particular ##k##)
Since ##f(x)## is continuous in ##\mathbb R##, it has a primitive function in ##\mathbb R## as well, so we have to define ##F(x)## also for points ## \frac{\pi}{2}+k\pi##.

##\lim_{x \to \left(\frac{\pi}{2}+k\pi \right)^-} F(x) =\frac{\pi}{2}+k\pi -\frac{\pi}{2\sqrt 2}+C_k ##
##\lim_{x \to \left(\frac{\pi}{2}+k\pi \right)^+} F(x) =\frac{\pi}{2}+k\pi +\frac{\pi}{2\sqrt 2}+C_{k+1}##

And since it is continuous, we can write that ##C_{k+1}= -\frac{\pi}{\sqrt 2}+C_k##. But now I don't know how to continue. Am I close to the solution? How to write the whole definition for ##F(x)##?
 
Last edited:
Physics news on Phys.org
Er, no , be careful. Not every a.e. continuous nor continuous function has a well-defined antiderivative. Absolute Continuity is the minimal condition. The derivative of the Cantor Function is a classic example. The derivative of any a.e. 0 not identically 0 function will integrate to 0, but its antiderivative isn't a constant.
 
Lotto said:
Homework Statement: My task is to find all primitive functions to ##f(x)= \frac{\sin^2 x}{1+\sin^2 x}## on maximal intervals.
Relevant Equations: For intervals ##\left(-\frac{\pi}{2}+k\pi, \frac{\pi}{2}+k\pi \right)## it is quite easy: ##F(x)=x-\frac{1}{\sqrt 2} \arctan{\left(\sqrt 2 \tan x\right)}+C_k##. (##C_k## corresponds to a particular ##k##)

Since ##f(x)## is continuous in ##\mathbb R##, it has a primitive function in ##\mathbb R## as well, so we have to define ##F(x)## also for points ## \frac{\pi}{2}+k\pi##.

##\lim_{x \to \left(\frac{\pi}{2}+k\pi \right)^-} F(x) =\frac{\pi}{2}+k\pi -\frac{\pi}{2\sqrt 2}+C_k ##
##\lim_{x \to \left(\frac{\pi}{2}+k\pi \right)^+} F(x) =\frac{\pi}{2}+k\pi +\frac{\pi}{2\sqrt 2}+C_{k+1}##

And since it is continuous, we can write that ##C_{k+1}= -\frac{\pi}{\sqrt 2}+C_k##. But now I don't know how to continue. Am I close to the solution? How to write the whole definition for ##F(x)##?
As ##x \to \frac \pi 2##, we have ##\arctan(\sqrt 2 \tan x) \to \frac \pi 2##. So, we can define ##F(\frac \pi 2) = \frac \pi 2 - \frac \pi {2\sqrt 2}## and we have a continous primitive function.
 
PeroK said:
As ##x \to \frac \pi 2##, we have ##\arctan(\sqrt 2 \tan x) \to \frac \pi 2##. So, we can define ##F(\frac \pi 2) = \frac \pi 2 - \frac \pi {2\sqrt 2}## and we have a continous primitive function.
So, could we write the solution in this form?

For ##\left(-\frac{\pi}{2}+k\pi, \frac{\pi}{2}+k\pi\right)##: ##F(x) = x-\frac{1}{\sqrt 2} \arctan{\left(\sqrt 2 \tan x\right)}+C_k-k \frac{\pi}{\sqrt 2}##
For ##\frac{\pi}{2}+k\pi##: ##F(x) = \frac{\pi}{2}+k\pi-\frac{1}{\sqrt 2}\frac{\pi}{2}+C_k - k \frac{\pi}{\sqrt 2}##

We subtract ##k \frac{\pi}{\sqrt 2}## because when we go from one interval with ##k## to another one with ##k+1## (when considering the unmodified function), the value at point ##\frac{\pi}{2}+k\pi## is greater by ##k \frac{\pi}{\sqrt 2}##, so if we want the function to be continuous, we always have to shift it on given interval by ## k \frac{\pi}{\sqrt 2}## downwards.

I imagine it this way because the graph of the unmodified function looks like this

graph.png
 
For a continuous function with period T (such as \sin^2 x/(1 + \sin^2 x)), its antiderivative on [0, \infty) can be written as \begin{split}<br /> F(t) &amp;= \int_0^t f(s)\,ds + C \\<br /> &amp;= \sum_{n=0}^{\lfloor t/T \rfloor - 1} \int_{nT}^{(n+1)T} f(s)\,ds + \int_{\lfloor t/T \rfloor T}^t f(s)\,ds + C \\<br /> &amp;= \left\lfloor \frac{t}{T} \right\rfloor \int_0^T f(s)\,ds +\int_0^{t - \lfloor t/T \rfloor T}f(s)\,ds + C \end{split} so that the only integral you actually need to calculate is \int_0^t f(s)\,ds for t \in [0,T]. This can be extended to negative t, but one must be mindful of how \lfloor \cdot \rfloor works for negative numbers.
 
Thread 'Use greedy vertex coloring algorithm to prove the upper bound of χ'
Hi! I am struggling with the exercise I mentioned under "Homework statement". The exercise is about a specific "greedy vertex coloring algorithm". One definition (which matches what my book uses) can be found here: https://people.cs.uchicago.edu/~laci/HANDOUTS/greedycoloring.pdf Here is also a screenshot of the relevant parts of the linked PDF, i.e. the def. of the algorithm: Sadly I don't have much to show as far as a solution attempt goes, as I am stuck on how to proceed. I thought...
Back
Top